Sam Goldstein Dana Princiotta Jack A. Naglieri Editors Evolutionary

Sam Goldstein Dana Princiotta Jack A. Naglieri Editors Handbook of Intelligence Evolutionary Theory, Historical Perspective, and Current Concepts Sam Goldstein • Dana Princiotta Jack A. Naglieri Editors

Handbook of Intelligence

Evolutionary Theory, Historical Perspective, and Current Concepts

2015 This volume is dedicated to the thousands of children to whom I have administered IQ tests and from whom I have learned that life is far more complex than a set of scores. To the researchers and clinicians who have come before and will follow me in the future. Science is never easy and not always popular but all we have to manage and temper our extraordinary and frequently expansive beliefs; and to Sherrie, my wonderful wife and partner. Sam Goldstein

I am eternally grateful to my mother, Alison Helen Princiotta , for her extraordinary combination of love and wisdom; to my partner, Paul Alexander Brighton , for his unrelenting optimism and devotion; and to my nephews, Tyler James and Hunter James , for their collective enrichment of my life, coursing from New York all the way to “Oootah” (Utah). Dana Princiotta

This book is dedicated to my wife, Kathleen Kryza, with love and admiration. Jack A. Naglieri “I believe in intuition and inspiration. Imagination is more important than knowledge. For knowledge is limited, whereas imagination embraces the entire world, stimulating progress, giving birth to evolution. It is, strictly speaking, a real factor in scientiﬁ c research.” Albert Einstein

“Man’s mind, once stretched by a new idea, never regains its original dimensions.” Oliver Wendell Holmes, Sr.

“It is not the strongest or the most intelligent who will survive but those who can best manage change.” Charles Darwin

vii Editor Biographies

Sam Goldstein, Ph.D. is a psychologist with areas of study in school psychology, child development, and neuropsychology. He is licensed as a psychologist and certifi ed as a developmental disabilities evaluator in the State of Utah. Dr. Goldstein is a Fellow in the National Academy of Neuropsychology and American Academy of Cerebral Palsy and Developmental Medicine. He is a Board Certifi ed Pediatric Neuropsychologist. Dr. Goldstein is an Adjunct Assistant Professor, Department of Psychiatry, University of Utah School of Medicine. Since 1980, Dr. Goldstein has worked in a private practice setting as the Director of a multidisciplinary team, providing evaluation, case management, and treatment services for children and adults with histories of neurological disease and trauma, autism, learning disability, adjustment diffi culties, and attention defi cit hyperactivity disorder. Dr. Goldstein is on staff at the University Neuropsychiatric Institute. He has served as a member of the Children’s Hospital Craniofacial Team. He has also been a member of the Developmental Disabilities Clinic in the Department of Psychiatry at the University of Utah Medical School. Dr. Goldstein has authored, coauthored, or edited 50 clinical and trade publications, including 18 textbooks dealing with managing children’s behavior in the classroom, genetics, autism, attention disorders, resilience, and adult learning disabilities. With Barbara Ingersoll, Ph.D., he has coauthored texts dealing with controversial treatments for children’s learning and attention problems and childhood depression. With Anne Teeter Ellison, he has authored Clinician’s Guide to Adult ADHD: Assessment and Intervention. With Nancy Mather, Ph.D., he has completed three texts for teachers and parents concerning behavioral and educational issues. With Michael Goldstein, M.D., he has completed two texts on attention defi cit hyperactivity disorder. He has edited three texts with Cecil Reynolds, Ph.D., on neurodevelopmental and genetic disorders in children. With Robert Brooks, Ph.D., he has authored 12 texts including Handbook of Resilience in Children, fi rst and second editions; Understanding and Managing Children’s Classroom Behavior, Second Edition; Raising Resilient Children; Nurturing Resilience in Our Children; Seven Steps to Help Children Worry Less; Seven Steps to Anger Management; The Power of Resilience; Raising a Self-Disciplined Child; and Raising Resilient Children with Autism Spectrum Disorders. With Jack Naglieri, he has authored a number of texts on autism, assessment of intelligence, and executive functioning. He has coauthored a parent training

ix x Editor Biographies program and is currently completing a number of additional texts on resilience, intelligence, and genetics. Dr. Goldstein is the Editor-in-Chief of the Journal of Attention Disorders and serves on six editorial boards. He is also the Coeditor of the Encyclopedia of Child Development and Behavior. With Jack Naglieri, Ph.D., Dr. Goldstein is the coauthor of the Autism Spectrum Rating Scales, Comprehensive Executive Functioning Inventory, Rating Scales of Impairment, and with Dr. Naglieri and J. P. Das the Cognitive Assessment System, Second Edition. Dr. Goldstein, a knowledgeable and entertaining speaker, has lectured extensively on a national and international basis to thousands of professionals and parents concerning attention disorders in children, resilience, depression, adjustment and developmental impairments, autism, and assessment of brain dysfunction. Dana Princiotta, Ph.D. is a licensed psychologist in the state of Arizona. She completed postdoctoral study at the Neurology, Learning, and Behavior Center in Salt Lake City, Utah. In addition to this text, she has coauthored fi ve book chapters and a peer-reviewed article. Jack A. Naglieri, Ph.D. is a Research Professor at the Curry School of Education at the University of Virginia, Senior Research Scientist at the Devereux Center for Resilient Children, and Emeritus Professor of Psychology at George Mason University. He is a Fellow of APA Divisions 15 and 16, recipient of the 2001 Senior Scientist Award for APA Division 16, and the 2011 Italian American Psychology Assembly Award for Distinguished Contributions to Psychology. He is a Diplomate in Assessment Psychology, has earned a license as a School Psychologist in Virginia and Ohio, and earned school psychology certifi cations in New York, Georgia, Arizona, and Ohio. Dr. Naglieri has focused his professional efforts on theoretical and psychometric issues concerning intelligence, cognitive interventions, diagnosis of learning and emotional disorders, and theoretical and measurement issues pertaining to protective factors related to resilience. Dr. Naglieri is the author or coauthor of more than 300 scholarly papers, books, and tests. His scholarly research includes investigations related to exceptionalities such as mental retardation, specifi c learning disabilities, giftedness, and attention defi cit disorder; psychometric studies of tests such as the Wechsler Scales of Intelligence, Cognitive Assessment System, and the Kaufman Assessment Battery for Children; examination of race, gender, and ethnic differences in cognitive processing; fair assessment using nonverbal and neurocognitive processing tests; identifi cation of gifted minorities, IDEA, and identifi cation of specifi c learning disabilities; and cognitively based mathematics interventions. He has authored various books, including Essentials of CAS Assessment (Naglieri, 1999), and coauthored other books including Assessment of Cognitive Processes: The PASS Theory of Intelligence (Das, Naglieri, and Kirby, 1994); Helping Children Learn: Intervention Handouts for Use at School and Home, Second Edition (Naglieri and Pickering, 2010); Essentials of WNV Assessment (Brunnert, Naglieri, and Hardy-Braz, 2009); and Helping All Gifted Children Learn: A Teacher’s Guide to Using the NNAT2 (Naglieri, Brulles, and Lansdowne, 2009). Dr. Naglieri has also coedited Editor Biographies xi

books such as Handbook of Assessment Psychology (Graham and Naglieri, 2002), Assessment of Autism Spectrum Disorders (Goldstein, Naglieri, and Ozonoff, 2009), Assessing Impairment: From Theory to Practice (Goldstein and Naglieri, 2009), and A Practitioner’s Guide to Assessment of Intelligence and Achievement (Naglieri and Goldstein, 2009). Dr. Naglieri’s scholarly efforts also include development and publication of tests and rating scales. He began this work in the mid-1980s with the publication of the Matrix Analogies Tests (Naglieri, 1985) and the Draw-A- Person Quantitative Scoring System (Naglieri, 1988) and DAP: Screening Procedure for Emotional Disturbance (Naglieri, McNeish, and Bardos, 1991). He published the Devereux Behavior Rating Scale—School Form (Naglieri, LeBuffe, and Pfeiffer, 1993), Devereux Scales of Mental Disorders (Naglieri, LeBuffe, and Pfeiffer, 1994), and the Devereux Early Childhood Assessments (LeBuffe and Naglieri, 2003). In 1997, he published the General Ability Scale for Adults (Naglieri and Bardos, 1997), Cognitive Assessment System (Naglieri and Das, 1997), and Naglieri Nonverbal Ability Test— Multilevel Form (Naglieri, 1997). He published the Naglieri Nonverbal Ability Test, Second Edition (Naglieri, 2008); the Wechsler Nonverbal Scale of Ability (Wechsler and Naglieri, 2008); and the Devereux Elementary Student Strength Assessment (LeBuffe, Shapiro, and Naglieri, 2009). Most recently, he published the Cognitive Assessment System, Second Edition (Naglieri, Das, and Goldstein, 2013); Comprehensive Executive Function Scale (Naglieri and Goldstein, 2013); and the Autism Spectrum Rating Scale (2010). For more information see: www.jacknaglieri.com. Contents

Part I Introduction

1 The Evolution of Intelligence ...... 3 Sam Goldstein

Part II Background

2 Intelligence in Nonprimates ...... 11 Thomas R. Zentall 3 Intelligence in Nonhuman Primates ...... 27 Tara Mandalaywala, Christine Fleener, and Dario Maestripieri 4 The Evolution of Language ...... 47 Philip Lieberman 5 Evolution of the Human Brain: From Matter to Mind ...... 65 Michel A. Hofman 6 Intelligence as a Conceptual Construct: The Philosophy of Plato and Pascal...... 83 Dana Princiotta and Sam Goldstein 7 The Life and Evolution of Early Intelligence Theorists: Darwin, Galton, and Charcot...... 93 Jordan Rigby 8 Social Competition and the Evolution of Fluid Intelligence ...... 105 David C. Geary

Part III Theories of Intelligence

9 Intelligence Defined: Wundt, James, Cattell, Thorndike, Goddard, and Yerkes ...... 123 John D. Greenwood 10 Piaget’s Theory of Intelligence ...... 137 Ulrich Müller, Kayla Ten Eycke, and Lesley Baker

xiii xiv Contents

11 Alfred Binet and the Children of Paris ...... 153 Amber Esping and Jonathan A. Plucker 12 From Psychometric Testing to Clinical Assessment: Personalities, Ideas, and Events That Shaped David Wechsler’s Views of Intelligence and Its Assessment ...... 163 Mark Benisz, Ron Dumont, and John O. Willis 13 A.R. Luria and Intelligence Defined as a Neuropsychological Construct ...... 181 Dana Princiotta and Sam Goldstein 14 Intelligence: Defined as Neurocognitive Processing ...... 193 Tulio M. Otero 15 CHC Theory of Intelligence ...... 209 Samuel O. Ortiz 16 Multiple Intelligences in the New Age of Thinking...... 229 Robert J. Sternberg 17 Emotional and Social Intelligence and Behavior ...... 243 Richard E. Boyatzis, James Gaskin, and Hongguo Wei 18 Intelligence as a Malleable Construct ...... 263 Lisa S. Blackwell, Sylvia Rodriguez, and Belén Guerra-Carrillo 19 Creativity and Intelligence ...... 283 Jonathan A. Plucker, Amber Esping, James C. Kaufman, and Maria J. Avitia

Part IV Assessment of Intelligence

20 Hundred Years of Intelligence Testing: Moving from Traditional IQ to Second-Generation Intelligence Tests ...... 295 Jack A. Naglieri 21 The Relationship Between Theories of Intelligence and Intelligence Tests ...... 317 W. Joel Schneider and Dawn P. Flanagan 22 Intelligence and Culture: History and Assessment ...... 341 Donald H. Saklofske, Fons J. R. van de Vijver, Thomas Oakland, Elias Mpofu, and Lisa A. Suzuki 23 Common and Variable Aspects of Intelligence ...... 367 Arthur MacNeill Horton Jr. and Cecil R. Reynolds 24 Current Concepts in the Assessment of Emotional Intelligence ...... 381 Steven J. Stein and Justin M. Deonarine Contents xv

Part V Applications of Intellectual Theory

25 Intelligence and Success ...... 405 Tarmo Strenze 26 The Use of Intelligence Tests in the Diagnosis of Specific Reading Disability ...... 415 Nancy Mather and Deborah Schneider 27 Executive Functioning and Intelligence ...... 435 Emily C. Duggan and Mauricio A. Garcia-Barrera 28 The Evolution of Intelligence: Implications for Educational Programming and Policy ...... 459 Christopher Jones, Peggy L. Tarpley, and Douglas Blancero

Part VI Conclusion

29 The March of Reason: What Was Hidden in Our Genes ...... 471 James R. Flynn 30 Closing Comments: Intelligence and Intelligence Tests – Past, Present, and Future ...... 487 Jack A. Naglieri and Sam Goldstein

Index ...... 491 Contributors

Maria J. Avitia University of Connecticut Storrs, Willington, CT, USA Lesley Baker Department of Psychology, University of Victoria, Victoria, BC, Canada Mark Benisz, Psy.D. Kiryas Joel School District, Monroe, NY, USA Lisa S. Blackwell, Ph.D. Mindset Works, Walnut, CA, USA Douglas Blancero JP Associates, Red Hook, NY, USA Richard E. Boyatzis Case Western Reserve University, Cleveland Heights, OH, USA Justin M. Deonarine Multi-Health Systems, Toronto, ON, Canada Emily C. Duggan, M.Sc. Department of Psychology, University of Victoria, Victoria, BC, Canada Ron Dumont, Ed.D. Department of Psychology, Fairleigh Dickinson University, Teaneck, NJ, USA Amber Esping College of Education, Texas Christian University, Fort Worth, TX, USA Dawn P. Flanagan, Ph.D. Department of Psychology, St. John’s University, Queens, NY, USA Christine Fleener, M.A. Department of Comparative Human Development, Institute for Mind and Biology, University of Chicago, Chicago, IL, USA James R. Flynn POLS Department, University of Otago, Dunedin, New Zealand Mauricio A. Garcia-Barrera, Ph.D. Department of Psychology, University of Victoria, Victoria, BC, Canada James Gaskin Marriott School of Management, Brigham Young University, Provo, UT, USA David C. Geary, Ph.D. Department of Psychological Sciences, Interdisciplinary Neuroscience Program, University of Missouri, Columbia, MO, USA Sam Goldstein, Ph.D. University of Utah, Salt Lake City, UT, USA

xvii xviii Contributors

John D. Greenwood, Ph.D. CUNY Graduate Center, New York, NY, USA Belén Guerra-Carrillo, B.A. University of California, Berkeley, CA, USA Michel A. Hofman, Ph.D. Netherlands Institute for Neuroscience, Amsterdam, The Netherlands Arthur MacNeill Horton Jr., Ed.D., A.B.P.P., A.B.N. Neuropsychology section, Psych Associates of Maryland, LLC, Towson, MD, USA Christopher Jones Longwood University, Hull Building, Farmville, VA, USA James C. Kaufman Department of Educational Psychology, Neag School of Education, University of Connecticut, Storrs, CT, USA Philip Lieberman George Hazard Crooker University, Providence, RI, USA Dario Maestripieri, Ph.D. Comparative Human Development, Evolutionary Biology, and Neurobiology, The University of Chicago, Chicago, IL, USA Tara Mandalaywala, Ph.D. Department of Psychology, New York University, New York, NY, USA Department of Comparative Human Development, Institute for Mind and Biology, University of Chicago, Chicago, IL, USA Nancy Mather, Ph.D. University of Arizona, Tucson, AZ, USA Elias Mpofu Faculty of Health Sciences, University of Sydney, Lidcombe, NSW, Australia Ulrich Müller Department of Psychology, University of Victoria, Victoria, BC, Canada Jack A. Naglieri, Ph.D. Department of Virginia, Centerville, VA, USA Thomas Oakland University of Florida, Gainesville, FL, USA Samuel O. Ortiz St. John’s University, Jamaica, NY, USA Tulio M. Otero, Ph.D. The Chicago School of Professional Psychology, Chicago, IL, USA Jonathan A. Plucker Department of Educational Psychology, Univeristy of Connecticut, Neag School of Education, Storrs, CT, USA Dana Princiotta, Ph.D. Tucson, AZ, USA Cecil R. Reynolds, Ph.D. Texas A&M University, Austin, TX, USA Jordan Rigby, Psy.D. Neurology, Learning and Behavior Center, Salt Lake City, UT, USA Sylvia Rodriguez Mindset Works, Walnut, CA, USA Donald H. Saklofske Department of Psychology, University of Western Ontario, Social Science Centre (7th ﬂ oor), London, ON, Canada Deborah Schneider Doctoral Candidate and Department Fellow in Special Education, University of Arizona, Tucson, AZ, USA Contributors xix

W. Joel Schneider Department of Psychology, Illinois State University, Normal, IL, USA Steven J. Stein, Ph.D. Multi-Health Systems, Toronto, ON, Canada Robert J. Sternberg Department of Human Development, B44 MVR, Cornell University, Ithaca, NY, USA Tarmo Strenze, Ph.D. Institute of Social Studies, University of Tartu, Estonia, Tartu, Estonia Lisa A. Suzuki, Ph.D. Department of Applied Psychology, New York University, New York, NY, USA Peggy L. Tarpley Department of Education and Special Education, Longwood University, Farmville, VA, USA Kayla Ten Eycke Behavioural Research Unit, Alberta Children’s Hospital, University of Calgary, Calgary, AB, Canada Fons J.R. van de Vijver Department of Culture Studies, Tilburg University, Tilburg, The Netherlands Hongguo Wei Case Western Reserve University, Cleveland, OH, USA John O. Willis, Ed.D. Department of Education, Rivier University, Peterborough, NH, USA Thomas R. Zentall Department of Psychology, University of Kentucky, Lexington, KY, USA P a r t I Introduction The Evolution of Intelligence 1 Sam Goldstein

Viewed narrowly, there seem to be almost as many deﬁ nitions of intelligence as there are experts asked to deﬁ ne it. R.J. Sternberg

In the last 3,000 years of the written word, the environmental pressures inherent in the intelligence has been defi ned in multiple ways, ecosystem of any species? In this volume, we, including the capacity for abstract thought, along with our contributors, will explore and understanding, communication, planning, learn- attempt to answer these questions. ing, reasoning, and, most importantly, problem solving. Though most widely studied within the human species, the concept and behaviors related The Vocabulary of Intelligence to intelligence have been observed and studied in animals and even plants. It is still the case, how- The word intelligence derives from the Latin ever, that there is no scholarly consensus as to verb intelligere . This verb fi nds its roots in inter- what exactly defi nes intelligence. It is very legere , meaning to “pick out” or discern clearly a construct that has resonated in the minds (Traupman 2007 ). A form of this verb, intellec- of philosophers, scientists, psychologists, and tus, was the medieval technical term for someone educators. It is a concept that has intrigued the with a good understanding as well as a translation general public and been used as the defi ning cri- for the Greek philosophical term nous. Nous, teria for Mensa, an organization touting itself as however, was strongly linked to the metaphysi- allowing membership to only the “most intelli- cal, cosmological theories of teleological scho- gent.” Certainly in the last 100 years, a psycho- lasticism, including theories of the immortality metric approach has been used to defi ne the of the soul and the concept of the active intellect. concept in humans and in doing so offer a means Its entire approach to the study of nature, how- of comparison between individuals. ever, was rejected by modern philosophers, That being said, how can intelligence be con- including Francis Bacon, Thomas Hobbes, John ceptualized within an evolutionary framework? Locke, and David Hume, all of whom preferred What drove all species to develop abilities and the word “understanding” in their English philo- acquire knowledge to enhance their survival? sophical works. Hobbes, in his work Latin De Has the evolution of intelligence been driven by Corpore, used the term “intellectus intelligit” or “the understanding understandeth” as a typical example of a logical absurdity. The term intelligence, therefore, has become less common in the * S. Goldstein ( ) English language philosophy but has been widely University of Utah , 230 South 500 East, Suite 100 , Salt Lake City , UT 84102 , USA adopted in contemporary psychology, absent the e-mail: [email protected] scholastic theories which it once implied.

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 3 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_1, © Springer Science+Business Media New York 2015 4 S. Goldstein

In 1994, an editorial statement in the Wall …the ability to plan and structure one’s behavior Street Journal by 52 researchers defi ned intelli- with an end in view. (Das 1984 , p. 35) An intelligence is the ability to solve problems gence as: or to create products that are valued within one or A very general mental capacity that among other more cultural settings. (Gardner 1993 ) things involves the ability to reason, plan, solve … performing an operation on a specifi c type problems, think abstractly, comprehend complex of content to produce a particular product. ideas, learn quickly and learn from experience. It is (Guilford 1967 ) not merely book learning, a narrow academic skill Intelligence is the ability to learn, exercise judg- or test taking smarts. Rather it refl ects a broader ment and be imaginative. and deeper capability for comprehending our sur- Intelligence is a general factor that runs through roundings, catching on, making sense of things, or all types of performance. (Jensen 1998 ) fi guring out what to do. (Gottfredson 1997 p. 20) Intelligence is assimilation to the extent that it incorporates all the given data of experience within A report published by the Board of Scientifi c its framework. (Piaget 1963 ) Affairs at the American Psychological Association … ability to carry on abstract thinking. (Terman in 1996 titled Intelligence: Knowns and Unknowns 1922 ) The term intelligence designates a complexly defi ned intelligence as (Neisser et al. 1996 ): interrelated assemblage of functions, no one of Individuals differ from one another in their ability which is completely or accurately known in man…. to understand complex ideas to adapt effectively to (Yerkes and Yerkes 1929 ) the environment, to learn from experience, to The ability to deal with cognitive complexity. engage in various forms of reasoning, to overcome (Gottfredson 1998 , p. 25) obstacles by taking thought. Although these indi- … goal directed adaptive behavior. (Sternberg vidual differences can be substantial, they are 1984 , p. 3) never entirely consistent: A given person’s intel- Unique propensity of human beings to change lectual performance will vary on different occa- or modify the structure of their cognitive function- sions, in different domains as judged by different ing to adapt to the change and demands of a life criteria. Concepts of “intelligence” are attempts to situation. (Feurestein et al. 1979 /2002) clarify and organize this complex set of phenom- … the mind is not a complex network of gen- ena. Although considerable clarity has been eral capabilities such as observation, attention, achieved in some areas, no such conceptualization memory, judgment and so forth, but a set of spe- has yet answered all the important questions and cifi c capabilities, each of which is to some extent non-commands universal assent. Indeed, when two independent of the others and has developed inde- dozen prominent theorists are recently asked to pendently. (Vygotsky 1978 ) defi ne intelligence they gave two dozen somewhat different defi nitions. (p. 77) Most certainly, the evolution of intelligence as It seems to us that in intelligence there is a a force driven by need over millions of years fundamental faculty, the alteration or the lack of across multiple species has led to often fascinat- which is of utmost importance in practical life. ing ideas and conceptual processes. This faculty of judgment is otherwise called good The time line of human evolution spans sense, practical sense, initiative, or the faculty of approximately seven million years (Berstrom and adapting one’s self to circumstances (Binet and Dugatkin 2011 ) from the separation of the Pan Simon 1905 ). Many of the most infl uential psy- genus until the emergence of behavioral moder- chologists in the fi eld of intelligence have also nity 50,000 years ago. The fi rst three million weighed in on the concept of intelligence, defi n- years of this time line concern Sahelanthropus . ing intelligence as: The next two million years concern The aggregate or global capacity of the individual Australopithecus , while the fi nal two million to act purposely, to think rationally and deal effec- years span the actual history of human tively with his environment. (Wechsler 1944 , p. 3) (Paleolithic) traits considered to refl ect human Intelligence is not a single, unitary ability, but intelligence, including empathy, theory of mind, rather a composite of several functions. (Anastasi 1992 , p. 613) mourning, and ritual, and the use of symbols of We shall use the term intelligence to mean the tools is apparent in great apes as well as other ability of an organism to solve new problems…. species although in lesser sophistication than in (Bingham 1937 ) humans (Matsuzawa 2001 ). However, the great 1 The Evolution of Intelligence 5 apes demonstrate considerable ability for information, and sequence their activities in cognition and empathy. Chimpanzees are capable order to enhance survival. of making tools, using them to acquire food and Homo habilis appeared in East Africa 2.4 mil- for social displays. They are strategic hunters lion years ago. The fi rst known human species, working cooperatively and organized by senior- Homo habilis , was also the fi rst known species to ity. They are status conscious, manipulative, and make stone tools. The use of such tools conferred capable of deception. They can learn to use sym- a crucial evolutionary advantage that required a bols and understand human language, including larger, more sophisticated brain to coordinate the some level of relational syntax and concepts of fi ne hand movements required for these tasks. number and numerical sequence (Boysen and The evolution of a larger brain created a problem Berntson 1989 ; Hirata 2009 ). for early humans, however. A larger brain It is thought that ten million years ago when requires a larger skull, and thus, females have to the Earth’s climate entered a cooler and dryer develop a wider birth canal for a newborn’s larger phase leading to the ice age 2.6 million years ago, skull to pass through. However, if the female’s one consequence was that the North African trop- birth canal grew too wide, her pelvis would be so ical forest began to retreat, being replaced fi rst by wide that she would lose the ability to run, still a open grasslands and eventually by desert. It is necessary skill in the dangerous world of two believed this forced tree-dwelling animals to million years ago. Evolution’s solution was to adapt to this new environment. When the envi- give birth at an early stage of fetal development ronment changed from continuous forest to before the skull was too large to pass through the patches of forest separated by expanses of grass- birth canal and have the plates of the skull uncon- land, some primates adapted to a partly or fully nected. These adaptations enabled the human ground-dwelling life. brain to continue to grow, but it imposed a new Early hominid species may have adapted to discipline to care for helpless infants for long this challenge by evolving to pedalism or walk- periods of time and force humans to become less ing on their hind legs. This gave their eyes mobile. Unlike salmon, snakes, bears, or pri- greater elevation and the ability to see approach- mates, the human species appeared to abandon ing danger from further off. It also freed the all efforts at self-preservation (thick skin or hair) arms from the task of walking and made the in exchange for developing big brains. This hands available for tasks such as gathering food. greatly expanded childhood and required humans At some point, bipedal primates developed to increasingly stay in one place for longer peri- handedness, giving them the ability to pick up ods of time so females could care for infants sticks, bones, and stones and use them as weap- while males hunted and fought with other bands ons or as tools for tasks such as killing smaller competing for food sources. As a result, humans animals, cracking nuts, or cutting up carcasses. became even more dependent on tool making to These primates developed the use of primitive compete with other animals and other humans, technology. Bipedal tool-using primates from a relying less on body size and strength for sur- Hominina subtribe of which the earliest species vival. About 200,000 years ago, Europe and the such as Sahelanthropus tchadensis date to about Middle East were colonized by Neanderthals. seven to fi ve million years ago. From that point They were extinct by 20,000 following the forward, the hominid brain began to develop appearance of modern humans in the region some rapidly in both size and differentiation of func- 40,000 years ago (Klein 2003 ). In this same tion. These early patterns of behavior very period of time, Neanderthals inhabited Europe, clearly refl ect an effort to problem solve and and Homo sapiens fi rst appeared in East Africa. adapt to the environment and thus likely the ori- It is unclear to what extent these early modern gins of modern intelligence. These early species humans had developed language, music, or reli- needed to learn to plan, attend to relevant detail, gion, but they spread through Africa over the fol- simultaneously process multiple sources of lowing 50,000 years or so. It is thought that their 6 S. Goldstein ability to harness language to communicate and successfully judge male intelligence, they must problem solve as well as the capacity to think one be intelligent themselves. thing but say or do something else facilitated As advancement, survival, and reproduction their spread and offered an advantage over other within an increasing complex social structure hominid species. Rapidly increasing sophistica- favored even more advanced social skills, con- tion and tool making and behaviors apparent by ceptual ability, reasoning, and problem solving, it 80,000 years ago in the migration out of Africa is not diffi cult to appreciate the rapid growth of followed toward the very end of the Middle intelligence in our species. Since competition Paleolithic some 60,000 years ago. Fully shifted bit by bit from controlling nature to infl u- modern behaviors, including art, music, self- encing other humans, it becomes of relevance to ornamentation, trade, burial rights, etc., are evi- outmaneuver other members of the group, seek- dent by 30,000 years ago. The oldest unequivocal ing leadership or acceptance by means of more example of prehistoric art dates to this period. advanced social skills and problem-solving Thus, it would appear that human intelligence did abilities. not evolve as a means of just solving ecological There is also no doubt that higher intellectual problems but as a means of surviving and repro- functioning develops better in an environment ducing in large and complex social groups. Some with adequate nutrition. Defi ciency in iron, zinc, of the behaviors associated with living in large protein, iodine, B vitamins, omega 3 fatty acids, groups include reciprocal altruism, deception, magnesium, and other nutrients either in the and coalition formation. These group dynamics mother during pregnancy or in the child during relate to the theory of mind or the ability to development results in impaired development understand the thoughts and emotions of others. (Frensham et al. 2012 ). The contribution of nutri- As hominids started living in large groups, selectional factors, including shifting from agrarian- tion must have favored effective problem solving ism to meat-eating diets, is thought to also have leading to greater intelligence. Yet not all theo- made a signifi cant contribution to the rapid rists agree. Geoffrey Miller argues that human growth in intelligence and human species over intelligence is unnecessarily sophisticated for the the last 50,000 years. needs of hunter-gatherers to survive. He argues that intelligent behaviors such as language, music, and art were of no utilitarian value to the About This Book survival of ancient hominids. Rather, intelligence may have been a fi tness indicator. Hominids Tracing the evolutionary and cultural roots of would have selected for greater intelligence as a intelligence is neither easy nor straightforward. proxy for healthy genes, and a positive feedback As far as we are aware, these two topics have loop of sexual selection would have led to the never been extensively presented simultaneously evolution of human intelligence in a relatively in a single volume. Following the introduction the short period of time. Such a theory must also book begins with seven background chapters. Our explain why both genders are intelligent. In many contributors explore the concept of intelligence in species, only males have impressive ornaments non-primates and nonhuman primates. As author and show-off behaviors. However, it is thought Philip Lieberman points out, our species took a that sexual selection may act on males and fast track in developing intelligence as we con- females even in species that are at least partially comitantly developed language. Author Michael monogamous. Sexual selection has clearly Hoffman emphasizes our knowledge of the evolu- occurred for other female-specifi c human traits, tion of the human brain and the conceptual devel- for example, breasts and buttocks. Sexual selec- opment of intelligence. The fi nal three chapters tion for intelligence and judgment can act as indi- of the Background section explore philosophers cators of success such as highly visible displays beginning with Plato and Pascal through Darwin, of wealth. Finally, if it is possible for females to Charcot, and Galton. The Background section 1 The Evolution of Intelligence 7 ends with an important discussion by David Geary Feurestein, R., Feurerstein, S., Falik, L., & Rand, Y. on the role social competition has likely played in (1979/2002). Dynamic assessments of cognitive modifi ability. ICELP Press: Jerusalem. pushing the evolution of fl uid intelligence. The Frensham, L. J., Bryan, J., & Parletta, N. (2012). third section contains eleven chapters explor- Infl uences of micronutrient and omega-3 fatty acid ing and differentiating historical roots of intelli- supplementation on cognition, learning, and behavior: gence beginning with early scientifi c researchers Methodological considerations and implications for children and adolescents in developed societies. such as James, Wundt, and Cattell, discussing the Nutrition Reviews, 70 , 594–610. work of Piaget and Binet and progressing through Gardner, H. (1993). Frames of mind: Theory of multiple the pivotal work of A. R. Luria, setting the stage intelligences . California: Fontana Press. for current conceptualization of multiple intelli- Gottfredson, L. S. (1997). Mainstream science on intelligence: An editorial with 52 signatories, history and gences, parallel types of intelligence, and intelli- bibliography. Intelligence, 24 (1), 13–23. gence as a malleable, creatively driven concept. Gottfredson, L. S. (1998). The general intelligence factor. The next section contains fi ve chapters presenting Scientifi c American Presents, 9 , 24–29. the long and fascinating history between theories Guilford, J. P. (1967). The nature of human intelligence . New York: McGraw-Hill. of intelligence and intelligence tests. Chapter 20 Hirata, Y. (2009). Factors affecting the perception of Japanese by Jack A. Naglieri explores the past 100 years moraic rhythm by second language learners. Journal of of intelligence testing and offers a glimpse into the Phonetic Society of Japan, 13 (3), 33–43. the future. The next three chapters discuss appli- Jensen, A. R. (1998). The g factor: The science of mental ability (Human evolution, behavior and intelligence). cations of intellectual theory as it relates to the Westport: Praeger. economic structure of our society, executive func- Klein, R. G. (2003). Whither the Neanderthals. Science, tioning, and educational achievement. Our book 299 , 1525–1527. concludes with a seminal chapter by James Flynn Matsuzawa, T. (2001). Primate foundations of human intelligence: A view of tool use in non-human pri- and our concluding comments. mates and fossil hominids. In M. Matsuzawa (Ed.), Primate origins of human cognition and behavior . New York: Springer. Neisser, U., Boodoo, G., Bouchard, T. J., Boykin, W., References Brody, N., et al. (1996). Intelligence: Knowns and unknows. American Psychologist, 51 , 77–101. Anastasi, A. (1992). What counselors should know about Piaget, J. (1963). The psychology of intelligence . the use and interpretation of psychological tests. New York: Routledge. Journal of Counseling and Developments, 70 , Sternberg, R. J. (1984). Toward a triarchic theory of 610–615. human intelligence. Behavioral and Brain Sciences, 7 , Berstrom, C. T., & Dugatkin, L. A. (2011). Evolution . 269–287. New York: W.W. Norton and Company. Terman, L. M. (1922). A new approach to the study of Binet, A., & Simon, T. (1905). Methods nouvelles por le genius. Psychological Review, 29 (4), 310–318. diagnostic du niveal intellectual des anormaux. Nal’ Traupman, J. (2007). The Bantam new college Latin and Annë Psychologique, 11 , 191–244. English dictionary (3rd revised ed.). New York: Bantam. Bingham, W. V. (1937). Aptitudes and aptitude testing . Vygotsky, L. S. (1978). Interaction between learning and New York: Harper and Brothers. development. In L. Vygotsky (Ed.), Mind and society Boysen, S. T., & Berntson, G. G. (1989). Numerical com- (pp. 79–91). Cambridge: Harvard University Press. petence in a chimpanzee (Pantroglodytes). Journal of Wechsler, D. (1944). Measurement of adult intelligence . Comparative Psychology, 103 , 23–31. Baltimore: The Williams and Wilkins Company. Das, J. P. (1984). Aspects of planning. In J. R. Kirby (Ed.), Yerkes, R. M., & Yerkes, A. W. (1929). The great apes: A Cognitive strategies and educational performance study of anthropoid life . New Haven: Yale University (pp. 35–50). Orlando: Academic. Press. Part II Background Intelligence in Nonprimates 2 Thomas R. Zentall

Humans tend to have an anthropocentric view of were hit by a large asteroid or suffered a self- intelligence that views them at the top and quite infl icted nuclear disaster), many simpler organ- often animals that look like us close behind. isms would likely survive much better than large Although the notion of an evolutionary scale with intelligent animals like us. humans at the top is popularly held, it is also self- From a purely biological perspective, the ideal serving. We tend to overvalue our problem- survival machine is a simple, one-celled, organ- solving ability, our capacity to modify our ism (e.g., the amoeba) that has survived as a spe- environment, and our ability to communicate with cies in one of two ways. Either it has needed to each other. Conversely, we tend to undervalue the undergo little change in morphology or behavior exceptional sensory skills of other animals, for for millions of years because it exists in a remark- example, the tracking and drug-detecting ability ably stable (predictable) environment, in which of dogs; the navigational abilities of homing case there has been little need for change, or if its pigeons, whales, and monarch butterfl ies; and the environment does change, it relies on natural ability of birds of prey to detect the minute move- selection by means of very rapid reproduction ment of a small animal on the ground far below and mutation (e.g., bacteria and viruses). This them. The role of our intelligence in the domina- ability to reproduce quickly and often, ensures tion of our species over others seems obvious, but the survival of many of these organisms (albeit in the broader perspective of evolutionary suc- not necessarily in the same form) even in the cess, as measured by the number of surviving event of a major catastrophe. Many other organ- members of a species, intelligence, as a general isms whose rate of reproduction has not been characteristic, correlates somewhat negatively able to keep up with relatively rapid changes in with most measures of evolutionary success. the environment have relied on the ability to Consider the relatively small numbers of our clos- modify their behavior during their lifetime. est relatives, the great apes, compared with the Intelligence, in its simplest form, can be thought large numbers of physiologically simpler insects, of as the fl exibility endowed by our genes that bacteria, and viruses. And it is estimated that if a allows organisms to adjust their behavior to rela- massive disaster were to occur (e.g., if the Earth tively rapidly changing environments. For some animals, a stable supply of a highly specifi c food may be predictable (e.g., eucalyptus leaves for * T. R. Zentall ( ) the koala or bamboo leaves for the giant panda)— Department of Psychology , University of Kentucky Lexington , KY 40506 , USA at least until recently. For most animals, however, e-mail: [email protected] environments are much less predictable, and their

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 11 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_2, © Springer Science+Business Media New York 2015 12 T.R. Zentall predisposed eating preferences have had to be running speed), and our relative physical weak- much more fl exible. For still other animals, the ness may not have allowed us to hunt competi- environment is suffi ciently unpredictable that it is tively with other predators (e.g., large cats). The impossible to for them to be predisposed to know competition with other animals for food must (by genetic means) what food will be available have come about slowly enough for us to develop (consider the varied diet of the city-dwelling rat). weapons and tools, complex forms of communi- For these animals to survive, more general cation (language), and complex social structure (abstract) rules must be available. Rules about (allowing for cooperation, teamwork, and recip- what to eat may not be based on the sight or taste rocation). According to this view, although our of what is ingested but on its consequences. intellect appears to have given us a clear advan- Instead of instructing the animal to eat eucalyp- tage over other animals, its evolution is likely to tus leaves or to eat a certain class of seeds, these have emerged because of our relative weakness genes instruct the animal that if it feels sick after in other areas. Other animals have compensated eating a new food, it should avoid eating more of for their weaknesses by developing strengths in that food. Such general rules allow for the behav- nonintellectual areas (e.g., the snail compensates ioral fl exibility that we call learning. for its lack of rapid mobility by building a pro- But there is a price to pay for this added fl ex- tective shell around itself). Discussions of ani- ibility. The animal must sometimes suffer the mal intelligence often assume, inappropriately, consequences of eating something bad. If the that intelligence is inherently good. In our case, novel food is poisonous, the animal may not sur- it has turned out to be generally true (at least to vive to use its newfound knowledge. The creation the present). For us, intelligence has had a run- and maintenance of a nervous system capable of away effect on our ability to adapt to change (an such learning represents a cost as well. For many effect that Dawkins 1976 calls hypergamy), animals, the benefi ts of the capacity for simple which has allowed us to produce radical changes associative learning outweigh the cost, but for in our environment. However, from a biological some animals, the negative consequences of trial perspective, in general, intelligence can be and error learning are suffi ciently costly that sim- viewed as making the best out of a bad situation, ple learning rules are not enough. or producing a complex solution to problems Some animals have found ways to reduce this that other species have often solved in simpler cost. Rats, which live in highly unpredictable ways. As we evaluate the various intellectual environments, have evolved the ability to learn, capacities of nonhuman animals, let us try to in a single experience, the consequences of eat- keep in mind that they have survived quite well ing a small amount of a novel food, even when (until recently) without the need for our complex those consequences are experienced hours after intellectual skills. the food was ingested (Garcia and Koelling 1966 ). Rats also have developed the ability to transmit food preferences socially. If a rat experi- The Comparative Approach: ences the smell of a novel food on the breath of Two Caveats another rat, it will prefer food with that smell over another equally novel food (Galef 1988 ), First, most people have a vague idea of the rela- and it may also be able to assess the consequences tive intelligence of animals. As a general rule, to the other rat of having eaten a novel food those species that are more like us physically are (Kuan and Colwill 1997 ). judged to be more intelligent. But we must be But what if this degree of fl exibility in learn- careful in making such judgments because we ing is still not enough to allow for survival? In humans are the ones who are defi ning intelligent the case of humans, for example, our poorly behavior. We make up the rules and the testing developed sense of smell, our relatively poorly procedures, and those tests may be biased in favor developed gross motor response (e.g., slow of our particular capacities. Isn’t it interesting that 2 Intelligence in Nonprimates 13 animals that are more similar to us, that have similar sensory, motor, and motivational systems, Absolute Versus Relational Learning just happen to be judged as more intelligent? Bitterman (1975 ) has suggested that a rela- One of the most basic cognitive functions tional view of animal learning can be used to cor- involves not being bound to the absolute proper- rect for peripheral differences in sensory capacity ties of a stimulus. Although Hull (1943 ) claimed and motor coordination. He suggests that rather that learning involves solely the absolute proper- than looking for differences in the rate at which ties of a stimulus, he proposed that animals will different species can learn, we might look at dif- appear to respond relationally because they will ferences, for example, in an animal’s ability to respond similarly to similar stimuli, a process learn from the experience of learning. In other known as stimulus generalization. Spence (1937 ) words, to what extent can learning facilitate new elaborated on this theory by proposing that dis- learning (learning to learn)? Then, using the rate crimination learning establishes predictable gra- of original learning as a baseline, one can deter- dients of excitation (approach) and inhibition mine the degree to which later learning, presum- (avoidance) that summate algebraically. And this ably involving the same processes, is facilitated. theory of generalization gradient summation can However, this approach is not always possible, account for a number of phenomena that were and we must be aware that our assessment may formerly explained as relational learning (see be biased by the use of testing procedures not Riley 1968 ). The fact that one sees little discus- well suited for the species we are studying. sion of this issue in the modern literature sug- Second, we must guard against the opposite gests that animals are capable of using either the bias—the tendency to interpret behavior as intel- absolute or relative properties of a stimulus in ligent because of its similarity to intelligent making discriminations. human behavior. In evaluating research addressing the cognitive capacity of animals, we should adopt C. Lloyd Morgan’s (1894 ) position that it is Learning to Learn not necessary to interpret behavior as complex (more cognitive) if a simpler (less cognitive) Can an animal use prior learning to facilitate new account will suffi ce. This is the principle of par- learning? That is, can animals learn to learn? If simony. Thus, higher-level cognitive interpreta- an animal learns a simple discrimination between tions should always be contrasted with simpler, two stimuli (an S+, to which responses are rein- contiguity- and contingency-based, associative- forced, and an S−, to which responses are extin- learning accounts. I will start with several classi- guished) and then, following acquisition, the cal issues concerned with the nature of learning discrimination is reversed (the S+ becomes S− and intelligence in animals, move to more com- and the S− becomes S+), and then reversed again, plex behavior thought to be uniquely human, and repeatedly, are successive reversals learned faster end with examples of presumably complex than earlier reversals? Animals trained on such a behavior that are likely to be based on simpler serial-reversal task often show improvement predisposed processes. within a few reversals, and the rate of improve- This review will focus primarily on the fl exi- ment can be used as a measure of learning to ble behavior of nonprimates because the cogni- learn. For example, rats show more improvement tive behavior of primates is covered elsewhere in than pigeons, and pigeons show more improve- this volume, and thus, it will not address several ment than gold fi sh (Bitterman and Mackintosh areas of research that have been conducted 1969 ). Mackintosh ( 1969 ) attributes these differ- uniquely with primates, such as analogical rea- ences in serial-reversal learning to the differential soning, conservation of volume and mass, lan- ability of these species to maintain attention to guage, perspective taking, theory of mind, and the relevant dimension and ignore irrelevant deception. dimensions. 14 T.R. Zentall

A different approach to learning to learn is to the exemplars and the similarities in the underly- look for improvement in the rate at which dis- ing processes for different species. For example, criminations involving new stimuli are learned. pigeons make errors similar to those of young This phenomenon, known as learning set (Harlow children (e.g., they often erroneously assign a 1949 ), has been studied primarily using visual picture of a bunch of celery or an ivy- covered discriminations with monkeys, but good evidence wall to the category “tree”). for such effects has also been found with olfactory discriminations with rats (Slotnick and Katz 1974 ). In the limit, learning of a new discrimina- Equivalence Relations tion, or of a reversal, can occur in a single trial. When it does, it is referred to as a win-stay-lose- The emergent relations that may arise when arbi- shift strategy because stimulus choice is com- trary, initially unrelated stimuli are associated pletely controlled by the consequences of choice with the same response are often referred to as on the preceding trial. One means of developing functional equivalence because they belong to a such a strategy is to learn to forget the conse- common stimulus class (see Zentall and Smeets quences of trials prior to the immediately preced- 1996 ). The best example of equivalence relations ing trial. In fact, research has shown that memory in humans is that aspect of language known as for the specifi c characteristics of the stimuli from semantics—the use of symbols (words) to stand prior discriminations does decline as the number for objects, actions, and attributes. What makes of discriminations learned increases (Meyer these relations so powerful is what one learns 1971 ). Thus, animals approach optimal learning about one member of the stimulus class (i.e., a by learning to ignore the effects of all but the word) will transfer to others (i.e., the object that most recent experience. it represents). Thus, a child can be told about the varied behavior of dogs (sometimes friendly but not always) without having to actually experi- Stimulus Class Formation ence them (and getting bitten). Thus, stimuli that belong to the same stimulus class can be thought Perceptual Classes of as having the same meaning. The most common procedure for demonstrating the develop- Pigeons are remarkably adept at responding ment of functional equivalence in animals selectively to photographs of natural scenes, involves training on two conditional discrimina- depending on whether the scene involves a tions. In the fi rst, for example, a red hue (sample) human form (Herrnstein and Loveland 1964 ) or signals that a response to a circle will be rein- trees or water (Herrnstein et al. 1976 ) and those forced (but not a response to a dot), and a green objects need not be anything that they might have hue signals that a response to a dot will be rein- actually encountered in their past (e.g., underwa- forced (but not a response to a circle; see ter pictures of fi sh; Herrnstein and deVilliers Fig. 16.1). In the second conditional discrimina- 1980 ). To demonstrate that the pigeons do not tion, a vertical line signals that a response to the simply memorize a list of pictures and their circle will be reinforced (but not a response to the appropriate responses, Herrnstein et al. showed dot), and a horizontal line signals that a response that the pigeons would respond appropriately to to the dot will be reinforced (but not a response to new examples of the positive and negative stimu- the circle). Thus, the red hue and vertical line can lus sets. be described as meaning choose the circle and the What is interesting about perceptual classes is green hue and horizontal line as choose the dot. that it is diffi cult to specify what features humans This procedure has been referred to as many-to- or pigeons use to discriminate members from one matching because training involves the asso- nonmembers of the perceptual class. However, ciation of two samples with the same comparison examination of the kinds of errors made can tell stimulus. To show that an emergent relation has us about the attributes that were used to categorize developed between the red hue and the vertical 2 Intelligence in Nonprimates 15

Memory Strategies

The task most often used to study memory in animals is delayed matching-to-sample, in which following acquisition of matching-to-sample, a delay is inserted between the offset of the sample and the onset of the comparison stimuli (Roberts and Grant 1976 ). However, the retention functions typically found with this procedure generally greatly underestimate the animal’s memory capacity for two reasons. First, in many studies, the novel delay interval is quite similar in appearance to the time between trials. This leads to an ambiguity in the meaning of the delay. When the delay interval and the intertrial interval are made distinctive, the retention functions obtained often provide a very different picture of the animal’s memory (Sherburne et al. 1998). Second, the novelty of the delays may result in a generalization decrement that is con- founded with memory loss. When pigeons are trained with delays, considerably better memory has been found (Dorrance et al. 2000 ). Of more interest in the assessment of animal intelligence is Fig. 16.1 Many -to-one matching training used to show an animal’s ability to actively affect its memory. that pigeons will learn that red and vertical (as well as green and horizontal) “mean the same thing.” If red and green samples are now associated with new comparison stimuli, blue and white , respectively, there is evidence that Prospective Processes vertical and horizontal lines are also associated with the blue and white stimuli, respectively Traditionally, animal memory has been viewed as a rather passive process. According to this view, sen- line and between the green hue and the horizontal sory events can leave a trace that may control line, one can train new associations between one responding even when the event is no longer pres- pair of the original samples (e.g., the red and ent (Roberts and Grant 1976 ). However, there is green hues) and a new pair of comparison stimuli evidence that animals can also actively translate or (e.g., blue and white hues, respectively). Then on code the representation of a presented stimulus into test trials, one can show that emergent relations an expectation of a yet-to-be- presented event have developed when, without further training, (Honig and Thompson 1982). What does it mean to an animal chooses the blue hue when the sample have an expectation of a future event? Imagine a is a vertical line and chooses the white hue when delayed matching task in which vertical- and hori- the sample is a horizontal line (Urcuioli et al. zontal-line samples are mapped onto red and green 1989 ; Wasserman et al. 1992 ; Zentall 1998 ). comparison stimuli. During the delay, one can Although pigeons are not capable of language imagine that some representation of the just seen learning, the ability of small-brained organisms sample stimulus would be remembered. But it is like pigeons to develop arbitrary stimulus classes, also possible that the sample is translated into a the main characteristic of symbolic representa- response intention to select one of the comparison tion, suggests that this capacity is much more stimuli. The ability to use expectations, or prospec- pervasive than once thought. tive coding processes, has important implications 16 T.R. Zentall for the cognitive capacities of animals. If the expec- Planning Ahead One of the hallmarks of tation of a stimulus, response, or outcome can serve human cognitive behavior is our ability to as an effective cue for comparison choice, it sug- consciously plan for the future. Although gests that animals may be capable of exerting active animals sometimes appear to plan for the future control over memory, and in particular, it may sug- (birds build nests; rats hoard food), these gest they have the capacity for active planning. behaviors are likely to be under genetic control. The notion of expectancy as an active pur- That is, animals do it but it is not likely to be posive process can be attributed to Tolman with the expectation of later use. Alternatively, (1932 ). Although one can say that a dog sali- what appears to be future planning just may be vates when it hears a bell because it expects the ability to delay reinforcement. To distinguish food to be placed in its mouth, the demonstra- between planning for the future and learning tion that an expectation can serve as a discrimi- with a long delay of reinforcement, Suddendorf native stimulus (i.e., as the basis for making a and Corballis (1997 ) have suggested that the choice) suggests that the expectancy has addi- behavior indicative of planning must occur in tional cognitive properties. the absence of the relevant motivation. Roberts (2002 ) reported the absence of planning by The Differential Outcome Effect If a monkeys. After they had eaten a portion of their conditional discrimination or matching task is daily allotment of food, they threw out of their designed such that a correct response following cage whatever food remained but then requested one sample results in one kind of outcome (e.g., more food later in the day. However, convincing food) and following the other sample results in a evidence for planning was reported by Raby different kind of outcome (e.g., water), one can et al. (2007 ). Western scrub jays, which cache show that acquisition of the conditional food for future use, learned that unpredictably, discrimination is faster (Trapold 1970 ) and they would either spend the night in a compartment retention is better when a delay is inserted in which in the morning they would fi nd one kind between the conditional and choice stimuli of food (peanuts) or in a compartment in which (Peterson et al. 1980 ). Furthermore, there is they would fi nd a different kind of food (kibble). evidence from transfer-of- training experiments On test trials, the night before, they were allowed that in the absence of other cues, outcome to eat and cache food in either compartment. anticipations can serve as suffi cient cues for When they were given peanuts, they tended to comparison choice. That is, if the original samples cache them in the kibble compartment, and when are replaced by other stimuli associated with the they were given kibble, they tended to cache them same differential outcomes, positive transfer has in the peanut compartment (i.e., they cached the been found (Edwards et al. 1982 ; Peterson 1984 ). food in the compartment in which they would not This line of research indicates that presentation of fi nd that particular food in the morning). a sample creates an expectation of a particular kind of outcome and that expectation alone can then serve as the basis for comparison choice. In Directed (Intentional) Forgetting most cases, the differential outcomes have differential hedonic value (e.g., a high probability The notion of directed or intentional forgetting of food versus a low probability of food), and it is is borrowed from human memory research. It possible that outcome anticipation can elicit implies that memory is an active rather than a differential emotional states in the animal. But passive (automatic) process. Presumably, fol- there is also evidence that nondifferentially lowing presentation, items that human partici- hedonic events such as the anticipation of a pants are instructed to forget may not be well particular neutral stimulus can affect response stored or maintained in memory and, thus, accuracy (Kelly and Grant 2001 ; Miller et al. should not be well retained. In a directed 2009 ; Williams et al. 1990 ). forgetting task with animals, for example, 2 Intelligence in Nonprimates 17 pigeons are trained on a matching task, and then Tulving ( 1972 ) proposed that an episodic a delay of a fi xed duration is introduced between memory should include the what, where, and the sample and the comparisons. On remember when of an experience. Clayton and Dickinson trials, during the delay, the pigeons are cued that (1999 ) showed that western scrub jays that there will be a test of sample memory, whereas cached peanuts and wax worms (what) on one on forget trials, the pigeons are cued that there side or the other of an ice cube tray (where) will be no test of sample memory. On selected learned that their preferred wax worms would be probe trials, the forget cue is presented, but edible after one day but after four days only the there is a test of sample memory. Matching peanut would be edible (when; see also Babb and accuracy on these probe trials is generally below Crystal 2006 , for a similar fi nding with rats). But that of remember trials on which there was an it can be argued that it is insuffi cient to retrieve expected test of sample memory (Grant 1981 ). the what, where, and when of an episode because But this design confounds differential motiva- those have been explicitly trained (i.e., they are tion on remember and forget trials with sample likely to be semantic or rule-based memories). memory effects because food can be obtained Instead, better evidence for episodic memory only on remember trials. In a more complex would come from the fi nding that animals can design that controls for motivational effects and retrieve information about a past episode when that better approximates the human directed for- there is no expectation that they will be requested getting procedure by allowing the animal to to do so in the future (Zentall et al. 2001 ). That is, reallocate its memory from the sample to an imagine that pigeons are fi rst trained to report the alternative memory on forget trials in training, location where they recently pecked (instruc- better evidence for directed forgetting in pigeons tions) and then they are trained on an unrelated has been demonstrated (Roper et al. 1995 ). conditional discrimination in which choice of a Thus, under certain conditions, it appears that vertical line was correct when the sample was animals do have at least some active control blue and choice of the horizontal line was correct over memory processes. when the sample was yellow. Singer and Zentall ( 2007 ) found that on probe trials on which following a vertical- or horizontal-line comparison Episodic Memory response the pigeons were asked unexpectedly to report the location that they had pecked, they reli- Human memory can be identifi ed by the kinds of ably did so. Thus, by either criterion (what- processes presumed to be involved. Procedural where- when or responding to an unexpected memory involves memory for actions (e.g., riding question), pigeons show some evidence of a bicycle), and it has been assumed that most episodic- like memory. learned behavior by animals involves this kind of memory. Human declarative memory is assumed to be more cognitive because it Navigation involves memory for facts (semantic memory) and memory of personal experiences (episodic Compared to many animals, humans have rela- memory). Although animals cannot typically tively poor navigational skills. Consider how describe factual information, their conditional dependent we are on external supports such as rule-based learning can be thought of as a kind compasses, maps, and more recently global posi- of semantic memory (e.g., if the sample is red, tioning devices. Many animals (e.g., migrating choose the vertical line; if the sample is green, whales, birds, monarch butterfl ies) can navigate choose the horizontal line). But do animals over many hundreds of miles using magnetic have episodic memory? Can they retrieve per- fi elds, chemical gradients, and star patterns. And sonal experiences or do they simply remember homing pigeons use a number of these naviga- the rules. tional systems including landmarks consisting of 18 T.R. Zentall natural and man-made geographic features (Lipp obtainable in infra humans, its occurrence et al. 2004 ). requires considerable environmental support” However, many humans have the ability to (Davis and Memmott, p. 566). In contrast, imagine a route that they will take and even to Capaldi (1993 ) concludes that under the right imagine how to get to a familiar destination by a conditions, animals count routinely. In simple but novel path. This ability, known as cognitive map- elegant experiments, Capaldi and Miller (1988 ) ping, consists of mentally knitting together land- demonstrated that following training, rats can marks one has experienced, such that the relation anticipate whether they will get fed or not for among them can be used to determine a novel path running down an alley depending solely on the to arrive at a goal. Landmarks are needed to form number of successive times they have run down a cognitive map, but they should not be necessary that alley and found food or the absence of food to use it. Can animals form a cognitive map? on successive earlier trials. Before trying to answer this question, we need The difference in the conclusions reached by to make an important distinction. Some animals Davis and Memmott (1982 ) and by Capaldi and have the remarkable ability to navigate in the Miller (1988 ) has general implications for the absence of landmarks or other external cues. This study of intelligence in animals (including ability, known as path integration (or dead reck- humans). The context in which one looks for a oning), involves the representation of direction particular capacity may determine whether one and distance one has traveled from a starting will fi nd evidence for it. As noted earlier, because point. Dessert ants are particularly adept at path we, as human experimenters, devise the tasks that integration as can be shown not only by the direct serve as the basis for the assessment of intelli- path that they take to return to their nest after a gence, we must be sensitive to the possibility that foraging trip but also by the systematic error these tasks may not be optimal for eliciting the incurred if they are displaced just before they behavior we are assessing. That is, much of our attempt to return home (Collette and Graham view of the evolutionary scale of intelligence 2004). The distinction between path integration may be biased in this way by species differences and cognitive mapping has been a point of con- in sensory, response, and motivational factors. troversy. However, under conditions that cannot be accounted for with either landmark use or path integration, there is evidence for the development Reasoning of a simple cognitive map in rats (Singer et al. 2007 ) and dogs (Chapuis and Varlet 1987 ). Reasoning can be thought of as a class of cognitive behavior for which correct responding on test trials requires an inference based on incomplete Counting experience. Although, for obvious reasons, most research on reasoning in animals has been done The term numerical competence is often used in with higher primates (e.g., chimpanzees), there is animal research because the more common term, evidence that some reasoning-like behavior can counting, carries with it the surplus meaning that be demonstrated in a variety of species. accompanies the human verbal labels given to In its simplest form, the transitive inference numbers. That this distinction is an arbitrary one, task can be described as follows: if A is greater based on limitations of response (output) capac- than B (A > B), and B is greater than C (B > C), ity rather than conceptual ability, is suggested by then it can be inferred that A > C (where the let- Pepperberg’s (1987 ) work with generalized ver- ters A, B, and C represent arbitrary stimuli). bal number use in an African gray parrot. Correct responding on this relational learning An excellent review of the animal counting task requires that an inference be made about the literature is provided by Davis and Memmott relation between A and C that can only be derived ( 1982 ), who conclude that “although counting is from the two original propositions. To avoid 2 Intelligence in Nonprimates 19 potential problems with end-point effects that value that transfers from the positive to the nega- could produce a spurious nonrelational solution tive stimulus in a simultaneous discrimination. (i.e., A is always greater, and C is never greater), experimental research typically uses a task that involves four propositions: A > B, B > C, C > D, Taking the Perspective of Others and D > E, and the test involves the choice between B and D, each of which is sometimes An organism can take the perspective of another greater and sometimes lesser. when it demonstrates an understanding of what When humans are tested for transitive infer- the other may know. For example, when Susan ence, the use of language allows for the proposi- sees a hidden object moved to a second hidden tions to be completely relational. Relative size location after Billy has left the room and Susan may be assigned to individuals identifi ed only by understands that Billy will probably look for the name (e.g., given that Anne is taller than Betty, object in the fi rst location rather than second, we and Betty is taller than Carol, who is taller, Anne would say that Susan can take the perspective of or Carol?). With animals, however, there is no Billy or she has a theory of mind because she way to present such relational propositions with- understands that Billy doesn’t know that the out also presenting the actual stimuli. And if the object has been moved (see Frye 1993 ). To dem- stimuli differ in observable value (e.g., size), then onstrate perspective taking in an animal is a bit a correct response can be made without the need more complex because, in the absence of lan- to make an inference. guage, theory of mind must be inferred from McGonigle and Chalmers (1977 ) suggested other behavior (see, e.g., Hare et al. 2001 ). that a nonverbal relational form of the task could be represented by simple simultaneous discriminations in which one stimulus is associated with Self-recognition reinforcement (+) and the other is not (−). A > B can be represented as A + B−, B > C as B + C−, and Recognition of the similarity between ourselves so on. With four propositions, an animal would and other humans would seem to facilitate per- be exposed to A + B−, B + C−, C + D−, and D + E−. spective taking. If we can recognize ourselves in a A is always positive and E is always negative, but mirror, we can see that we are similar to others of B and D, stimuli that were never paired during our species. Gallup (1970 ) has shown that not training, would share similar reinforcement histo- only will chimpanzees exposed to a mirror use it ries. If animals order the stimuli from A is best to for grooming, but if their face is marked while E is worst, then B should be preferred over D. they are anesthetized, they will use the mirror to Findings consistent with transitive inference explore the mark visually and tactually (i.e., they have been reported in research with species as pass the mark test). Furthermore, both prior expe- diverse as chimpanzees (Gillan 1981 ), rats (Davis rience with the mirror and the presence of the mir- 1992), and pigeons (Fersen et al. 1991 ). Although ror following marking appear to be necessary for some have argued that these results can be mark exploration to occur. Mirror-directed mark accounted for without postulating that an infer- exploration appears to occur generally in the great ence has been made (Couvillon and Bitterman apes (orangutans and perhaps also in gorillas) but 1992; Fersen et al. 1991; Steirn et al. 1995 ), tran- not in monkeys even with extensive mirror experi- sitive inference effects have been found when ence (Gallup and Suarez 1991 ). However, using these presumably simpler mechanisms have been the mark test, there is some evidence of self-rec- controlled (Lazareva and Wasserman 2006 ; ognition in dolphins (Reiss and Marino 2001), Weaver et al. 1997 ). Thus, although it is not clear elephants (Plotnik et al. 2006 ), and magpies (Prior what mechanism produces it, pigeons clearly et al. 2008 ). Thus, self-recognition appears to show transitive choice that is not produced by occur in several nonprimate species thought to differential reinforcement history or differential show other kinds of cognitive skills. 20 T.R. Zentall

Imitation What Animals Can Tell Us About A more direct form of perspective taking Human Reasoning involves the capacity to imitate another (Piaget 1951 ), especially opaque imitation for which I have saved for last the discussion of several lines the observer cannot see itself perform the of research with animals directed to biases and response (e.g., clasping one’s hands behind heuristics characteristic of humans that appear to one’s back). But evidence for true imitative be somewhat irrational or at least suboptimal. The learning requires that one rule out (or control results of these studies are important, not so much for) other sources of facilitated learning follow- because of their implications for animals, but pri- ing observation (see Whiten and Ham 1992 ; marily for their implications for how we interpret Zentall 1996 , 2012 ). A design that appears to human behavior. That is, if other animals have control for artifactual sources of facilitated these same biases, then the basis for those biases learning following observation is the two-action does not depend on language or human culture as procedure based on a method developed by is sometimes proposed. Dawson and Foss (1965 ). For example, imitation is said to occur if observers, exposed to a demonstrator performing a response in one of Cognitive Dissonance two behaviorally different ways, perform the response with the same behavior as their dem- One of these biases has to do with a phenomenon onstrator. Akins and Zentall ( 1996 ) trained extensively studied in humans called cognitive Japanese quail demonstrators to either step on a dissonance. Cognitive dissonance is the discom- treadle or peck the treadle for food reinforce- fort that comes when there is a discrepancy ment. When observer quail were exposed to one between one’s beliefs and one’s behavior. For or the other demonstrator, they matched the example, if one believes that one should tell the behavior of their demonstrator with a high prob- truth, one is likely to feel dissonance on occa- ability (see also Zentall et al. 1996 , for similar sions when one fails to do so. That dissonance evidence with pigeons). Furthermore, there is may be resolved by deciding that there are some some evidence that pigeons can imitate a conditions under which lying is appropriate or sequence of two responses, operating a treadle the person lied to may have deserved it. Cognitive (by stepping or pecking) and pushing a screen dissonance presumably comes about because of a (to the left or to the right; Nguyen et al. 2005 ), need to be consistent or to avoid being labeled a an example of what Byrne and Russon (1998 ) hypocrite. Does this represent a kind of social refer to as program-level imitation. intelligence? And if so, would nonhuman ani- If Piaget is correct, the ability to imitate mals show a similar effect? But how would one requires the ability to take the perspective of go about asking this question of animals? another. But children do not develop the ability One approach involves a version of cognitive to take the perspective of another until they are dissonance called justiﬁ cation of effort (Aronson 5–7 years old, yet they are able to imitate oth- and Mills 1959 ). In their study, undergraduates, ers at a much earlier age. Furthermore, if who underwent an unpleasant initiation to pigeons and Japanese quail can imitate, it is become part of a group, reported that they unlikely that they do so by taking the perspec- wanted to join the group more than those who tive of the demonstrator, in the sense that underwent a less unpleasant initiation. It is Piaget implied. Thus, although cognitively assumed that those individuals who underwent interesting, imitation may not provide evidence an unpleasant initiation gave more value to for the kind of cognitive behavior implied by membership in the group to justify undergoing perspective taking. the unpleasantness. 2 Intelligence in Nonprimates 21

The justifi cation of effort design allows for a have recently found conditions under which direct test of cognitive dissonance in animals. For pigeons will prefer an average of 2 pellets of food example, if on some trials a pigeon has to work over a predictable 3 pellets of food (Zentall and hard to receive signal A that says food is coming Stagner 2011 ). The procedure is as follows: If the and on other trials the pigeon does not have to pigeon chooses the left alternative, on 20 % of work hard to receive signal B that says the same the trials, a green stimulus appears and is fol- food is coming, will the pigeon show a prefer- lowed by 10 pellets of food. The remainder of the ence for signal A over signal B? Several studies time it chooses the left alternative; a red stimulus have shown that they will (e.g., Clement et al. appears and is never followed by food. Thus, on 2000 ; Kacelnik and Marsh 2002 ). But is this cog- average the pigeon receives 2 pellets of food for nitive dissonance? Do animals need to justify to choosing that alternative. If the pigeon chooses themselves why they worked harder for one sig- the right alternative, it received either a blue or a nal than the other? yellow stimulus but in either can it receives 3 pel- Alternatively, we have suggested that this lets of food. Curiously, the pigeons prefer the left choice behavior results from the contrast between alternative overwhelmingly over the right alterna- the relatively negative emotional state of the tive, and they do so in spite of the fact that they organism at the end of the effort and upon presen- would get 50 % more food for choosing the right tation of the signal (Zentall and Singer 2007 ). alternative. That difference would be greater when more This result suggests that gambling behavior is effort was involved. Thus, the subjective value of likely to have a simple biological basis, and the signal for reinforcement might be judged to although social and cognitive factors may contrib- be greater. Contrast provides a more parsimoni- ute to human gambling behavior, the underlying ous account of the pigeons’ choice behavior. mechanism is likely to be simpler. The mecha- Could contrast also be involved in the similar nisms responsible for this suboptimal behavior behavior shown by humans? This possibility appear to involve the enhanced effect of the signal should be examined by social psychologists. for the large magnitude of reinforcer (the 10-pel- let “jackpot”) and the reduced effect of the signal for nonreinforcement with training (Stagner et al. Maladaptive Gambling Behavior 2012). This account appears to be consistent with research with humans that has found that gam- Humans often gamble (e.g., play the lottery) even blers overvalue wins in spite of their low proba- though the odds against winning are very high. bility of occurrence and they give too little This behavior may be attributable to an inaccu- negative value to their losses in spite of their high rate assessment of the probability of winning, probability of occurrence (Blanco et al. 2000 ). perhaps resulting in part from public announce- ments of the winners but not the losers (an avail- ability heuristic). Would animals show a similar Sunk Cost kind of maladaptive gambling behavior? According to optimal foraging theory, they The sunk cost effect occurs when one allows an should not because such inappropriate behavior amount of money, time, or other resource already should have been selected against by evolution. invested to affect one’s decision to invest more Furthermore, if the choice is to have any meaning resources. For example, one may sit through a for the animal, it would have to have experienced fi lm that one does not like because to leave would the probability associated with winning (rein- be to waste one’s investment of the price of the forcement) and that should reduce the likelihood ticket. But in so doing, one is spending additional that the animal would not be able to assess the resources, one’s time, and there is no way to probability of winning and losing. However, we recoup the money already invested. Similarly, 22 T.R. Zentall one may choose to continue with a failing busi- to judge the value of sets of dishes and sets of ness because of the past investment one has made baseball cards (Hsee 1998), and it also has been in it. This phenomenon comes under the general found when academics are asked to judge the rubric of prospect theory (Kahneman and quality of a curriculum vita (Hayes 1983 ). For Twersky 1979 ) which suggests that humans will example, a vita with three publications in excel- take greater risks to avoid a loss than to obtain a lent journals is judged better than one with the gain. The question is to what extent does this same three publications in excellent journals plus behavior stem from the cultural tenet to avoid six more in lesser quality journals. wasting resources and to complete what one Recent evidence suggests that even pigeons starts. If one could show that animals show the are susceptible to this bias. We found that pigeons same behavior, it would suggest that sunk cost is will work for dried peas and dried milo seeds, but a general phenomenon that has basic behavioral when given a choice between the two, they prefer origins. the peas. However, when they are given a choice In fact, evidence for sunk cost has been found between a pea and a pea together with a milo in pigeons (Navarro and Fantino 2005 ; Pattison seed, they prefer the pea alone (Zentall et al. et al. 2012 ). For example, pigeons learn that 2013). Apparently, the pigeons too are averaging pecking a green light requires 30 pecks, whereas the high-quality pea with the lower-quality milo pecking a red light requires only 10 pecks. They seed and value the pair less than the pea by itself then learn that after pecking the green light a (see also Kralik et al. 2012 ). The basis of this bias number of times (that varied from trial to trial), may originate in the need to make rapid deci- they would be able to choose to continue with the sions, presumably because of intense competi- green light (to complete the 30 pecks) or switch tion from conspecifi cs and the possibility of to the red light for which 10 pecks were required. predation, and they use it in the laboratory even Results indicated that the pigeons often choose to when speed is not a factor. Once again, the fact return to the green light even when 20 more pecks that other animals show this suboptimal choice are required. Thus, pigeons show a sunk cost indicates that the bias is probably not dependent effect that is very similar to that shown by on human cultural infl uence. humans. Why pigeons show the sunk cost effect is not clear. One can speculate that it arises from the fact that in nature switching to a different Conclusions patch often involves uncertainty, some travel time, and possible danger, but one can certainly The broad range of positive research fi ndings that conclude that culture and language are not neces- have come from investigating the cognitive abili- sary components. ties of animals suggests that many of the “special capacities” attributed to humans may be more quantitative than qualitative. In the case of many When Less Is More cognitive learning tasks, once we learn how to ask the question appropriately (i.e., in a way that When humans are asked to judge the value of a is accommodating to the animal), we may often set of objects of excellent quality, they often give be surprised with the capacity of animals to use it higher value than those same objects with the complex relations. addition of some objects of lesser quality (Hsee In evaluating the animal (and human) intelli- 1998). This bias is an example of the affect heu- gence literature, we should be sensitive to both ristic in which it appears that the average quality overestimation of capacity (what appears to be of a set is used to determine the value of the set higher-level functioning in animals that can be rather than the quantity of items in the set. The accounted for more parsimoniously at a lower phenomenon has become known as a less is more level; see Zentall 1993 ) and underestimation of effect. It has been found when humans are asked capacity (our bias to present animals with tasks 2 Intelligence in Nonprimates 23 convenient to our human sensory, response, and Capaldi, E. J., & Miller, D. J. (1988). Counting in rats: Its motivational systems). Underestimation can also functional signifi cance and the independent cognitive processes that constitute it. Journal of Experimental come from the diffi culty in providing animals Psychology: Animal Behavior Processes, 14 , 3–17. with task instructions as one can quite easily do Chapuis, N., & Varlet, C. (1987). Short cuts by dogs in with humans (see Zentall 1997 ). The accurate natural surroundings. Quarterly Journal of assessment of animal intelligence will require Experimental Psychology, 39 , 49–64. Clayton, N. S., & Dickinson, A. (1999). Scrub jays vigilance, on the one hand, to evaluate cognitive ( Aphelocoma coerulescens ) remember the relative functioning against simpler accounts and, on the time of caching as well as the location and content of other hand, to determine the conditions that will their caches. Journal of Comparative Psychology, 113 , maximally elicit the animal’s cognitive capacity. 403–416. Clement, T. S., Feltus, J., Kaiser, D. H., & Zentall, T. R. The study of human biases by examining ani- (2000). “Work ethic” in pigeons: Reward value is mals for the presence of similar phenomena in directly related to the effort or time required to obtain animals can also help us to determine that sim- the reward. Psychonomic Bulletin & Review, 7 , pler mechanisms are involved. Thus, the study of 100–106. Collette, T. S., & Graham, P. (2004). Animal navigation: animal intelligence can inform us not only of the Path integration, visual landmarks and cognitive maps. cognitive abilities of animals but also can suggest Current Biology, 14 , 475–477. the bases of certain human phenomena thought to Couvillon, P. A., & Bitterman, M. E. (1992). A conven- have complex social origins. tional conditioning analysis of “transitive inference” in pigeons. Journal of Experimental Psychology: Animal Behavior Processes, 18 , 308–310. Acknowledgments Preparation of this chapter was made Davis, H. (1992). Transitive inference in rats (Rattus nor- possible by grant MH 63726 from the National Institute of vegicus ). Journal of Comparative Psychology, 106 , Mental Health and grant HD60996 from the National 342–349. Institute of Child Health and Human Development. Davis, H., & Memmott, J. (1982). Counting behavior in animals: A critical evaluation. Psychological Bulletin, 92 , 547–571. Dawkins, R. (1976). The selfi sh gene. New York: Oxford References University Press. Dawson, B. V., & Foss, B. M. (1965). Observational learn- Akins, C., & Zentall, T. R. (1996). Evidence for true imi- ing in budgerigars. Animal Behaviour, 13 , 470–474. tative learning in Japanese quail. Journal of Dorrance, B. R., Kaiser, D. H., & Zentall, T. R. (2000). Comparative Psychology, 110 , 316–320. Event duration discrimination by pigeons: The choose- Aronson, E., & Mills, J. (1959). The effect of severity of short effect may result from retention-test novelty. initiation on liking for a group. Journal of Abnormal Animal Learning & Behavior, 28 , 344–353. and Social Psychology, 59 , 177–181. Edwards, C. A., Jagielo, J. A., Zentall, T. R., & Hogan, Babb, S. J., & Crystal, J. D. (2006). Discrimination of D. E. (1982). Acquired equivalence and distinctive- what, when, and where is not based on time of day. ness in matching-to-sample by pigeons: Mediation by Learning & Behavior, 34 , 124–130. reinforcer-specifi c expectancies. Journal of Bitterman, M. E. (1975). The comparative analysis of Experimental Psychology: Animal Behavior learning. Science, 188 , 699–709. Processes, 8 , 244–259. Bitterman, M. E., & Mackintosh, N. J. (1969). Habit Fersen, L. V., Wynne, C. D. L., Delius, J. D., & Staddon, reversal and probability learning: Rats, birds, and fi sh. J. E. R. (1991). Transitive inference formation in In R. M. Gilbert & N. S. Sutherland (Eds.), Animal pigeons. Journal of Experimental Psychology: Animal discrimination learning (pp. 163–185). New York: Behavior Processes, 17 , 334–341. Academic. Frye, D. (1993). Causes and precursors of children’s the- Blanco, C., Ibáñez, Sáiz-Ruiz, A., Blanco-Jerez, J., & ory of mind. In D. F. Hay & A. Angold (Eds.), Nunes, E. V. (2000). Epistemology, pathophysiology, Precursors and causes of development and psychopa- and treatment of pathological gambling. CNS Drugs, thology . Chichester: Wiley. 13 , 397–407. Galef, B. G., Jr. (1988). Imitation in animals: History, Bryne, R. W., & Russon, A. E. (1998). Learning by imita- defi nition, and interpretation of data from the psycho- tion: A hierarchical approach. Behavioral and Brain logical laboratory. In T. R. Zentall & B. G. Galef Jr. Sciences, 21 , 667–721. (Eds.), Social learning: Psychological and biological Capaldi, E. J. (1993). Animal number abilities: perspectives (pp. 3–28). Hillsdale: Erlbaum. Implications for a hierarchical approach to instrumen- Gallup, G. G. (1970). Chimpanzees self-recognition. tal learning. In S. T. Boysen & E. J. Capaldi (Eds.), Science, 167 , 86–87. The development of numerical competence (pp. 191– Gallup, G. G., & Suarez, S. D. (1991). Social responding 209). Hillsdale: Erlbaum. to mirrors in rhesus monkeys: Effects of temporary 24 T.R. Zentall

mirror removal. Journal of Comparative Psychology, Pigeon homing along highways and exits. Current 105 , 376–379. Biology, 14 , 1239–1249. Garcia, J., & Koelling, R. A. (1966). Relation of cue to Mackintosh, N. J. (1969). Comparative studies of reversal consequence in avoidance learning. Psychonomic and probability learning: Rats, birds, and ﬁ sh. In R. M. Science, 4 , 123–124. Gilbert & N. S. Sutherland (Eds.), Animal discrimina- Gillan, D. J. (1981). Reasoning in the chimpanzee: tion learning (pp. 137–162). New York: Academic. II. Transitive inference. Journal of Experimental McGonigle, B. O., & Chalmers, M. (1977). Are monkeys Psychology: Animal Behavior Processes, 7 , 150–164. logical? Nature, 267 , 694–696. Grant, D. S. (1981). Stimulus control of information pro- Meyer, D. R. (1971). Habits and concepts of monkeys. In cessing in pigeon short-term memory. Learning and L. E. Jarrard (Ed.), Cognitive processes of nonhuman Motivation, 12 , 19–39. primates (pp. 83–102). New York: Academic. Hare, B., Call, J., & Tomasello, M. (2001). Do chimpan- Miller, H. C., Friedrich, A. M., Narkavic, R. J., & Zentall, zees know what conspeciﬁ cs know? Animal Behaviour, T. R. (2009). A differential outcomes effect using 61 , 139–151. hedonically-nondifferential outcomes with delayed Harlow, H. F. (1949). The formation of learning sets. matching-to-sample by pigeons. Learning & Behavior, Psychological Review, 56 , 51–65. 37 , 161–166. Hayes, S. C. (1983). When more in less: Quantity versus Morgan, C. L. (1894). An introduction to comparative quality of publications in the evaluation of vitae. The psychology . London: Scott. American Psychologist, 38 , 1398–1400. Navarro, A. D., & Fantino, E. (2005). The sunk cost effect Herrnstein, R. J., & deVilliers, P. A. (1980). Fish as a in pigeons and humans. Journal of the Experimental natural category for people and pigeons. Psychology Analysis of Behavior, 83 , 1–13. of Learning and Motivation, 14 , 59–95. Nguyen, N. H., Klein, E. D., & Zentall, T. R. (2005). Herrnstein, R. J., & Loveland, D. H. (1964). Complex Imitation of two-action sequences by pigeons. visual concept in the pigeon. Science, 146 , 549–551. Psychonomic Bulletin & Review, 12 , 514–518. Herrnstein, R. J., Loveland, D. H., & Cable, C. (1976). Pattison, K. F., Zentall, T. R., & Watanabe, S. (2012). Natural concepts in pigeons. Journal of Experimental Sunk cost: Pigeons ( Columba livia ) too show bias to Psychology: Animal Behavior Processes, 2 , 285–301. complete a task rather than shift to another. Journal of Honig, W. K., & Thompson, R. K. R. (1982). Retrospective Comparative Psychology, 126 , 1–9. and prospective processing in animal working mem- Pepperberg, I. M. (1987). Interspecies communication: A ory. In G. Bower (Ed.), The psychology of learning tool for assessing conceptual abilities in an African and motivation (Vol. 16, pp. 239–283). Orlando: Grey parrot. In G. Greenberg & E. Tobach (Eds.), Academic. Language, cognition, and consciousness: Integrative Hsee, C. K. (1998). Less is better: When low-value levels (pp. 31–56). Hillsdale: Erlbaum. options are valued more highly than high-value Peterson, G. B. (1984). How expectancies guide behav- options. Journal of Behavioral Decision Making, 11 , ior. In H. L. Roitblat, T. G. Bever, & H. S. Terrace 107–121. (Eds.), Animal cognition (pp. 135–148). Hillsdale: Hull, C. L. (1943). Principles of behavior. New York: Erlbaum. Appleton. Peterson, G. B., Wheeler, R. L., & Trapold, M. A. (1980). Kacelnik, A., & Marsh, B. (2002). Cost can increase pref- Enhancement of pigeons’ conditional discrimination erence in starlings. Animal Behaviour, 63 , 245–250. performance by expectancies of reinforcement and Kahneman, D., & Tversky, A. (1979). Prospect theory: An nonreinforcement. Animal Learning & Behavior, 8 , analysis of decision under risk. Econometrica, 47 , 22–30. 263–291. Piaget, J. (1951). Play, dreams, and imitation in child- Kelly, R., & Grant, D. S. (2001). A differential outcomes hood . New York: W. W. Norton. effect using biologically neutral outcomes in delayed Plotnik, J. M., de Waal, F. B. M., & Reiss, D. (2006). Self- matching-to-sample with pigeons. Quarterly Journal recognition in an Asian elephant. Proceedings of the of Experimental Psychology, 54B , 69–79. National Academy of Sciences, 103 , 17053–17057. Kralik, J. D., Xu, E. R., Knight, E. J., Khan, S. A., & Prior, H., Schwarz, A., & Gunturkun, O. (2008). Mirror- Levine, J. W. (2012). When less is more: Evolutionary induced behavior in the magpie ( Pica pica ): Evidence origins of the affect heuristic. PLoS One, 7 (10), of self-recognition. PLoS Biology, 6 (8), e202. e46240. doi: 10.1371/journal.pone.0046240 . doi: 10.1371/journal.pbio.0060202 . Kuan, L. A., & Colwill, R. (1997). Demonstration of a Raby, C. R., Alexis, D. M., Dickinson, A., & Clayton, socially transmitted taste aversion in the rat. N. S. (2007). Empirical evaluation of mental time Psychonomic Bulletin & Review, 4 , 374–377. travel. Behavioral Brain Sciences, 30 , 330–331. Lazareva, O. F., & Wasserman, E. A. (2006). Effect of Reiss, D., & Marino, L. (2001). Self-recognition in the stimulus orderability and reinforcement history on bottlenose dolphin: A case of cognitive convergence. transitive responding in pigeons. Behavioural Proceedings of the National Academy of Sciences, 98 , Processes, 72 , 161–172. 5937–5942. Lipp, H.-P., Vyssotski, A. L., Wolfer, D. P., Renaudineau, Riley, D. A. (1968). Discrimination learning . Boston: S., Savini, M., Tröster, G., & Dell’Omo, G. (2004). Allyn & Bacon. 2 Intelligence in Nonprimates 25

Roberts, W. A. (2002). Are animals stuck in time? Wasserman, E. A., DeVolder, C. L., & Coppage, D. J. Psychological Bulletin, 128 , 473–489. (1992). Non-similarity based conceptualization in Roberts, W. A., & Grant, D. S. (1976). Studies of short- pigeons via secondary or mediated generalization. term memory in the pigeon using the delayed Psychological Science, 6 , 374–379. matching-to-sample procedure. In D. L. Medin, W. A. Weaver, J. E., Steirn, J. N., & Zentall, T. R. (1997). Roberts, & R. T. Davis (Eds.), Processes of animal Transitive inference in pigeons: Control for differen- memory (pp. 79–112). Hillsdale: Erlbaum. tial value transfer. Psychonomic Bulletin and Review, Roper, K. L., Kaiser, D. H., & Zentall, T. R. (1995). 4 , 113–117. Directed forgetting in pigeons: The role of alternative Whiten, A., & Ham, R. (1992). On the nature and evolu- memories in the effectiveness of forget cues. Animal tion of imitation in the animal kingdom: Reappraisal Learning & Behavior, 23 , 280–285. of a century of research. Advances in the Study of Sherburne, L. M., Zentall, T. R., & Kaiser, D. H. (1998). Behavior, 21 , 239–283. Timing in pigeons: The choose-short effect may Williams, D. A., Butler, M. M., & Overmier, J. B. (1990). result from “confusion” between delay and intertrial Expectancies of reinforcer location and quality as cues intervals. Psychonomic Bulletin & Review, 5 , for a conditional discrimination in pigeons. Journal of 516–522. Experimental Psychology: Animal Behavior Singer, R. A., Abroms, B. D., & Zentall, T. R. (2007). Processes, 16 , 3–13. Formation of a simple cognitive map by rats. Zentall, T. R. (1993). Animal cognition: An approach to International Journal of Comparative Psychology, 19, the study of animal behavior. In T. R. Zentall (Ed.), 417–425. Animal cognition: A tribute to Donald A. Riley Singer, R. A., & Zentall, T. R. (2007). Pigeons learn to (pp. 3–15). Hillsdale: Erlbaum. answer the question ‘Where did you just peck?’ and Zentall, T. R. (1996). An analysis of imitative learning in can report peck location when unexpectedly asked. animals. In C. M. Heyes & B. G. Galef Jr. (Eds.), Learning & Behavior, 35 , 184–189. Social learning and tradition in animals (pp. 221– Slotnick, B. M., & Katz, H. M. (1974). Olfactory learning- 243). New York: Academic. set formation in rats. Science, 185 , 796–798. Zentall, T. R. (1997). Animal memory: The role of instruc- Spence, K. W. (1937). The differential response in ani- tions. Learning and Motivation, 28 , 248–267. mals to stimuli varying within a single dimension. Zentall, T. R. (1998). Symbolic representation in pigeons: Psychological Review, 44 , 430–444. Emergent stimulus relations in conditional discrimina- Stagner, J. P., Laude, J. R., & Zentall, T. R. (2012). tion learning. Animal Learning & Behavior, 26 , Pigeons prefer discriminative stimuli independently of 363–377. the overall probability of reinforcement and of the Zentall, T. R. (2012). Perspectives on social learning. number of presentations of the conditioned reinforcer. Journal of Comparative Psychology, 126 , 114–128. Journal of Experimental Psychology: Animal Behavior doi: 10.1037/a0025381 . Processes, 38 , 446–452. Zentall, T. R., & Singer, R. A. (2007). Within-trial contrast: Steirn, J. N., Weaver, J. E., & Zentall, T. R. (1995). Pigeons prefer conditioned reinforcers that follow a Transitive inference in pigeons: Simpliﬁ ed procedures relatively more rather than less aversive event. Journal and a test of value transfer theory. Animal Learning & of the Experimental Analysis of Behavior, 88 , 131–149. Behavior, 23 , 76–82. Zentall, T. R., & Smeets, P. M. (Eds.). (1996). Stimulus Suddendorf, T., & Corballis, M. C. (1997). Mental time class formation in humans and animals . Amsterdam: travel and the evolution of the human mind. Genetic, North Holland. Social, and General Psychology Monographs, 123 , Zentall, T. R., & Stagner, J. P. (2011). Maladaptive choice 133–167. behavior by pigeons: An animal analog of gambling Tolman, E. C. (1932). Purposive behavior in animals and (sub-optimal human decision making behavior). men . New York: Century. Proceedings of the Royal Society B: Biological Trapold, M. A. (1970). Are expectancies based on differ- Sciences, 278 , 1203–1208. ent reinforcing events discriminably different? Zentall, T. R., Sutton, J. E., & Sherburne, L. M. (1996). Learning and Motivation, 1 , 129–140. True imitative learning in pigeons. Psychological Tulving, E. (1972). Episodic and semantic memory. In Science, 7 , 343–346. E. Tulving & W. Donaldson (Eds.), Organization of Zentall, T. R., Clement, T. S., Bhatt, R. S., & Allen, memory (pp. 382–403). New York: Academic. J. (2001). Episodic-like memory in pigeons. Urcuioli, P. J., Zentall, T. R., Jackson-Smith, P., & Steirn, Psychonomic Bulletin & Review, 8 , 685–690. J. N. (1989). Evidence for common coding in many- Zentall, T. R., Laude, J. R., Daniels, C. W., & Case, J. P. to- one matching: Retention, intertrial interference, (2013, August). When less means more. Paper pre- and transfer. Journal of Experimental Psychology: sented at the meeting of the American Psychological Animal Behavior Processes, 15 , 264–273. Association, Honolulu. Intelligence in Nonhuman Primates 3

Tara Mandalaywala , Christine Fleener , and Dario Maestripieri

With such variability in animal intelligence, we Introduction must ask why and how intelligence evolved and why species differences in intelligence exist. Intelligence is a multidimensional construct that Intelligence is a metabolically costly brain func- can include a wide range of cognitive abilities. tion, as it is associated with growing larger brains Historically, intelligence in humans and other over longer developmental periods. Most life forms animals has been measured and defi ned by fl exi- on Earth show highly sophisticated adaptations to bility in problem solving, learning, memory, rea- their environment but little or no general intelli- soning, abstract thinking, planning, and gence, suggesting that such general intelligence is communication and language. Humans are highly not a prerequisite for survival or successful repro- profi cient in all of these domains, but among non- duction on our planet. However, intelligence abili- human animals there is more variability, and ties have emerged in many different evolutionary some species perform well on one or a few mea- lineages, indicating that the benefi ts of intelligence sures of intelligence but poorly on others. can outweigh its costs under certain selective pressures. In the primate lineage in particular, general intelligence has evolved to a greater degree than in T. Mandalaywala , Ph.D.(*) almost any other taxonomic group, suggesting that Department of Psychology , New York University , 6 any effort to understand the evolution of human Washington Place , New York , NY 10003 , USA intelligence must take into account broad evolu- Department of Comparative Human Development, tionary trends within the Primate order as well as Institute for Mind and Biology , University of similarities and differences between humans and Chicago , 940 E. 57th Street , Chicago , IL 60637 , USA nonhuman primates (NHPs). e-mail: [email protected] The order Primates is 60–85 million years old C. Fleener , M.A. and is split into two main groups: prosimians and Department of Comparative Human Development, Institute for Mind and Biology , University of anthropoids. Prosimians, which include lemurs, Chicago, 940 E. 57th Street , Chicago , IL 60637 , USA lorises, and tarsiers, are the most ancestral pri- e-mail: cfl [email protected] mates. Anthropoids include monkeys, apes, and D. Maestripieri , Ph.D. hominins and are split into platyrrhines, fl at- Comparative Human Development, nosed New World monkeys of South and Central Evolutionary Biology, and Neurobiology , America, and catarrhines, narrow-nosed Old The University of Chicago , 5730 S. Woodlawn Avenue , Chicago , IL 60637 , USA World monkeys and apes of Africa and southeast e-mail: [email protected] Asia. Many prosimians and New World monkeys

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 27 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_3, © Springer Science+Business Media New York 2015 28 T. Mandalaywala et al. communicate mainly through olfaction or simple vocalizations, while Old World monkeys and The Evolution of Intelligence apes rely more heavily on complex visual, vocal, in Primates and tactile signals than on olfaction. These forms of communication in Old World monkeys and The two predominant theories advanced to apes demand more behavioral fl exibility and as explain the evolution of intelligence in primates such involve more complex cognitive abilities. emphasize the cognitive challenges imposed by Using comparative studies, we can see how intel- (1) feeding ecology or (2) the social environment. ligence has evolved in the order Primates, with These are known as the ecological intelligence prosimians showing the least cognitive complex- and social intelligence hypotheses, respectively. ity, New World monkeys showing greater com- Although recent research more strongly supports plexity, Old World monkeys showing even greater the social intelligence hypothesis, it is possible complexity than New World monkeys, and a sig- that both theories, and possibly other not yet nifi cant jump in the cognitive abilities of apes and developed theories, are partially correct (e.g., humans (Deaner et al. 2006 ). Reader et al. 2011 ; Holekamp 2007 ). Both the Despite their cognitive differences, all pri- ecological and social intelligence hypotheses mates are characterized by having large brains share the assumption that brain tissue is one of and, in particular, a large neocortex, which is the most metabolically expensive tissues both to involved in sensory perception, motor commands, grow and to maintain (e.g., Aiello and Wheeler spatial reasoning, and conscious thought. Primates 1995). In human infants, brain tissue accounts for develop more slowly and live longer than most approximately 60 % of the basal metabolic rate, mammals, which facilitates a long process of and even in adulthood brain tissue requires a brain growth and maintenance of brain plasticity. large proportion of basal metabolism, around Large, plastic brains allow for advanced, fl exible 20 % (Martin 2013 ). cognitive abilities. Using their intelligence, primates can keep track of elaborate dominance hierarchies and recognize conspecifi cs and kin. They Ecological Intelligence Hypothesis can manipulate, deceive, or cooperate. Some can solve problems in novel ways, make tools, and The ecological intelligence hypothesis (Milton even be taught to use symbols and language. 1981 ) proposes that foraging challenges posed by To explore how large brains and intelligence the environment were the impetus for complex arose in primates, it is useful to consider the cognitive evolution in primates. Most primates broad range of ecological and social pressures consume a predominantly plant-based diet, with primates experience in their environment. Some different species of primates consuming different primate species are mostly arboreal and their diet types of plants (e.g., folivorous primates con- consists mainly of leaves, fruit, or insects; others sume mostly leaves, whereas frugivorous pri- are mostly terrestrial and their diet can include mates consume a mixture of leaves and fruits). grass, leaves, fruit, or meat. Some primates live Different plant species have different distribu- in social groups made up of hundreds of individ- tions in space and time, and each type of plant uals, while others live more solitary lifestyles. As imposes different foraging demands on the pri- we will see in the next section, some theories mate species that consumes them. For example, concerning the evolution of primate intelligence leaves tend to be widely distributed but are of low emphasize the importance of ecological pres- nutritional quality; thus primates must consume sures, while others focus on the role of complex more leaves to gain the same energy intake social environments. obtained from consuming fewer plants of higher 3 Intelligence in Nonhuman Primates 29 quality, such as fruit (see Fish and Lookwood of higher rates of tool use and object manipula- 2003 for evidence of dietary constraints on brain tion of food items in species that displayed more size). Fruits tend to be patchily distributed in both cognitively complex behaviors, such as chimpan- space and time, imposing a unique challenge on zees and capuchins (Parker and Gibson 1977 ; the frugivorous species that depend on them. This reviewed in King 1986 ). However, both the eco- environmental complexity requires that frugivo- logical intelligence and the extractive foraging rous species possess a suite of strong cognitive hypotheses are problematic as they fail to explain skills to forage effi ciently, and the ecological cognitive behavior in a number of species (see intelligence hypothesis proposes that species fac- King 1986 for a review of species that are consid- ing the patchiest and most challenging foraging ered to have high intelligence but do not engage environments should be the most cognitively in extractive foraging). advanced (Milton 1981 ). Consistent with this hypothesis, frugivores have a larger cranial capacity than folivores (Clutton-Brock and Social Intelligence Hypothesis Harvey 1980 ). In an empirical test of the ecological intelli- The social intelligence hypothesis postulates that gence hypothesis, Platt et al. ( 1996 ) compared increased group size and the demands imposed Wied’s marmosets and golden lion tamarins on by increased social complexity selected for big- spatial and visual memory tasks. Although the ger brains and enhanced cognition (Dunbar 1998 ; diets of these two species overlap, signifi cant dif- Humphrey 1976). Although group living confers ferences in spatial and temporal qualities of their many advantages (e.g., Silk 2007 ), it also has foraging led Platt et al. (1996 ) to hypothesize that drawbacks such as increased competition for the tamarins would outperform the marmosets on resources, which includes both mates and food. the cognitive tasks. In accordance with the eco- Both Humphrey ( 1976) and Dunbar ( 1998 ) pro- logical intelligence hypothesis, Platt et al. (1996 ) posed that primates solve these social challenges found that, after a long delay (24 and 48 h), the with the use of fl exible cognitive strategies in real tamarins outperformed the marmosets, perhaps time, rather than evolved “rules-of-thumb” and because of the increased memory demands on the heuristics. tamarins when seeking out food sources that had The social intelligence hypothesis predicts more complex spatial and temporal distribution that all social species should show enlarged brain than the food sources of the marmosets. size relative to more solitary species because of Parker and Gibson (1979 ) proposed that it was the increased demands of navigating a more com- not the complexity of the food environment but plex social environment and keeping track of an rather the complexity of the foraging tasks posed exponentially greater number of relationships. In by the foods consumed that selected for increased line with this prediction, Dunbar (1998 ) found intelligence in primates. Those species that that neocortex size correlates with social group engage in “extractive foraging,” in which a food size but does not correlate with any measure pro- object is not immediately or obviously percepti- posed by the ecological intelligence or extractive ble and must be extracted from a nonfood item, foraging hypotheses. The neocortex has increased would possess enhanced intelligence relative to in size exponentially across primate evolution those species that only consume readily percep- (Passingham 1982 ), and as brain tissue is pro- tible foods. Parker and Gibson (1979 ) suggest hibitively expensive to grow and maintain, the that these complex foraging tasks present two selection pressures for a dramatic increase in cognitively taxing problems: the ability to under- neocortex size must have outweighed the meta- stand that a nonfood item might contain an edible bolic cost (Dunbar 1998 ) suggesting the particu- substance and the ability to engage in complex lar importance of the neocortex in primates. object manipulation. Evidence in favor of the Moreover, neocortex size correlates with a vari- extractive foraging hypothesis came in the form ety of measures of social complexity in primates 30 T. Mandalaywala et al. such as the number of females in a group, the size ferent ways in which nonhuman primates display of grooming groups, as well as the amount of tac- their intelligence across physical and social tical deception and social learning (reviewed in domains, in the fi eld and in the laboratory. Dunbar and Shultz 2007 ; Reader and Laland 2002 ; but see Grueter et al. 2013 ). For example, research in various species of lemurs has found Physical Cognition that species that live in larger groups are better at tasks of transitive inference than species that live Primates navigate many different types of envi- in smaller groups (MacLean et al. 2008 ). ronment, and as such, have evolved fl exibility in A related theory, the Machiavellian intelli- how they process and interact with their physical gence hypothesis, suggests that it was the need to world. Through both natural selection and expe- engage in social manipulation that helped drive rience, primates have acquired an understanding increased cognitive ability (reviewed in Whiten of how objects should behave in space and time and Byrne 1997). Evidence for this hypothesis and are sometimes able to apply simple princi- comes from studies of deception. Tactical decep- ples of physics and mathematics to solving spe- tion “occurs when an individual is able to use an cifi c problems. Physical cognition refers to the ‘honest’ act from his normal repertoire in a dif- expectation that objects in an environment will ferent context to mislead familiar individuals” follow consistent laws in their behavior and inter- (Whiten and Byrne 1988 ), and it is not found in action. For example, without knowing anything all primate species; it is distributed unevenly with about an object, we expect that all parts of that greater rates of deception occurring in the great object can only trace one connected path through apes, and virtually no instances of deception space and time, that two distinct objects cannot found in prosimians thus far (Byrne and Whiten occupy the same space at the same time, that the 1992 ). Moreover, there is a relationship between object will not deform as it moves, and that two neocortical volume and frequency of deception in distinct objects will move together if and only if primates (Byrne and Corp 2004 ). Implicit in the they make contact in some way. These physical defi nition of deception is the application of a concepts are termed continuity, boundedness, behavior to novel situations, emphasizing the rigidity, and no action at a distance, respectively utility of fl exibility in thinking (e.g., applying a (Spelke 1990 ), and there is evidence to show that behavior or response to a novel situation when NHPs comprehend these concepts as well (see appropriate), not just increased memory or visuo- below). spatial skills. Flexible thinking is often seen as a Both apes and monkeys have shown the abil- hallmark of true intelligence. Unfortunately, fl ex- ity to represent the existence and movements of ible thinking is not well studied. Some hypothe- objects even when they cannot be seen (capu- size that one should be able to distinguish chins, chimpanzees: Mireille et al. 1976 ; Rhesus hard-wired intelligence—that is, intelligence macaques: de Bois and Novak 1994 ; orangutans, arising as a result of specifi c challenges in the squirrel monkeys: de Bois et al. 1998 ), but this environment—from higher-order intelligence research reveals qualitative differences in how and fl exible thinking, in which a subject can apes and monkeys understand physical concepts, solve a completely novel problem (see also cul- illuminating how physical cognition has evolved tural intelligence: van Schaik et al. 2012 ). in primates. In visible displacement tasks where Just as there are problems with the ecological an object moves behind an occluder, both apes intelligence hypothesis, some question the neces- and monkeys are able to track the object when it sity of enhanced cognition to deal with social cannot be seen, but in invisible displacement complexity (e.g., Barrett et al. 2007 ). Having tasks, where the object is fi rst placed in a con- reviewed the main theories as to how and why tainer and then moved behind an occluder, only enhanced intelligence may have evolved in the apes can track the object. Other research shows order Primates, we will now review the many dif- that NHPs can understand that a solid object can- 3 Intelligence in Nonhuman Primates 31 not move through another solid object (rhesus solve problems and construct limitless structures macaques: Santos and Hauser 2002 ), but this and apparatuses, and there are many reports research again reveals qualitative differences in showing that NHPs are also capable of making how the concept of solidity is understood. For and using tools to hunt, forage, or solve simple example, while monkeys are surprised when an tasks. Chimpanzees fi sh for termites by carefully object appears to fall through a solid barrier, they selecting twigs and stripping off their leaves so are not surprised if the object appears to roll they can fi t through small termite holes (Boesch down a ramp and through a solid barrier. This and Boesch 1990). Individual chimpanzees also error may be due to a “gravity bias” that is also modify their tools for more effective use, fraying present in infant humans, where most attention is the tips of their twigs to provide more surface given to an item that is dropped with the expecta- area for termites to bite. Great apes also show the tion that it will fall on the surface directly below capacity to use stone tools to forcefully access where it was dropped (Hood et al. 1999 ). When otherwise inaccessible foods (chimpanzees: multiple forces act on a falling object (i.e., ramps, Boesch and Boesch 1990 ; orangutans: van Schaik tubes), both adult NHPs and infant humans are et al. 1996; gorillas: Breuer et al. 2005), and this unable to track the object’s movement in space ability is present in some species of monkey as and time (Santos 2004). This comparison of well. Some macaques have been known to use infant humans and nonhuman primates suggests stone axes and pounding hammers to process similar foundational cognitive circuitry. shelled foods (Gumert et al. 2009 ), and capuchins Other research shows that NHPs are capable use carefully selected stone anvils and hammers of recognizing that objects cannot move without to break open nuts (Ottoni and Mannu 2001 ; fi rst being contacted by another object (cotton- Fragaszy et al. 2004 ). top tamarins: Hauser 1998), can differentiate Some NHPs are not only capable of choosing between adequate and inadequate physical sup- and modifying tools to accomplish a task, they are port (chimpanzees: Cacchione and Horst 2004 ), also capable of using tools to make other tools. and can infer the location of rewards based on the Bearded capuchins have been reported selecting effect of weight (chimpanzees: Hanus and Call small stones to loosen large quartz fragments in 2008 ). Having a foundation of these simple phys- conglomerate rock so that the quartz can be used ical concepts allows primates to more quickly as heavier, more effective hammers in nut crack- negotiate their environment and inform their ing (Mannu and Ottoni 2009 ). This kind of problem solving. While there is extensive evi- sequential tool use begs the question of the depth dence showing that NHPs comprehend many of knowledge NHPs use when they are utilizing physical properties, the question is how they are tools. We have established that NHPs differentiate capable of applying this information to the way between different physical properties of objects, they reason about their physical world. but can they use these properties to choose their tools, and does this show that they understand the causal relationship between the physical property Tool Use and Causal Understanding and the effectiveness of the tool? When chimpanzees and human children were The primary morphological feature that distin- shown how to get into an opaque puzzle box to guishes primates from most other animals is their receive a reward, both the chimpanzees and the hands. Primates evolved manual dexterity for children could repeat all the steps they had arboreal living and for obtaining and processing observed in order to receive the treat. However, food (Passingham 1982 ), but along with their when a translucent box was introduced, it large brains, this adaptation allows primates to revealed that most of the steps that had been interact with and manipulate their environment in observed served no functional purpose in retriev- unique ways. With the use of their hands, humans ing the treat. When chimpanzees and children have become profi cient at manufacturing tools to were again tested on this translucent box, the 32 T. Mandalaywala et al. humans continued to repeat all of the unneces- als to specifi c quantities and do simple arithmetic sary steps, but the chimpanzees skipped right to (e.g., chimpanzees: Boysen and Berntson 1989 ; the steps required to access the reward (Horner squirrel monkeys: Olthof et al. 1997 ). and Whiten 2005 ). This result is strong evidence for an NHP’s understanding of causality. The human child’s inability to complete this task in Numerical Cognition fewer steps may not necessarily indicate an inability to reason about causality but rather that Usually, instead of counting, animals and human they may default to overimitating more knowl- infants mentally represent quantities approxi- edgeable sources at these early ages. mately in an analog format, and many studies Other recent experiments show that NHPs are show that both children and NHPs share cognitive indeed attentive to the physical properties of the algorithms for encoding and comparing numeri- tool, selecting tools on the basis of their properties cal values and applying simple arithmetic that would best crack open nuts (chimpanzees: (Cantlon et al. 2009; Feigenson et al. 2004 ; Schrauf et al. 2012 ; capuchins: Visalberghi et al. Gallistel 1989 ; Meck and Church 1983 ). Even 2009 ). This suggests that NHPs can encode the the brain regions recruited during approximate requirements that a tool should meet in order to be number representations are present in both effective. We also know that the selection and uti- humans and NHPs (Dehaene et al. 2003 ; Nieder lization of tools are acquired and perfected 2005 ), and these cumulative fi ndings indicate throughout several years in an individual’s life that NHPs and humans share the foundation for with the aid of social observation and trial and complex, sophisticated mathematical thought. error (Pouydebat et al. 2006 ), and understanding Extensive research has shown that NHPs are causality is not necessary when repeated exposure able to spontaneously represent numerosities of to a problem allows an individual to learn what up to 9, at least on an ordinal scale (e.g., chimpan- does or does not work in order to obtain a desired zees: Boysen and Berntson 1989; squirrel mon- outcome. Solving problems with tools on the keys: Olthof et al. 1997; rhesus macaques: basis of previous experience does, however, Brannon and Terrace 1998; cotton-top tamarins: require an excellent memory, which NHPs most Uller et al. 2001 ). When one can mentally com- certainly have evolved. bine these values to create a new value without having to directly observe that new value, we call this arithmetic. Research has shown that rhesus Memory macaques are capable of simple addition and sub- traction. Rhesus are able to spontaneously com- Great apes are able to employ an incredible mem- pute addition operations over large numbers when ory; they show foresight in selecting, transport- the ratio difference between the two numbers is ing, and saving appropriate tools for an apparatus signifi cant (Flombaum et al. 2005 ), they are able in a different room that they have not seen for up to select a larger quantity of food following sub- to 14 h (Mulcahy and Call 2006 ) and replicating tractions of up to three pieces, and they are capa- novel actions after 24 h (Hopper 2010 ). ble of representing zero and equality when two Furthermore, both monkeys and apes are able to quantities are contrasted (Sulkowski and Hauser recall serial lists of symbolic images, and this 2001 ). This simple computational system of ana- working memory ability often outperforms that of log numerical representation allows for complex humans (Inoue and Matsuzawa 2007 ). This mem- algebra, calculus, and differential equations when ory for symbols requires that NHPs be capable of the human ability to represent numerical values a certain amount of mental representation. While symbolically is incorporated. However, most no NHP species spontaneously uses symbols, NHPs only need an understanding of quantity that they do have the capacity to associate semantic is suffi cient for comparing resources or social meaning to an arbitrary symbol. Both apes and group size, and with these selective priorities, monkeys have been able to apply Roman numer- more complex computations are a long way off. 3 Intelligence in Nonhuman Primates 33

dyads as compared to unrelated female-infant Social Cognition dyads and to match a photograph of an infant with the corresponding photograph of its mother. It is not only the physical environment that Work in wild vervet monkeys found similar imposes challenges on an individual; the social results; using a playback method in which an environment supplies its own unique set of chal- auditory stimulus of an infant vocalization was lenges. There are many benefi ts to living in played in the presence its mother and two addi- social groups, such as increased protection from tional adult females, Cheney and Seyfarth (1980) predators and greater success in intergroup con- found that the mother oriented toward the origin fl icts and resource defense (e.g., Majolo et al. of the vocalization, whereas the adult females 2008). However, there are potentially signifi cant oriented toward the mother prior to the mother’s costs imposed by increased sociality, such as movement. Both studies illustrate that group intensifi ed intragroup competition for food and members are aware of the special relationship mates and greater exposure to stronger, more between mother and infant and are able to differ- aggressive group members. Over evolutionary entiate that relationship on the basis of visual and history, NHPs have developed behavioral strate- auditory clues. gies and cognitive abilities to cope with the chal- NHPs that live in social groups have to keep lenges that arise from heightened sociality. In track of both kin and non-kin relationships, espe- this section, we will address what NHPs know cially with respect to the relative ranks of their about their social world and what cognitive skills social partners. The majority of primate species are necessary to coexist in social groups, such as have clearly established dominance hierarchies, knowledge of third-party relationships, transitive which reduce the incidence of physically aggres- inference of social relationships, and under- sive encounters between individuals (reviewed in standing what others see—and possibly what Kaufmann 1983 ; e.g., gelada baboons: Bergman they know as well. et al. 2005 ; olive baboons: Sapolsky 1992 ). However, having dominance hierarchies is of little use if individuals are unable to recognize and Knowledge of Third-Party remember specifi c relationships between conspe- Relationships cifi cs and other members of their social group. There is ample observational evidence from stud- Ample evidence demonstrates that NHPs are able ies of both captive and wild populations of NHPs to discriminate between individual members of illustrating that NHPs are aware of their own their species in visual (e.g., chimpanzees: Parr social status and respond appropriately when et al. 2000; Parr and de Waal 1999 ; rhesus they encounter a conspecifi c of higher or lower macaques: Parr et al. 2000; Pascalis and rank (e.g., in rhesus macaques, a subordinate Bachevalier 1998 ), auditory (e.g., vervets: individual fear-grimacing or presenting their Cheney and Seyfarth 1980, 1982 ; savannah hindquarters to a more dominant individual). baboons: Cheney et al. 1995; squirrel monkeys: Additionally, both observational and experimen- Snowdon and Cleveland 1980; rhesus macaques: tal evidence suggest that some species of pri- Rendall et al. 1996 ; Hansen 1976 ), and olfactory mates are able to “eavesdrop” on interactions (e.g., ring-tailed lemurs: Scordato and Drea 2007 ; between others and extrapolate from observed Palagi and Dapporto 2006 ) modalities, as well as domi nance relationships between two indivi- cross-modally (rhesus macaques: Adachi and duals to infer their own dominance to an unknown Hampton 2011 ; Sliwa et al. 2010 ). participant in that interaction, as long as their Recognition of social associations has been own dominance relationship to one of the part- documented across a variety of kin and non-kin ners in the interaction is known (e.g., if A knows relationships. For example, work by Dasser it is dominant to B, and A observes B being dom- (1988 ) has shown that in long-tailed macaques, inant to C, A is aware that it is dominant to C, conspecifi cs are able to identify mother-offspring without ever having to interact with C directly). 34 T. Mandalaywala et al.

ships can be determined by observation of Transitive Inference obvious behavioral interactions, such as affi lia- tion (e.g., grooming, mating) or aggression (e.g., The ability to produce appropriate responses to biting, chasing, threatening, etc.). However, there novel pairings of nonadjacent members in an is some evidence that NHPs are also capable of ordered series without having previous experi- picking up on more subtle social cues as well, ence with those pairings is known as transitive such as the direction in which an individual is inference and can be applied to a variety of social looking and what it looks at. interactions, not simply dominance interactions It has been well established that NHPs pay (e.g., reputation learning in chimpanzees: Subiaul attention to conspecifi cs; however, they do not et al. 2008 , kin-biased redirected aggression in attend to all physical aspects of conspecifi cs vervets: Cheney and Seyfarth 1999 , and recruit- equally. NHPs attend preferentially to the faces of ment of allies in bonnet macaques: Silk 1999 ). others, paying particular attention to the eye Most evidence for transitive inference comes region (Perrett and Mistlin 1990; Keating and from observational data (e.g., vervets: Cheney Keating 1982 ); preferential attention to eyes and and Seyfarth 1982 , 1986 ; rhesus macaques: eye movements increases gradually across devel- Judge 1982; Japanese macaques: Schino et al. opment (Ferrari et al. 2000 ). The primate brain 2006 ; Aureli et al. 1992 ; baboons: Bergman et al. contains neurons that preferentially respond to 2003 ), but it has been studied experimentally in a head orientation and eye gaze (Perrett et al. 1992 ; few species as well (e.g., ring-tailed and mon- Felleman and Van Essen 1991), and the facial goose lemurs: MacLean et al. 2008 ; capuchins: musculature of primates is more highly developed D’Amato et al. 1985 ; vervets: Borgeaud et al. than non-primate animals (Huber 1961 ; Huber 2013; chimpanzees: Gillan 1981 ). In an experi- 1931 ). Research on captive NHPs has revealed mental study on prosimians, MacLean et al. that many species engage in gaze following (e.g., (2008 ) found that although both ring-tailed and ring-tailed lemurs: Shepherd and Platt 2008 ; mongoose lemurs displayed some evidence of sooty mangabeys, rhesus macaques, stump-tailed transitive inference on a computer task, the ring- macaques, pig-tailed macaques, chimpanzees: tailed lemurs were markedly better than the mon- Tomasello et al. 1998 ), and anecdotal and obser- goose lemurs at this task, which the authors vational data from wild populations suggest that speculate is due to their larger social group size gaze following occurs in other primate species as and thus more cognitively complex social envi- well (e.g., long-tailed macaques: de Waal et al. ronment (but see le Roux and Bergman 2011 for 1976 ; hamadryas baboons: Kummer 1967 ; evidence in gelada baboons that knowledge of Savannah baboons: Byrne and Whiten 1992 ; third-party relationships has limits in extremely chimpanzees: Plooij 1978 ). large social groups). Gaze following can provide important information about the socio-ecological environment. By paying attention to what others are attending Gaze Following: Seeing What to, individuals can increase their own knowledge Others See or Knowing What by exploiting the information that others have Others Know? (e.g., food source location, predator detection, potential mates, etc.). For instance, an experi- Having established that NHPs pay attention to ment conducted on free-ranging rhesus macaques social relationships between other individuals, found that when subjects were given a choice and that some species are capable of inferring between stealing a grape from a human experi- additional relationships that they do not observe menter who was looking at the food or an experi- or experience directly, it is necessary to establish menter who was looking away from the food, the what behavioral clues individuals use to interpret monkey chose to steal from the one not gazing at these relationships. Many of the above relation- the grape (Flombaum and Santos 2005 ). This 3 Intelligence in Nonhuman Primates 35 held true whether the eyes alone or eyes and could not. They varied the means by which one head were oriented away from the food, as well human was unable to see the begging gesture as when the eyes of one experimenter were (i.e., gazing away from the chimpanzee, eyes occluded by a physical barrier (Flombaum and blindfolded, head covered by a bucket, or back Santos 2005 ). turned), and the chimpanzees begged Gaze can also be affected by the social context. indiscriminately from both humans in all In a study on wild black-crested macaques, conditions except for one—when the human Micheletta and Waller (2012 ) found that individ- experimenters back was turned. Povinelli and uals were quicker to follow the gaze of friends Eddy (1996a ) took this as evidence that chimpan- (i.e., conspecifi cs with whom an individual zees can be aware of what others can and cannot engaged in frequent, affi liative interactions) than see but only when the visual clue was very obvi- non-friends, suggesting that black-crested ous. Regardless of whether behavioral responses macaques prefer information-gathering from to gaze or even head orientation are the product friends, or that they are more attuned to social of learned associations or a deeper understand- cues from friends. In addition to differences in ing, several species of NHPs appear to be able to gaze following within a species, there appear to use gaze to gather information about or exploit be differences in how gaze is followed between the attentional state of another. However, the species. For example, Santos and Hauser ( 1999 ) debate about what NHPs know regarding the found that cotton-top tamarins cannot use eye mental states of others is ongoing and stretches gaze alone to predict future behavior of a human from the domains of gaze following to the under- experimenter but rather need head and eye orien- standing (or lack thereof) of intentional versus tation in order to predict future action accurately. accidental action and to the major debate over Using a combination of gaze and gesture, whether NHPs possess a Theory of Mind. MacLean and Hare (2011 ) found that chimpanzees and bonobos were more likely to look at a target object that had been looked at and gesticu- Understanding the Mental States lated toward previously, as compared to a previ- of Others: Theory of Mind ously ignored object, suggesting that they are attuned to what others are paying attention to. Theory of Mind refers to the ability to attribute It has been suggested that gaze following is a mental states, such as perception, beliefs, inten- prerequisite for other cognitive abilities such as tions, desires, and knowledge, to others. Premack joint attention, predicting the actions of others, and Woodruff (1978 ) investigated if one adult social learning, and imitation (Emery et al. 1997 ). chimpanzee was able to infer the goals and inten- However, what gaze following actually means in tions of a human actor, and although they origi- terms of cognitive understanding is still up for nally came to the conclusion that their subject debate, with some arguing that gaze following is understood what the human actor was attempting a simple reaction to cues such as body or eye ori- to do (i.e., the actor’s original goal), subsequent entation (e.g., Bering and Povinelli 2003 ), and research over the next three decades suggested others contending that gaze following implies that Premack and Woodruff’s result was prob- that individuals are aware of what others see and lematic due to methodological issues and that, in possibly what others know (e.g., chimpanzees: fact, chimpanzees and other NHPs were unable Hare et al. 2001; debate reviewed in Seed and to understand the mental states of others Tomasello 2010 , as well as Povinelli and Vonk (reviewed in Tomasello and Call 1997 ). 2004). In part, this debate results from ambigu- However, the past decade has brought a sea ous data. For example, Povinelli and Eddy change in how many primatologists and compar- (1996a ) presented young chimpanzees with two ative psychologists study Theory of Mind, and humans, one of whom could see the chimpanzee these new methods have produced evidence giving a food begging gesture and one of whom which some believe suggests that a few species of 36 T. Mandalaywala et al.

NHPs do possess some components of a Theory zees, but not capuchins, are aware of what others of Mind (e.g., Seed and Tomasello 2010 ; can and cannot see, and that in this competitive Tomasello et al. 2003 ). Perhaps the most signifi - context, they will exploit that information to gain cant factor to spur this change was the realization food rewards while avoiding negative social con- that previous tests of Theory of Mind in NHPs sequences. In addition to demonstrating that were not utilizing the natural behaviors and chimpanzees can use gaze to infer what others environmental conditions of each species (e.g., can see, and perhaps what they know (Hare et al. Hare and Tomasello 2004 ). 2001), this fi nding also illustrates the importance As complex cognition is hypothesized to have of conducting tests in an ecologically relevant evolved to solve problems encountered in the manner, which allows animals to behave in a nat- socio-ecological environment, researchers real- ural way (e.g., Bräuer et al. 2007 ; Hare and ized that they could exploit natural tendencies in Tomasello 2004 ). behavior to gain insight into the NHP’s mind, and These new evidence and methods have not that the negative results that had been found thus convinced everyone, however, and the ensuing far might be a consequence of trying to elicit controversy and debate have caused a schism in responses that were contrary to a species’ natural the research community on Theory of Mind in behaviors (Hare and Tomasello 2004 ). For exam- nonhumans. In contrast to those who accept that ple, many studies had found that chimpanzees the current methods allow researchers to investi- did not utilize gaze cues from human experiment- gate what one individual knows about the mind ers to select a container that had food, and that of another (as reviewed in Seed and Tomasello they would choose indiscriminately between two 2010 ), there are those who believe that research- containers (e.g., Itakura et al. 1999 ; Call et al. ers have yet to devise a test that accurately distin- 1998 ; Tomasello et al. 1997; Povinelli and Eddy guishes between subjects who only use behavioral 1996b ). This was taken as evidence that chim- cues and learned associations and subjects who panzees did not understand that gaze is an indica- use a combination of behavioral cues/associa- tor of what others were attending to or interested tions as well as inferring the mental states of oth- in and thus that chimpanzees were not able to ers (Povinelli and Vonk 2004). Although those reason about the knowledge states of others. who espouse the latter position do not deny the However, chimpanzees do not typically coop- potential existence of a Theory of Mind in NHPs erate to acquire food, and Hare et al. ( 2000 ) (but see Penn and Povinelli 2007 ), they fi nd the hypothesized that they would perform better current data inadequate in distinguishing between when tested in a competitive situation. Putting a whether individuals are only paying attention to dominant and subordinate individual in the same behavior or whether they are inferring something testing area, in sight of each other but separated about the mental states of others in addition to by two clear barriers, Hare et al. (2000 ) placed reading behavioral clues (e.g., Povinelli and Vonk two pieces of food in the arena; one food was in 2004 ; Karin-D’Arcy and Povinelli 2002 ). an open area where both animals could see it, and It has been suggested that the best way to rec- the other was placed behind a visual barrier tify the current crisis is to use versions of tasks where only the subordinate had visual access to that test false beliefs in verbal subjects (e.g., the it. Hare et al. (2000 ) found that when the subordi- Sally-Anne or Smarties task often used to study nate was released into the food arena prior to the Theory of Mind in human children) that are dominant individual, the subordinate went behind adapted for use in non- or preverbal individuals the visual barrier and ate the piece of food that (Lurz and Krachun 2011 ). Thus far, few studies the dominant could not see. Using an almost have used false-belief tasks to investigate Theory identical task, Hare et al. (2003 ) did not fi nd evi- of Mind in nonhuman primates (but see Call and dence that capuchin monkeys are cognizant of Tomasello 1999 ). One study in rhesus macaques, what another individual sees. Hare et al. ( 2001 ) using a violation-of-expectancy paradigm com- suggested that this demonstrates that chimpan- monly used in studies of human infants (Spelke 3 Intelligence in Nonhuman Primates 37

1985 ), found that subjects exhibited surprise (by toward themselves, grooming body parts that looking longer) when a human experimenter were not visible without the aid of the mirror looked in the wrong location for an object when (Gallup 1970 ). In light of these observations and that experimenter knew where the object was, by with the hypothesis that in order to display self- virtue of having seen it placed there earlier recognition, an individual must have a concept of (Marticorena et al. 2011). However, the rhesus self, Gallup developed the mark test to investigate macaques did not appear to make any prediction whether non- or preverbal individuals displayed (i.e., they looked equally at all possible locations) mirror self-recognition (MSR) (Gallup 1977 ). In about where the experimenter would look the mark test, while a subject is unconscious or when the object was moved from its original distracted, a mark in a conspicuous color is location without the experimenter’s knowledge painted onto a part of the body that the individual (Marticorena et al. 2011 ). This suggests that cannot feel or see without a mirror (usually its although rhesus macaques can predict what an forehead). Once awake, the subject is placed in individual will do when that individual has a true front of a mirror and the number of touches belief, they cannot predict behavior when an indi- directed to the mark are counted, as are the num- vidual has a false belief (Marticorena et al. 2011 ). ber of touches that an individual makes to other Importantly, none of the species that have been parts of their body. This test has been applied to a tested thus far have demonstrated an understand- wide variety of species, and at present it is estab- ing of false beliefs (e.g., Krachun et al. 2009a , b ; lished that humans and most great apes pass the Kaminski et al. 2008 ; Hare et al. 2001 ; Call and mark test by touching the mark more frequently Tomasello 1999 ). in front of the mirror than when no mirror is pres- Keeping in mind the possibility that what ent (chimpanzees: de Veer et al. 2002 ; Gallup appears to be a cognitively complex skill could in 1977), but monkeys do not pass this test (reviewed fact be attributed to learned associations between in Anderson and Gallup 2011 ; Povinelli 1987 ) behaviors and subsequent actions, we will review and show persistent social responses to mirrors the literature which suggests that NHPs might over prolonged exposure. have an understanding of the mental states of However, as with most areas of primate intel- both themselves and others. ligence, the line in the sand between monkeys and apes is under dispute, with some suggesting that an inability to pass the mark test does not Concept of Self mean that monkeys do not have the ability for self-recognition (e.g., Seyfarth and Cheney 2000 ; Before scientists became interested in the ques- Heyes 1994 ). There is some evidence that mon- tion of whether NHPs had a concept of the minds keys do in fact show self-directed behaviors when of others, they were interested in whether indi- in front of a mirror that do not take place in its viduals had a concept of self (e.g., Gallup 1970). absence (e.g., rhesus macaques: Rajala et al. The relationship between understanding that one 2010 ) and that they are capable of equating vid- has his/her own mind and being able to attribute eos of parts of their own bodies with themselves mentality to others is not well understood, even in (e.g., capuchins: Anderson et al. 2009 ; Japanese humans (e.g., Happé 2003 ). However, it is almost macaques: Iriki et al. 2001 ), and this indicates certain that a concept of self is necessary for a that at least precursors to self-recognition may be fully developed Theory of Mind (Gallup 1982 ). present in some monkey species (e.g., pygmy Gallup ( 1970) observed that when chimpan- marmosets: Eglash and Snowdon 1983 ). zees were placed in front of a mirror, they ini- Moreover, one species of great apes does not pass tially reacted as though their refl ection were the mark test—gorillas—suggesting that the evo- another chimpanzee and acted socially towards lution of self-recognition does not divide mon- it. However, after a few days of mirror exposure, keys and apes, but that performance on the mark the chimpanzees began to use the mirror to act test is affected by other factors such as a natural 38 T. Mandalaywala et al. inclination not to look directly at others that may et al. (2004 ) found that the subject’s rate of frus- diminish an individual’s willingness to partici- tration behaviors was much higher when the pate in the test (reviewed in Suarez and Gallup experimenter was unwilling to dispense food 1981). Drawing from inconsistencies in the data than when the experiment was unable. on MSR, it has been suggested that the visual Additionally, the subjects left the testing area and modality might not always be the best test for terminated the session earlier in the unwilling self-recognition, and that some species might condition, again suggesting that chimpanzees can perform better in an auditory or olfactory modal- discriminate between humans who will not help ity or even simply by testing for an understanding them because they cannot and those who will not of oneself versus others in a more social setting help because they do not want to (Call et al. (Seyfarth and Cheney 2000 ). 2004 ). Although few monkey species have been tested on their understanding of intentions, capuchins have also shown the ability to discriminate Inferring Intentions between an experimenter that is unwilling and one that is unable (Phillips et al. 2009 ). Some of the fi rst work to investigate what NHPs Third, we can test attentiveness to intentional- understand about the mental states of others came ity by observing what types of helping behavior from studies asking whether primates could pre- NHPs offer when they observe an experi- dict the goals of others. This cognitive ability menter struggling to achieve a particular goal. would provide benefi ts for individuals in their Using this method, Warneken and Tomasello natural environment, enabling them to predict not ( 2006) had a human experimenter accidentally only what others will do in highly prescribed sit- or intentionally drop an object they were carry- uations but also in novel ones (Call and Tomasello ing and observed whether or not chimpanzees 2008 ). What NHPs understand about the goals of would retrieve the object and return it to the others can be tested in multiple ways. experimenter; they found that chimpanzees First, we can ask whether NHPs are able to would help in the accidental but not the inten- discriminate between intentional and accidental tional condition. Warneken et al. ( 2007) also actions. For example, Povinelli et al. ( 1998 ) employed a helping paradigm using two chim- tested whether chimpanzees preferred to receive panzees, where one could assist the other in their food from an experimenter who had previously goal of opening a door, and found similar results. been clumsy (e.g., accidentally spilling juice Capuchins also demonstrated an understanding when trying to hand it to the chimp) or an experi- of intentions, by offering help to a human experimenter who intentionally spilled the juice instead menter (Barnes et al. 2008 ), but in contrast to the of handing it. Although the results were not chimpanzees (Warneken et al. 2007 ), the capu- strong, they found a general preference for the chins would not assist the experimenter if there clumsy experimenter as compared to the one who was a cost to themselves. This illustrates an intentionally did not deliver the juice (Povinelli interesting species difference that should be et al. 1998), suggesting that chimpanzees might taken into consideration when testing other pri- be able to discriminate between accidents and mates, as other species may abandon helping intentional actions (see Call and Tomasello 1998 behavior in certain contexts, resulting in null test for evidence of this ability in orangutans). results, even if they are mentally capable of Second, this inspires a similar question of understanding the experimenter’s goal. whether NHPs are able to differentiate between Fourth, we can test NHPs understanding of individuals who are unable to help and those who intentionality by testing their imitation of rational are unwilling to help. To test this, Call et al. or irrational behavior. In an experiment using ( 2004) had a human experimenter give food to a methods co-opted from studies of human infants chimpanzee until the experimenter became either (Gergely et al. 2002 ), enculturated chimpanzees unwilling or unable to give food any longer. Call were exposed to a human experimenter carrying 3 Intelligence in Nonhuman Primates 39 out an action in an unusual way because the intentions of others, and they can discriminate experimenter was physically constrained (e.g., between an intentional and an accidental action. the experimenter opens a door with his foot Still, this is not well studied outside of great apes, instead of his hands because his hands are occu- and even within great apes, most research has pied carrying an object: Buttelmann et al. 2007 ) been conducted on chimpanzees. More work is and to an experimenter carrying out the same necessary to gain insight into whether other unusual action even though there was no reason- NHPs have a concept of intentionality, both using able constraint. Buttelmann et al. (2007 ) found experimental methods as well as through obser- that the chimpanzees were more likely to imitate vation of wild populations. the experimenter’s unusual actions (using their foot) to achieve the goal of opening the door after observing the unconstrained condition, but they Deception would simply open the door with their hands following observation of the constrained condition. Observation of wild and free-ranging populations This suggests that the chimpanzees were aware has actually produced some of the most compel- of the goal of the experimenter and were able to ling evidence that NHPs understand the mental conclude that the experimenter in the constrained states of others, especially in the domain of condition was only using his/her feet because deception. For example, observational research they could not use their hands, but the experi- has revealed evidence of tactical deception that menter in the unconstrained condition was using emerges in multiple modalities. NHPs will stay their feet for some other reason (see Buttelmann quiet and withhold vocalizations during certain et al. 2012 for similar results using a novel situations (e.g., sneaky mating, competing for method). food, territory defense in wild chimpanzees: Interestingly, Buttelmann et al. ( 2007 ) did not Watts and Mitani 2001 ; Wilson et al. 2001 ), espe- fi nd the same result with non-enculturated chim- cially when alerting others to their location could panzees (see also Tennie et al. 2012 ), suggesting result in aggression (e.g., evidence of punishment that imitation is a learned behavior in chimpan- for withholding food calls in rhesus macaques: zees. Previous studies have also suggested that Hauser 1992 ). Le Roux et al. (2013 ) found evi- chimpanzees and the other great apes are actually dence for tactical deception in gelada baboons, better emulators (carrying out an intended goal where males and females were less likely to without carrying out the specifi c actions demon- vocalize during extra-pair copulations and were strated in achieving that goal) than imitators more likely to engage in this sneaky mating when (mimicking specifi c actions step by step), as the female’s primary mate was further away, thus compared to human children who are prone to avoiding an aggressive punishment. overimitation (e.g., Tennie et al. 2006 , 2010 ; Experimental studies of deception can also Horner and Whiten 2005 ; but see Whiten et al. shed light on NHPs understanding of the per- 2009 for evidence of imitation in wild chimpan- spective and knowledge of others. For example, zees). Compared to great apes, monkeys have both rhesus macaques (Santos et al. 2006 ) and been studied much less, yet there is some evi- chimpanzees (Melis et al. 2006 ) choose to steal dence that capuchin monkeys are able to imitate food from a silent (nonfunctional bells attached) the actions of others to achieve a goal (van de as opposed to noisy (functional bells attached) Waal and Whiten 2012 ). However, observational container when there is a food competitor present data from wild and free-ranging populations of (but see Bräuer et al. 2008 for a contradictory monkeys typically suggest that imitation is infre- result in chimpanzees). Additionally, Hare et al. quent and rarely utilized in natural contexts ( 2006 ) found that when chimpanzees were put in (reviewed in Visalberghi and Fragaszy 1990 ). a competitive situation with a human experi- Together, these methods and studies indicate menter over a contested piece of food, the chim- that NHPs are capable of understanding the panzees would manipulate the visual perception 40 T. Mandalaywala et al. of the experimenter to make their approach to the in cognition could very well be both shaped by food item less obvious (i.e., by hiding behind an and have consequences for social relationships occluder) and thereby hiding their intention to across a given species’ lifespan. grab the food from the experimenter. Although Despite the acknowledged utility for a truly there is evidence that rhesus macaques are able to comparative approach to studies of intelligence, deceive others in visual (Flombaum and Santos few species of primates have been studied as 2005 ) and auditory (Santos et al. 2006 ) domains, thoroughly as the great apes, with a particular not all species of monkeys appear able to do so attention on chimpanzees and humans. We expect (e.g., long-tailed macaques: Kummer et al. 1996 ). vast differences in the cognitive abilities of great Returning to the idea that challenges in the apes and humans, as the human brain size is three social environment spurred the evolution of times that of the other great apes and humans increased cognition and larger brains, multiple show a much larger behavioral repertoire, but relationships between deception and various spe- previous research has shown that humans, at least cies of NHPs have been noted to better under- early in development are not any more intelligent stand its evolution in primates. In conducting a in general than great apes (reviewed in Tomasello meta-analysis of previously published papers, and Hermann 2010 ). This was best demonstrated Byrne and Whiten ( 1992 ) found that deception is through the results of a battery of tests adminis- not found across all species and clades of NHPs tered to chimpanzees, orangutans, and 2-year-old equally; instead, deception is found most fre- humans (Hermann et al. 2007 ) in which humans’ quently in great apes and infrequently, if at all, in socio-cognitive abilities (e.g., reading intentions, prosimians. Additionally, Byrne and Corp (2004 ) communicating, and learning socially) were found a relationship between neocortical volume greatly enhanced relative to the other great apes, and use of deception across the Primate order. In but all three species were equally able to reason order to understand this relationship more fully, about the physical world (e.g., understanding additional studies on previously untested species causality, space, and quantity) (see also Hermann of NHPs are necessary. et al. 2010 ). Tomasello et al. ( 2005) hypothesized that the crucial difference between humans and nonhu- A Comparative Approach man primates is humans’ ability to reason, understand, learn from, and act effectively in their A call for study of additional species is common unique social environments in a concept we call and is often repeated in the articles and studies culture. But while humans are undeniably special described above. Within the order Primates, there and intelligent in a way that is not seen in nonhu- is a remarkable diversity of species, and exten- man species, it is perhaps premature to conclude sive studies of many primate species have shown that the crucial difference between humans and that these species exist in a wide variety of habi- NHPs is their ability to reason and know about tats and exhibit many types of mating and social the social world. Many species of NHPs live in systems as well as behaviors. While much extraordinarily complex social groups (e.g., research thus far has described socio-ecological gelada baboons: Snyder-Mackler et al. 2011 ), differences between species, far fewer studies even larger and more complex than the social have examined species-specifi c differences in groups early hominids were thought to live in. As cognition. As primates are considered one of the shown by Hare et al. (2001 ), sometimes the ques- most cognitively complex taxa, with the evolution might simply be investigated in an unsuit- tion of primate intelligence hypothesized to be able way, making it unintelligible to the species driven by their high degree of sociality (Dunbar being tested. As we are humans, we better under- 1998), the utility of a comparative approach to stand the best way to ask other humans what they the study of cognition cannot be overstated think and know about themselves and others, but (MacLean et al. 2012 ), especially as differences more thought and consideration, coupled with an 3 Intelligence in Nonhuman Primates 41 understanding of the social and ecological con- Borgeaud, C., van de Waal, E., & Bshary, R. (2013). Third- text a species inhabits, is necessary to devise party ranks knowledge in wild vervet monkeys (Chlorocebus aethiops pygerythrus). PLoS One, 8, 1–4. studies that are relevant and sensible to NHPs and Boysen, S. T., & Berntson, G. G. (1989). Numerical com- allow us to distinguish between the true mental petence in a chimpanzee (Pan troglodytes). Journal of states of NHPs and simple behavioral responses. Comparative Psychology, 103 , 23–31. The past decade has brought remarkable advances Brannon, E. M., & Terrace, H. S. (1998). Ordering of the numerosities 1 to 9 by monkeys. Science, 282 , in our understanding of primate intelligence 746–749. (Whiten 2013 ), and by utilizing a comparative Bräuer, J., Call, J., & Tomasello, M. (2007). Chimpanzees approach and devising clever, species-specifi c really know what others can see in a competitive situ- cognitive tests, the next decade promises to be ation. Animal Cognition, 10 , 439–448. Bräuer, J., Call, J., & Tomasello, M. (2008). Chimpanzees just as productive. do not take into account what others can hear in a competitive situation. Animal Cognition, 11 , 175–178. Breuer, T., Ndoundou-Hockemba, M., & Fishlock, V. (2005). First observation of tool use in wild gorillas. References PLoS Biology, 3 , 2041–2043. Buttelmann, D., Carpenter, M., Call, J., & Tomasello, M. Adachi, I., & Hampton, R. R. (2011). Rhesus monkeys (2007). Enculturated chimpanzees imitate rationally. see who they hear: Spontaneous cross-modal memory Developmental Science, 10 , F31–F38. for familiar conspecifi cs. PLoS One, 6 , 1–8. Buttelmann, D., Schütte, S., Carpenter, M., Call, J., & Aiello, L. C., & Wheeler, P. (1995). The expensive-tissue Tomasello, M. (2012). Great apes infer others’ goals hypothesis: The brain and the digestive system in based on context. Animal Cognition, 15 , 1037–1053. human and primate evolution. Current Anthropology, Byrne, R. W., & Corp, N. (2004). Neocortex predicts 36 , 199–221. deception rate in primates. Proceedings of the Royal Anderson, J. R., & Gallup, G. G., Jr. (2011). Which pri- Society of London: Series B, 271 , 1693–1699. mates recognize themselves in mirrors? PLoS Biology, Byrne, R. W., & Whiten, A. (1992). Cognitive evolution 9 , 1–3. in primates: Evidence from tactical deception. Man, Anderson, J. R., Kuroshima, H., Paukner, A., & Fujita, K. 27 , 609–627. (2009). Capuchin monkeys (Cebus apella) respond to Cacchione, T., & Horst, K. (2004). Recognizing impossi- video images of themselves. Animal Cognition, 12 , ble object relations: Intuitions about support in chim- 55–62. panzees (Pan troglodytes). Journal of Comparative Aureli, F., Cozzolino, R., Cordischi, C., & Scucchi, S. Psychology, 118 , 140–148. (1992). Kin-oriented redirection among Japanese Call, J., & Tomasello, M. (1998). Distinguishing inten- macaques—An expression of a revenge system. tional from accidental actions in orangutans (Pongo Animal Behaviour, 44 , 283–291. pygmaeus), chimpanzees (Pan troglodytes) and human Barnes, J. L., Hill, T., Langer, M., Martinez, M., & Santos, children (Homo sapiens). Journal of Comparative L. R. (2008). Helping behaviour and regard for others Psychology, 112 , 192–206. in capuchin monkeys (Cebus apella). Biology Letters, Call, J., & Tomasello, M. (1999). A nonverbal false belief 4 , 638–640. task: The performance of children and great apes. Barrett, L., Henzi, P., & Rendell, D. (2007). Social brains, Child Development, 70 (2), 381–395. simple minds: Does social complexity really require Call, J., & Tomasello, M. (2008). Does the chimpanzee cognitive complexity? Philosophical Transactions of have a theory of mind? 30 years later. Trends in the Royal Society B, 362 , 561–575. Cognitive Sciences, 12 , 187–192. Bergman, T. J., Beehner, J. C., Cheney, D. L., & Seyfarth, Call, J., Hare, B., & Tomasello, M. (1998). Chimpanzee R. M. (2003). Hierarchical classifi cation by rank and gaze following in an object-choice task. Animal kinship in baboons. Science, 302 , 1234–1236. Cognition, 1 , 89–99. Bergman, T. J., Beehner, J. C., Cheney, D. L., Seyfarth, Call, J., Hare, B., Carpenter, M., & Tomasello, M. (2004). R. M., & Whitten, P. L. (2005). Correlates of stress in ‘Unwilling’ versus ‘unable’: Chimpanzees’ under- free-ranging male chacma baboons, Papio hamadryas standing of human intentional action. Developmental ursinus. Animal Behaviour, 70 , 703–713. Science, 7 , 488–498. Bering, J. M., & Povinelli, D. J. (2003). Comparing cogni- Cantlon, J. F., Platt, M. L., & Brannon, E. M. (2009). tive development. In D. Maestripieri (Ed.), Primate Beyond the number domain. Trends in Cognitive psychology (pp. 205–233). Cambridge, MA: Harvard Sciences, 13 , 83–91. University Press. Cheney, D. L., & Seyfarth, R. M. (1982). Recognition of Boesch, C., & Boesch, H. (1990). Tool use and tool mak- individuals within and between groups of free-ranging ing in wild chimpanzees. Folia Primatologica, 54 , vervet monkeys. American Zoologist, 22 , 519–529. 86–99. Cheney, D. L., & Seyfarth, R. M. (1986). The recognition of social alliances among vervet monkeys. Animal Behaviour, 34 , 1722–1731. 42 T. Mandalaywala et al.

Cheney, D. L., & Seyfarth, R. M. (1999). Recognition of Flombaum, J. I., & Santos, L. R. (2005). Rhesus monkeys other individuals’ social relationships by female attribute perceptions to others. Current Biology, 15 , baboons. Animal Behaviour, 58 , 67–75. 447–452. Clutton-Brock, T. H., & Harvey, P. (1980). Primates, brains, Flombaum, J. I., Junge, J. A., & Hauser, M. D. (2005). and ecology. Journal of Zoology, 190 , 309–323. Rhesus monkeys (Macaca mulatta) spontaneously D’Amato, M. R., Salmon, P., Loukas, E., & Tomie, A. compute addition operations over large numbers. (1985). Symmetry and transitivity of conditional rela- Cognition, 97 , 315–325. tions in monkeys (Cebus paella) and pigeons (Columba Fragaszy, D., Izar, P., Visalberghi, E., Ottoni, E. B., & de livia). Journal of the Experimental Analysis of Oliveira, M. G. (2004). Wild capuchin monkeys Behavior, 44 , 35–47. (Cebus libidinosus) use anvils and stone pounding Dasser, V. (1988). A social concept in Java monkeys. tools. American Journal of Primatology, 64 , Animal Behaviour, 36 , 225–230. 359–366. de Bois, S. T., & Novak, M. A. (1994). Object perma- Gallistel, C. R. (1989). Animal cognition: The representa- nence in rhesus monkeys (Macaca mulatta). Journal of tion of space, time and number. Annual Review of Comparative Psychology, 108 , 318–327. Psychology, 40 , 155–189. de Bois, S. T., Novak, M. A., & Bond, M. (1998). Object Gallup, G. G., Jr. (1970). Chimpanzees: Self-recognition. permanence in orangutans (Pongo pygmaeus) and Science, 167 , 86–87. squirrel monkeys (Saimiri sciureus). Journal of Gallup, G. G., Jr. (1977). Self-recognition in primates: A Comparative Psychology, 112 , 137–152. comparative approach to the bidirectional properties of de Veer, M. W., Gallup, G. G., Theall, L. A., van den Bos, consciousness. American Psychologist, 32 , 329–338. R., & Povinelli, D. J. (2002). An 8-year longitudinal Gallup, G. G., Jr. (1982). Self-awareness and the emer- study of mirror self-recognition in chimpanzees (Pan gence of mind in primates. American Journal of troglodytes). Neuropsychologia, 1493 , 1–6. Primatology, 2 , 237–248. de Waal, F. B. M., van Hooff, J. A. R. A. M., & Netto, Gergely, G., Bekkering, H., & Király, I. (2002). Rational W. J. (1976). An ethological analysis of types of ago- imitation in preverbal infants. Nature, 415 , 755. nistic interaction in a captive group of Java-monkeys Gillan, D. J. (1981). Reasoning in the chimpanzee: (Macaca fascicularis). Primates, 17 , 257–290. II. Transitive inference. Journal of Experimental Deaner, R. O., van Schaik, C. P., & Johnson, V. (2006). Do Psychology: Animal Behavior Processes, 7 , 150–164. some taxa have better domain-general cognition than Grueter, C. C., Bissonnette, A., Isler, K., & Van Schaik, others? A meta-analysis of nonhuman primate studies. C. P. (2013). Grooming and group cohesion in pri- Evolutionary Psychology, 4 , 149–196. mates: Implications for the evolution of language. Dehaene, S., Piazza, M., Pinel, P., & Cohen, L. (2003). Evolution and Human Behavior, 34 , 61–68. Three parietal circuits for number processing. Gumert, M. D., Kluck, M., & Malaivijitnond, S. (2009). Cognitive Neuropsychology, 20 , 487–506. The physical characteristics and usage patterns of Dunbar, R. I. M. (1998). The social brain hypothesis. stone axe and pounding hammers used by long-tailed Evolutionary Anthropology, 6 , 178–190. macaques in the Andaman Sea region of Thailand. Dunbar, R. I. M., & Shultz, S. (2007). Evolution in the American Journal of Primatology, 71 , 594–608. social brain. Science, 317 , 1344–1347. Hansen, E. W. (1976). Selective responding by recently Eglash, A. R., & Snowdon, C. T. (1983). Mirror-image separated juvenile rhesus monkeys to the calls of their responses in pygmy marmosets (Cebuella pygmaea). mother. Developmental Psychobiology, 9 , 83–88. American Journal of Primatology, 5 , 211–219. Hanus, D., & Call, J. (2008). Chimpanzees infer the loca- Emery, N. J., Lorincz, E. N., Perrett, D. I., Oram, M. W., tion of a reward on the basis of the effect of its weight. & Baker, C. I. (1997). Gaze following and joint atten- Current Biology, 18 , 370–372. tion in rhesus monkeys (Macaca mulatta). Journal of Happé, F. (2003). Theory of mind and the self. Annals of Comparative Psychology, 111 , 286–293. the New York Academy of Sciences, 1001 , 134–144. Feigenson, L., Dehaene, S., & Spelke, E. (2004). Core Hare, B., & Tomasello, M. (2004). Chimpanzees are more systems of number. Trends in Cognitive Sciences, 8 , skillful in competitive than in cooperative cognitive 307–314. tasks. Animal Behaviour, 68 , 571–581. Felleman, D. J., & Van Essen, D. C. (1991). Distributed Hare, B., Call, J., Agnetta, B., & Tomasello, M. (2000). hierarchical processing in the primate cerebral cortex. Chimpanzees know what conspecifi cs do and do not Cerebral Cortex, 1 , 1–47. see. Animal Behaviour, 59 , 771–785. Ferrari, P. F., Kohler, E., Fogassi, L., & Gallese, V. (2000). Hare, B., Call, J., & Tomasello, M. (2001). Do chimpan- The ability to follow eye gaze and its emergence dur- zees know what conspecifi cs know? Animal Behaviour, ing development in macaque monkeys. Proceedings of 6 , 139–151. the National Academy of Science, 97 , 13997–14002. Hare, B., Addessi, E., Call, J., Tomasello, M., & Fish, J. L., & Lookwood, C. A. (2003). Dietary constraints Visalberghi, E. (2003). Do capuchin monkeys, Cebus on encephalization in primates. American Journal of apella, know what conspecifi cs do and do not see? Physial Anthropology, 120 , 171–181. Animal Behaviour, 65 , 131–142. 3 Intelligence in Nonhuman Primates 43

Hare, B., Call, J., & Tomasello, M. (2006). Chimpanzees Karin-D’Arcy, R., & Povinelli, D. J. (2002). Do chimpan- deceive a human competitor by hiding. Cognition, 10 , zees know what each other see? A closer look. 495–514. International Journal of Comparative Psychology, 15 , Hauser, M. D. (1992). Costs of deception: Cheaters are 21–54. punished in rhesus monkeys (Macaca mulatta). Kaufmann, J. H. (1983). On the defi nitions and functions Proceedings of the National Academy of Sciences of of dominance and territoriality. Biological Reviews, the United States of America, 89 , 12137–12139. 58 , 1–20. Hauser, M. D. (1998). A non-human primate’s expecta- Keating, C. F., & Keating, E. G. (1982). Visual scan pat- tions about object motion and destination: The impor- terns of rhesus monkeys viewing faces. Perception, 11 , tance of self- propelled movement and animacy. 211–219. Developmental Science, 1 , 31–38. King, B. J. (1986). Extractive foraging and the evolution Hermann, E., Call, J., Hernandez-Lloreda, M. V., Hare, of primate intelligence. Human Evolution, 1 , B., & Tomasello, M. (2007). Humans have evolved 361–372. specialized skills of social cognition: The cultural Krachun, C., Call, J., & Tomasello, M. (2009a). Can intelligence hypothesis. Science, 317 , 1360–1366. chimpanzees (Pan troglodytes) discriminate appear- Hermann, E., Hernandez-Lloreda, M. V., Call, J., Hare, B., ance from reality? Cognition, 112 , 435–450. & Tomasello, M. (2010). The structure of individual Krachun, C., Carpenter, M., Call, J., & Tomasello, M. differences in the cognitive abilities of children and (2009b). A competitive nonverbal false belief task for chimpanzees. Psychological Science, 21 , 102–110. children and apes. Developmental Science, 12 , Heyes, C. M. (1994). Refl ections on self-recognition in 521–535. primates. Animal Behaviour, 47 , 909–919. Kummer, H. (1967). Tripartite relations in hamadryas Holekamp, K. E. (2007). Questioning the social intelli- baboons. In S. A. Altmann (Ed.), Social communica- gence hypothesis. Trends in Cognitive Sciences, 11 , tion among primates (pp. 63–71). Chicago: The 65–69. University of Chicago Press. Hood, B. M., Hauser, M. D., Anderson, L., & Santos, L. Kummer, H., Anzenberger, G., & Hemelrijk, C. K. (1996). (1999). Gravity biases in a non-human primate? Hiding and perspective taking in long-tailed macaques Developmental Science, 2 , 35–41. (Macaca fascicularis). Journal of Comparative Hopper, L. M. (2010). Deferred imitation in children and Psychology, 110 , 97–102. apes: Children imitate after a delay, but can apes ape in le Roux, A., & Bergman, T. J. (2011). Indirect rival assess- a similar fashion? The Psychologist, 23 , 2–5. ment in a social primate, Theropithecus gelada. Horner, V., & Whiten, A. (2005). Causal knowledge and Animal Behaviour, 83 , 249–255. imitation/emulation switching in chimpanzees (Pan le Roux, A., Snyder-Mackler, N., Roberts, E. K., Beehner, troglodytes) and children (Homo sapiens). Animal J. C., &Bergman, T. J. (2013). Evidence for tactical Cognition, 8 , 164–181. concealment in a wild primate. Nature Huber, E. (1931). Evolution of facial musculature and Communications , 4. doi: 10.1038/ncomms2468 . facial expression . Baltimore: John Hopkins Press. Lurz, R. W., & Krachun, C. (2011). How could we know Huber, E. (1961). The facial musculature and its innerva- whether nonhuman primates understand others’ inter- tion. In C. G. Hartman & W. L. Straus (Eds.), Anatomy nal goals and intentions? Solving Povinelli’s problem. of the rhesus monkey (Macaca mulatta) (pp. 176–188). Review of Philosophy and Psychology, 2 , 449–481. New York: Hafner. MacLean, E. L., & Hare, B. (2011). Bonobos and chim- Humphrey, N. (1976). The social function of intellect. In panzees infer the target of another’s attention. Animal P. P. G. Bateson & R. A. Hinde (Eds.), Growing points Behaviour, 83 , 345–353. in ethology (pp. 303–317). Cambridge, UK: MacLean, E. L., Merritt, D. J., & Brannon, E. M. (2008). Cambridge University Press. Social complexity predicts transitive reasoning in pro- Inoue, S., & Matsuzawa, T. (2007). Working memory of simian primates. Animal Behaviour, 76 , 479–486. numerals in chimpanzees. Current Biology, 17 , MacLean, E. L., Matthews, L. J., Hare, B. A., Nunn, C. L., R1004–R1005. Anderson, R. C., Aureli, F., Brannon, E. M., Call, J., Iriki, A., Tanaka, M., Obayashi, S., & Iwamura, Y. (2001). Drea, C. M., Emery, N. J., Haun, D. B. M., Herrmann, Self-images in the video monitor coded by monkey E., Jacobs, L. F., Platt, M. L., Rosati, A. G., Sandel, intraparietal neurons. Neuroscience Research, 40 , A. A., Schroepfer, K. K., Seed, A. M., Tan, J., van 163–173. Schaik, C. P., & Wobber, V. (2012). How does cogni- Itakura, S., Agnetta, B., Hare, B., & Tomasello, M. (1999). tion evolve? Phylogenetic comparative psychology. Chimpanzees use human and conspecifi c social cues Animal Cognition, 15 , 223–238. to locate hidden food. Developmental Science, 4 , Majolo, B., de Bortoli Vizioli, A., & Schino, G. (2008). 448–456. Costs and benefi ts of group living in primates: Group Judge, P. (1982). Redirection of aggression based on kin- size effects on behaviour and demography. Animal ship in a captive group of pigtail macaques. Behaviour, 76 , 1235–1247. International Journal of Primatology, 3 , 301. Mannu, M., & Ottoni, E. B. (2009). The enhanced tool-kit Kaminski, J., Call, J., & Tomasello, M. (2008). of two groups of monkeys in the Caatinga: Tool mak- Chimpanzees know what others know, but not what ing, associative use, and secondary tools. American they believe. Cognition, 109 , 224–234. Journal of Primatology, 3 , 242–251. 44 T. Mandalaywala et al.

Marticorena, D. C. W., Ruiz, A. M., Mukerji, C., Goddu, Passingham, R. E. (1982). The human primate . San A., & Santos, L. R. (2011). Monkeys represent others’ Francisco: W.H. Freeman. knowledge but not their beliefs. Developmental Penn, D. C., & Povinelli, D. J. (2007). On the lack of evi- Science, 14 , 1406–1416. dence that non-human animals possess anything Martin, R. D. (2013). Growing a large brain. In R. D. remotely resembling a ‘theory of mind’. Philosophical Martin (Ed.), How we do it (pp. 121–146). New York: Transactions of the Royal Society of London B: Basic Books. Biological Sciences, 362 , 731–744. Meck, W. H., & Church, R. M. (1983). A mode control Perrett, D. I., & Mistlin, A. J. (1990). Perception of facial model of counting and timing processes. Journal of characteristics by monkeys. In W. C. Stebbins & M. A. Experimental Psychology: Animal Behavior Berkley (Eds.), Comparative perception (pp. 187– Processes, 9 , 320–334. 215). New York: Wiley. Melis, A. P., Call, J., & Tomasello, M. (2006). Perrett, D. I., Hietanen, J. K., Oram, M. W., & Benson, Chimpanzees (Pan troglodytes) conceal visual and P. J. (1992). Organization and functions of cells auditory information from others. Journal of responsive to faces in the temporal cortex. Comparative Psychology, 120 , 154–162. Philosophical Transactions of the Royal Society of Micheletta, J., & Waller, B. M. (2012). Friendship affects London B, 335 , 23–30. gaze following in a tolerant species of macaque, Phillips, W., Barnes, J. L., Mahajan, N., Yamaguchi, M., Macaca nigra. Animal Behaviour, 83 , 459–467. & Santos, L. R. (2009). ‘Unwilling’ versus ‘unable’: Milton, K. (1981). Distribution pattern of tropical plant Capuchin monkeys’ (Cebus apella) understanding of foods as an evolutionary stimulus to primate mental human intentional action. Developmental Science, 12 , development. American Anthropologist, 83 , 534–548. 938–945. Mireille, M., Bouchard, M. A., Granger, L., & Herscovitch, Platt, M. L., Brannon, E. M., Briese, T. A., & French, J. (1976). Piagetian object-permanence in Cebus J. A. (1996). Differences in feeding ecology predict capucinus, Lagothrica ﬂ avicauda and Pan troglodytes. differences in performance between golden lion Animal Behaviour, 24 , 585–588. tamarins (Leontopithecus rosalia) and Wied’s mar- Mulcahy, N. J., & Call, J. (2006). Apes save tools for mosets (Callithrix kuhli) on spatial and visual mem- future use. Science, 312 , 1038–1040. ory tasks. Animal Learning and Behavior, 24 , Nieder, A. (2005). Counting on neurons: The neurobiol- 384–393. ogy of numerical competence. Nature Reviews Plooij, F. X. (1978). Some basics traits of language in wild Neuroscience, 6 , 177–190. chimpanzees? In A. Lock (Ed.), Action, gesture and Olthof, A., Iden, C. M., & Roberts, W. A. (1997). symbol (pp. 111–131). London: Academic. Judgments of ordinality and summation of number Pouydebat, E., Gorce, P., Bels, V., & Coppens, Y. (2006). symbols by squirrel monkeys (Saimiri sciureus). Substrate optimization in nut cracking by capuchin Journal of Experimental Psychology, 23 , 325–339. monkeys (Cebus apella). American Journal of Ottoni, E. B., & Mannu, M. (2001). Semi free-ranging Primatology, 68 , 1017–1024. Tufted capuchins (Cebus apella) spontaneously use Povinelli, D. J. (1987). Monkeys, apes, mirror and minds: tools to crack open nuts. International Journal of The evolution of self-awareness in primates. Human Primatology, 22 , 347–358. Evolution, 2 , 493–509. Palagi, E., & Dapporto, L. (2006). Beyond odor discrimi- Povinelli, D. J., & Eddy, T. J. (1996a). What young chim- nation: Demonstrating individual recognition by scent panzees know about seeing. Monographs of the in Lemur catta. Chemical Senses, 31 , 437–443. Society for Research in Child Development, 61 , 1–152. Parker, S. T., & Gibson, K. R. (1977). Object manipula- Povinelli, D. J., & Eddy, T. J. (1996b). Chimpanzees: tion, tool use and sensorimotor intelligence as feeding Joint visual attention. Psychological Science, 7 , adaptations in cebus monkeys and great apes. Journal 129–135. of Human Evolution, 6 , 623–641. Povinelli, D. J., & Vonk, J. (2004). We don’t need a micro- Parker, S. T., & Gibson, K. R. (1979). A developmental scope to explore the chimpanzee’s mind. Mind and model for the evolution of language and intelligence in Language, 19 , 1–28. early hominids. Behavioral and Brain Sciences, 2 , Povinelli, D. J., Perilloux, H. K., Reaux, J. E., & 367–408. Bierschwale, D. T. (1998). Young and juvenile chim- Parr, L. A., & de Waal, F. B. M. (1999). Visual kin recog- panzees’ (Pan troglodytes) reactions to intentional nition in chimpanzees. Nature, 399 , 647–648. versus accidental and inadvertent actions. Behavioural Parr, L. A., Winslow, J. T., Hopkins, W. D., & de Waal, Processes, 42 , 205–218. F. B. M. (2000). Recognizing facial cues: Individual Premack, D., & Woodruff, G. (1978). Does the chimpan- discrimination by chimpanzees (Pan troglodytes) and zee have a theory of mind? Behavioral and Brain rhesus monkeys (Macaca mulatta). Journal of Sciences, 1 , 515–526. Comparative Psychology, 114 , 47–60. Rajala, A. Z., Reininger, K. R., Lancaster, K. M., & Pascalis, O., & Bachevalier, J. (1998). Face recognition in Populin, L. C. (2010). Rhesus monkeys (Macaca primates: A cross-species study. Behavioural mulatta) do recognize themselves in the mirror: Processes, 43 , 87–96. 3 Intelligence in Nonhuman Primates 45

Implications for the evolution of self-recognition. ies of Geladas (Theropithecus gelada). International PLoS One, 5 , 1–8. Journal of Primatology, 33 , 1054–1068. Reader, S. M., & Laland, K. N. (2002). Social intelli- Spelke, E. S. (1985). Preferential-looking methods as gence, innovation, and enhanced brain size in pri- tools for the study of cognition in infancy. In mates. Proceedings of the National Academy of G. Gottlieb & N. A. Krasnegor (Eds.), Measurement of Sciences, 99 , 4436–4441. audition and vision in the fi rst year of postnatal life: A Reader, S. M., Hager, Y., & Laland, K. N. (2011). The methodological overview (pp. 323–363). Westport: evolution of primate general and cultural intelligence. Ablex. Philosophical Transactions of the Royal Society B, Spelke, E. S. (1990). Principles of object perception. 366 , 1017–1027. Cognitive Science, 14 , 29–56. Rendall, D., Rodman, P. S., & Emond, R. E. (1996). Vocal Suarez, S. D., & Gallup, G. G., Jr. (1981). Self-recognition recognition of individuals and kin in free-ranging rhe- in chimpanzees and orangutans, but not gorillas. sus monkeys. Animal Behaviour, 51 , 1007–1015. Journal of Human Evolution, 10 , 175–188. Santos, L. R. (2004). ‘Core Knowledges’: A dissociation Subiaul, F., Vonk, J., Okamoto-Barth, S., & Barth, between spatiotemporal knowledge and contact- J. (2008). Do chimpanzees learn reputation by obser- mechanics in a non-human primate? Developmental vation? Evidence from direct and indirect experience Science, 7 , 167–174. with generous and selfi sh strangers. Animal Cognition, Santos, L. R., & Hauser, M. D. (1999). How monkeys see 11 , 611–623. the eyes: Cotton-top tamarins’ reaction to changes in Sulkowski, G. M., & Hauser, M. D. (2001). Can rhesus mon- visual attention and action. Animal Cognition, 2 , keys spontaneously subtract? Cognition, 79, 239–262. 131–139. Tennie, C., Call, J., & Tomasello, M. (2006). Push or pull: Santos, L. R., & Hauser, M. D. (2002). A non-human pri- Imitation vs. emulation in great apes and human chil- mate’s understanding of solidity: Dissociations dren. Ethology, 112 , 1159–1169. between seeing and acting. Developmental Science, 5 , Tennie, C., Call, J., & Tomasello, M. (2010). Evidence for F1–F7. emulation in chimpanzees in social settings using the Santos, L. R., Nissen, A. G., & Ferrugia, J. A. (2006). fl oating peanut task. PLoS One, 5 , 1–9. Rhesus monkeys (Macaca mulatta) know what others Tennie, C., Call, J., & Tomasello, M. (2012). Untrained can and cannot hear. Animal Behaviour, 71 , chimpanzees (Pan troglodytes schweinfurthii) fail to 1175–1181. imitate novel actions. PLoS One, 7 , 1–19. Sapolsky, R. M. (1992). Cortisol concentrations and the Tomasello, M., & Call, J. (1997). Primate cognition . social signifi cance of rank instability among wild New York: Oxford University Press. baboons. Psychoneuroendocrinology, 17 , 701–709. Tomasello, M., & Hermann, E. (2010). Ape and human Schino, G., Tiddi, B., & Polizzi Di Sorrentino, E. (2006). cognition: What’s the difference? Current Directions Simultaneous classifi cation by rank and kinship in in Psychological Science, 19 , 3–8. Japanese macaques. Animal Behaviour, 71, 1069–1074. Tomasello, M., Call, J., & Hare, B. (1998). Five primate Schrauf, C., Call, J., Fuwa, K., & Hirata, S. (2012). Do species follow the visual gaze of conspecifi cs. Animal chimpanzees use weight to select hammer tools? PLoS Behaviour, 55 , 1063–1069. One, 7 , 1–12. Tomasello, M., Call, J., & Hare, B. (2003). Chimpanzees Scordato, E. S., & Drea, C. M. (2007). Scents and sensi- understand psychological states—the question is bility: Information content of olfactory signals in the which ones and to what extent. Trends in Cognitive ringtailed lemur (Lemur catta). Animal Behaviour, 73 , Sciences, 7 , 153–156. 301–314. Tomasello, M., Carpenter, M., Call, J., Behne, T., & Moll, Seed, A., & Tomasello, M. (2010). Primate cognition. H. (2005). Understanding and sharing intentions: The Topics in Cognitive Science, 2 , 407–419. origins of cultural cognition. Behavioral and Brain Seyfarth, R. M., & Cheney, D. L. (2000). Social aware- Sciences, 28 , 675–735. ness in monkeys. American Zoologist, 40 , 902–909. Uller, C., Hauser, M., & Carey, S. (2001). Spontaneous Shepherd, S. V., & Platt, M. L. (2008). Spontaneous social representation of number in cotton-top tamarins orienting and gaze following in ringtailed lemurs (Saguinus oedipus). Journal of Comparative (Lemur catta). Animal Cognition, 11 , 13–20. Psychology, 115 , 248–257. Silk, J. B. (1999). Male bonnet macaques use information van de Waal, E., & Whiten, A. (2012). Spontaneous emer- about third-party rank relationships to recruit allies. gence, imitation and spread of alternative foraging Animal Behaviour, 58 , 45–51. techniques among groups of vervet monkeys. PLoS Sliwa, J., Duhamel, J., Pascalis, O., & Wirth, S. (2010). One, 7 , 1–7. Spontaneous voice-face identity matching by rhesus van Schaik, C., Fox, E., & Sitompul, A. (1996). monkeys for familiar conspecifi cs and humans. Manufacture and use of tools in wild Sumatran orang- Proceedings of the National Academy of Sciences, utans. Naturwissenschaften, 83 , 186–188. 108 , 1735–1740. van Schaik, C. P., Isler, K., & Burkart, J. M. (2012). Snyder-Mackler, N., Beehner, J. C., & Bergman, T. J. Explaining brain size variation: From social to cultural (2011). Defi ning higher levels in the multilevel societ- brain. Trends in Cognitive Sciences, 16 , 277–284. 46 T. Mandalaywala et al.

Visalberghi, E., & Fragaszy, D. (1990). Do monkeys ape? Whiten, A. (2013). Humans are not alone in computing In S. T. Parker & K. R. Gibson (Eds.), “Language” how others see the world. Animal Behaviour, 86 , and intelligence in monkeys and apes (pp. 247–273). 213–221. Cambridge, UK: Cambridge University Press. Whiten, A., & Byrne, R. W. (1988). Tactical deception in Visalberghi, E., Addessi, E., Truppa, V., Spagnoletti, N., primates. Behavioral and Brain Sciences, 11 , 233–273. Ottoni, E., Izar, P., & Fragaszy, D. (2009). Selection of Whiten, A., McGuigan, N., Marshall-Pescini, S., & effective stone tools by wild bearded capuchin mon- Hopper, L. M. (2009). Emulation, imitation, over- keys. Current Biology, 19 , 213–217. imitation and the scope of culture for child and chim- Warneken, F., & Tomasello, M. (2006). Altruistic helping panzee. Philosophical Transactions of the Royal in human infants and young chimpanzees. Science, Society B, 364 , 2417–2428. 311 , 1301–1303. Wilson, M., Hauser, M., & Wrangham, R. (2001). Warneken, F., Hare, B., Melis, A. P., Hanus, D., & Does participation in intergroup conﬂ ict depend Tomasello, M. (2007). Spontaneous altruism by chim- on numerical assessment, range location, or rank panzees and young children. PLoS Biology, 5 , e184. for wild chimpanzees? Animal Behaviour, 61 , Watts, D., & Mitani, J. (2001). Boundary patrols and 1203–1216. intergroup encounters in wild chimpanzees. Behaviour, 138 , 299–327. The Evolution of Language 4 Philip Lieberman

Humans are the only living species that possesses …any variation, however slight and from whatever language. It is diffi cult to conceive of any aspect cause proceeding, if it be in any degree profi table to an individual of any species, in its infi nitely of human behavior in which language does not complex relations to other organic beings and to play a role. Social interaction, mitigating or pro- external nature, will tend to the preservation of that moting violence, forming inclusive groups, individual and will generally be inherited by its sharing information, selecting mates, transmit- offspring. (Darwin 1859 , p. 61) ting technology, ethical values all involve com- The genetic basis of variation was unknown in munication through the medium of language. Darwin’s epoch, but he was acutely aware of the And speech, as noted below, has properties that variations that marked the individuals that com- enhance the transmission rate of information. prised a species, whether the species in question Other means can serve to communicate infor- was pigeons or people. The concept of a species mation, but language is the modality by which was to Darwin elastic, since his focus was on the human culture is transmitted and information is “transmutation” of species. Thus, distinction shared. Language also is a modality of thought between varieties and species thus necessarily that may be thought of as a cultural tool. Current was hazy. For example, when discussing the dif- studies point to the neural bases of language and ferent varieties of domesticated pigeons, Darwin cognition sharing common structures and observed that circuits. if shown to an ornithologist, and he were told that they were wild birds, would certainly, I think, be ranked by him as well-defi ned species. (1859 , p. 22) The Darwinian Framework Natural selection, blind to all consequences except the survival of progeny, was the engine Charles Darwin knew that the pace of evolution is that could gradually create a “new” species not even—both gradual changes and abrupt transi- through the aggregation of small variations. tions occur. Natural selection, the principle most However, Darwin also proposed another mecha- often associated with Darwin, generally results in nism to account for abrupt transitions, such as small gradual changes. In Darwin’s words, that from aquatic to terrestrial life. Gradual adaptations by means of natural selection to aquatic life that would produce fi sh that were more suc- * P. Lieberman ( ) cessful at having surviving progeny could not George Hazard Crooker University , 141 Elton Street , Providence , RI 02906-5433 , USA account for the appearance of animals who lived e-mail: [email protected] on land and had to breathe air. Darwin’s solution,

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 47 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_4, © Springer Science+Business Media New York 2015 48 P. Lieberman based on his observation of living species – This point of view contracts with the view lungfi sh, transitional species that could survive in expressed by Noam Chomsky and many other dried up river beds – was the linguists. Chomsky’s central claim for decades (e.g., Chomsky 1972 ) has been that humans pos- …fact that an organ originally constructed for one purpose…may be converted into one for a wholly sess innate neural mechanisms that are species different purpose…. (p. 190) specifi c and specifi c to language alone and that essentially instantiate the details of syntax of all language, past, present, and future. For many years Chomsky posited a “universal grammar” Communication and Cognition (UG) that instantiated hundreds of principles and parameters that were activated when a child came It has often been debated whether language into contact with a language. In The Science of serves primarily as a vehicle of thought or com- Language (Chomsky 2013) , the innate “faculty munication. The fi ndings of the studies that will of language” entails humans possessing only one be briefl y reviewed here point to language serv- innate mental capacity “merge” that accounts for ing both functions. Moreover, although human humans possessing language as well as cognitive cognitive capabilities are in a sense unique, the capabilities such as numeracy and arithmetic neural substrate implicated in human language skills. Merge is defi ned as the capacity for “… and cognition also regulates aspects of behav- taking two things and putting them together or ior that we share to various degrees with other taking one thing and taking a piece of it and stick- species. The biological bases of human lan- ing it at the edge” (Chomsky 2013, p. 16). Merge guage and cognition refl ect aspects of anatomy accounts for what Chomsky takes to be the defi n- and neural mechanisms that can be traced back ing feature of human language—being able to millions of years that initially served other construct and comprehend complex sentences ends, but took on new functions and subse- that include clauses or conjunctions. quently were adapted through the agency of However, Chomsky’s views on the evolution natural selection to enhance communication of human language unfortunately fall outside the and cognition. domain of biology in light of his views on natural It, moreover, is diffi cult to conceive of any selection. In Chomsky’s words, aspect of human behavior in which language It is perfectly safe to attribute this development [of does not play a role. Social interaction, mitigat- innate language structures] to “natural selection,” ing violence, promoting violence, forming so long as we realize that there is no substance to inclusive groups, denigrating other groups, this assertion, that it amounts to no more than a belief that there is some natural explanation for sharing information, selecting mates—all have these phenomena. (Chomsky 1972 , p. 97) been enhanced by the evolution of human linguistic and cognitive capabilities. That being the Discussions of the absence of evidence for case, it is unlikely that any single factor, such as natural selection pervade Chomsky (2013), The social interaction, provided the selective advan- Science of Language . Summing up its view on tage that drove the evolution of human language the evolution of human language, and cognition. you [Chomsky] emphasize in the case of human language—certainly the most distinctive and central mental faculty, one that no other creature has Initial Stages …there is no evidence that a long-term selectional story will work. There are reasons to believe that language was introduced at a single stroke with the The biological endowment that humans bring to introduction of Merge, perhaps some fi fty of sixty bear on the acquisition of language is innate, an thousand years ago. (Chomsky 2013, p. 103) aspect of human genetic endowment. However, children learn to speak, learn the meaning of The date for the sudden appearance of human words, and learn the syntax of their language. language on different pages of Chomsky (2013) 4 The Evolution of Language 49 shifts between 50,000 and 100,000 years in the past, but these dates are all implausible in light of Protolanguage evidence from the fossil record, archaeology, and genetics, which points to modern humans and The language capabilities of extinct hominins earlier extinct hominins such as the Neanderthals (species ancestral or related to present-day possessing language at least 500,000 years ago, humans) have been vigorously debated. Chomsky, with precursors dating back to the common as noted above, believes that humans possessing ancestor of present-day apes and humans six mil- language suddenly appeared 100,000 or lion years ago. 50,000 years ago. Mithin (2005) claimed that the Moreover, “merge” is not a species-specifi c Neanderthals, who became extinct about human attribute. The neural basis for “merge” 40,000 years ago, had a language restricted to clearly is present to some degree in many, per- communicating by humming. Other proposals haps all, living species. If the linguistic terminol- have claimed that early hominins possessed a ogy is stripped away, merge reduces to the “protolanguage” limited to isolated words. These associative learning, the process by which ani- claims can be assessed through comparative stud- mals learn to associate two or more things, acts, ies of ape communication. Present-day apes and and events—and for humans to associate con- humans share a common ancestor. Thus, any cepts. Ivan Pavlov, in the early years of the twen- aspect of language that apes can master in an tieth century, showed that dogs can connect the appropriate environment most likely was present sound of a bell and food. Subsequent studies in early hominin languages. show that species ranging from mollusks to apes The reference to “environment” is essential. A can merge. The mollusks studied by Carew et al. biological capacity may be present in any species (1981) were very slow at learning to associate that is not apparent until it is expressed in a par- electric shocks and a smell pleasant to mollusks, ticular environment. No one in the year 1805, for but they did so after several hundred trials. example, would have believed that humans could The mollusk study demonstrated that the neu- approach each other at closing speeds exceeding ral correlate of associative learning was, as Hebb 120 miles per hour, separated by three feet or less, (1949) had suggested, synaptic modifi cation. and survive while at the same time listening to Synapses transmit information between “neu- music or to someone talking, but that occurs rou- rons,” the basic computing elements of brains. tinely on roads throughout the world all day long. Synapses also code information acquired through When raised from infancy in a language using associative learning as their “weights.” The environment that makes use of manual gestures degree to which they transmit information is or other manual modalities such as pushing but- modifi ed. In the mollusk study a synapse involved tons on a speech synthesizer, chimpanzees can in motor control coded the association between acquire some aspects of human language. When electric shock and smell. American Sign Language (ASL) is used by Humans are capable at learning and executing human caretakers to communicate to another and complex acts, concepts, and language that tran- to infant chimpanzees in a fairly normal house- scend the capabilities of mollusks or any living hold, they can acquire active vocabularies of species. As the studies discussed below demon- about 150 words and master the infl ectional mor- strate, these human capabilities involve genetic phology inherent in ASL words (Gardner and events in the last 500,000 years that enhanced Gardner 1969 ). Formal tests show that chimpan- synaptic connectivity and plasticity. However, zees also can comprehend distinctions in mean- the basic mechanism, synaptic plasticity, is pres- ing conveyed by the syntax of spoken simple ent in mollusks and dates back hundreds of mil- English sentences (Savage-Rumbaugh et al. lions of years. 1985 ). The Gardner chimpanzees also appeared 50 P. Lieberman to comprehend spoken English sentences, but no Alvin Liberman and his colleagues ( 1967 ). formal tests were administered. It thus is improb- A brief account of the physiology of speech pro- able that any hominin species was limited to a duction and some aspects of speech perception is “protolanguage” in which they communicated by necessary to understand the complexity of this uttering isolated, single, words or that process and why the evolution of the species- Neanderthals were limited to humming. However, specifi c human tongue plays a critical role in this no nonhuman species can talk. Since apes instead process. can use sign language and other manual systems The source of energy for speech production is to signify words, this lends plausibility to ges- the airfl ow out of the lungs. This, in itself, neces- tures playing a greater role in the early stages of sitates a person’s executing a set of complex mus- the evolution of language (Hewes 1973 ). cular maneuvers, owing to the evolutionary history of the lungs. Darwin (1859 , p. 160) noted that that the lungs of air-breathing animals Talking evolved from the swim bladders of fi sh. Primitive fi sh, such as sharks, that lack swim bladders have Being able to talk entails being able to learn com- to constantly move to maintain a given depth in plex, coordinated motor acts involving the lungs, the ocean. Swim bladders enabled more larynx, jaw, lips, soft palate, and the intrinsic and “advanced” fi sh to hover by storing air extracted extrinsic muscles of the tongue, and then being from water in elastic swim bladders, which able to rapidly and precisely execute these adjusted their body’s size so as to displace an maneuvers. The learning process for human chil- equal weight of water at a given depth. This dren takes years, contrary to some claims by lin- allows such fi sh to hover while expending less guists that language is in place by age 5. Children energy than fi sh that must constantly move. at age 10 years still are not able to speak at adult Human lungs retain this elastic property; as they rates (Smith 1978). Though communication expand, they increase the volume of a person’s using manual gestures appears to be simpler body. No muscle directly acts on the lung sacks since apes can master a limited number of ASL to expand them. During quiet inspiration the dia- words, formal manual sign languages are an phragm, intercostal, and abdominal muscles invention dating to the eighteenth century and expand the space in which the elastic lungs are could not have played a dominant role in the placed, and the lungs, in turn, expand, storing early stages of hominin language claims for the energy in a manner similar to that of rubber bal- spontaneous invention of complex sign languages loons. During expiration, the elastic recoil of the (e.g., Nicaraguan Sign Language does not hold lung sacs, in a manner analogous to that of dis- up) (Polich 2006). tended balloons, expels air. At the start of an The obvious advantages of talking over man- expiration, where the lungs are at their maximum ual gestures are that it isn’t necessary to look at expansion, the air pressure within the lungs the person communicating, it works in the dark, (alveolar air pressure) is at a maximum and then and it frees the talkers’ hands so that they can be falls as the lungs defl ate. The durations of inspi- used to employ tools, carry infants, carry objects, ration and expiration during quiet breathing are and manipulate devices. In the 1960s it became almost equal. apparent that human speech has a critical linguis- The alveolar air pressure during expiration tic advantage over all other auditory signals—as impinges on the vocal cords of the larynx. If the we talk we can transmit information at rates that vocal cords of the larynx are tensioned and posi- exceed the fusion frequency of the auditory sys- tioned so as to phonate (rapidly opening and tem. Other discrete sound codes, such as Morse closing), a series of “puffs” of air result. The aver- code, transmitted at this rate merge into a buzz. age rate at which these puffs of air occur, the funda- Speech achieves this transmission rate by means mental frequency of phonation (Fo), is perceived as of the process of “encoding,” a term coined by the “pitch” of a speaker’s voice. Fo is determined 4 The Evolution of Language 51 by the magnitude of alveolar air pressure and the During spontaneous speech people usually tension placed on the vocal cords of the larynx. anticipate the length of the sentence that they will Since alveolar air pressure is high at the start of a produce and inspire more air before the start of a quiet expiration, F0 would start high and rapidly long sentence (Lieberman and Lieberman 1973 ). fall when people talk barring the complex com- It is diffi cult to establish when hominins might pensatory maneuvers that must be unconsciously have acquired the ability to control alveolar air carried out when we talk. pressure during speech. Attempts have been The human vocal cords are complex structures made to relate the size of bony channels through whose evolution can be traced back to lungfi sh. A which nerves enervate the diaphragm to alveolar series of adaptations changed their role from control, but the diaphragm does not enter into sealing the lungs from the intrusion of water to speech production. enhancing phonation in different species (Negus In most languages and dialects (there are 1949 ). Animals ranging from frogs to humans exceptions such as “valley-girl” English), the communicate information by means of variations fundamental frequency of phonation and ampli- in F0. In human speech, variations in F0 patterns tude of the speech signal remaining more or less often signal emotion, but they also transmit refer- level during a sentence, abruptly falling at its end. ential, linguistic information in most human lan- In many dialects of English and other languages, guages. The Chinese languages, for example, Fo remains instead level or rises for yes-no ques- differentiate words by means of F0 patterns spe- tions (Armstrong and Ward 1926 ; Pike 1945 ; cifi c to given words—the consonant-vowel Lieberman 1967 ). Frogs signal different states by sequence [ma] in Northern Chinese specifi es four means of calls that have different F0s. In pri- different words, differentiated by four different mates, independent studies such as Cheyney and lexical tones (Tseng 1981 ). Seyforth (1990 ) show that monkeys signal refer- The pattern of alveolar muscular control dur- ential information by means of calls that have dif- ing speech is quite different so as to produce an ferent F0 contours. This again points out the alveolar air pressure contour that is almost level implausibility of any stage in early hominin com- throughout a sentence until its end, unless a munication that exclusively relied on manual speaker wishes to accentuate a syllable gestures. (Lieberman 1967 ). The diaphragm is immobi- Mammalian infants and their mothers main- lized and the duration of expiration is keyed to tain contact and direct attention using calls that the length of the sentence that the speaker intends have varying F0 contours—that holds also for to produces. Alveolar air pressure is maintained humans (Fernald et al. 1989 ). The neural circuits at an almost uniform level by programing a set of that control this aspect of phonation appear to instructions to the intercostal and abdominal have their roots in therapsids, mammal-like rep- muscles that can expand the lungs, so that they tiles who lived in the age of dinosaurs. The ante- “hold back” in a graduated manner, keyed to the rior cingulate gyrus (ACC) of the paleocortex is length of a sentence, against the force generated linked by a neural circuit to the basal ganglia, by the elastic recoil force of the lungs. The alveo- subcortical structures deep within the brain. The lar air pressure resulting from the lungs’ elasticity circuits appear to be similar in monkeys and otherwise always would be high at the start of the humans (Alexander et al. 1998; Lehericy et al. expiration (often blowing the vocal cords apart, 2004 ). The inference that the therapsid brain had ending phonation) and rapidly falls as lung vol- a neural circuit linking an ACC to the basal gan- ume falls. The intercostal and abdominal muscles glia follows from therapsid fossils having the contain muscle “spindles” that can monitor the three middle ear bones that characterize all mam- force that they produce. The diaphragm, which is mals. In true reptiles these bones form a hinge in immobilized during speech and singing, contains the jaw, allowing the jaw to open wide so as to few spindles which may account for its being swallow large creatures. In the course of evolu- immobilized when we talk (Bouhuys 1974 ). tion, the former three-bone jaw hinge migrated 52 P. Lieberman into the mammalian middle ear, serving as a mechanical amplifi er that enhances the ability of The “Gift” of Tongue a mother’s and suckling infant’s maintaining contact. Lesion studies show that when the neural Speech encoding which accounts for the rapid circuit to the ACC is disrupted in a mouse mother, rate at which we can vocally transmit information she will not pay attention to her pups (Maclean derives from the inherent limit on how the shape and Newman 1982). General problems in main- and length of the airway above the larynx can be taining attention occur when neural circuits link- changed (Fig . 4.1 ). ing the ACC to basal neural are degraded by The airway above the larynx, the supralaryn- Parkinson disease in humans (Cummings. 1993 ). geal vocal tract (SVT), determines the sound Maclean and Newman showed that ACC neural quality of consonants and vowels in a manner circuits also control the laryngeal “mammalian analogous to a pipe organ. In a pipe organ a isolation cry.” An example is a human infant cry. source of acoustic energy that has a wide band of In human adults, disrupted basal ganglia circuits frequencies is fi ltered by pipes that allow energy of the ACC can result in mutism (Cummings to pass through them in narrow ranges of fre- 1993) and aberrant patterns of laryngeal control quency, producing particular musical notes. during speech (Lieberman et al. 1992 ; Pickett Phonation, the acoustic energy generated by the et al. 1998 ). larynx, is the “source” of energy for speech Thus, it is most likely that hominins, dating sounds such as the vowels and initial consonants back six million years ago to the common ances- of the words bit and map. The acoustic energy tor of humans and present-day apes, had some generated by the larynx occurs at the fundamen- form of spoken language in which words and, tal frequency of phonation and at its harmonics— most likely, some aspects of syntax were com- integral multiples of Fo. For a given shape and municated by modulating F0. length of the SVT, maximum acoustic energy will

Fig. 4.1 Sketch of the human tongue and supralaryngeal airway 4 The Evolution of Language 53 pass through it at a set of “formant frequencies.” unless a person intones a fi xed vowel. For example, The vowel of the word “see”—in phonetic nota- when producing a word such as bit, the positions tion [i]—could have local energy maxima at 270, of the tongue, jaw, lips, and larynx which deter- 2,300, and 3,000 Hz for an adult speaker whose mine the shape of the SVT and the resulting for- overall SVT length is 17 cm. The same speaker’s mant frequencies of /b/ must move to the different [a] vowel (the vowel of the word “ma”) would be positions necessary for /I/ (the vowel of the word about 700, 1,100, and 2,400 Hz. Consonants and “bit”) and then to /t/. As they move, albeit rapidly, vowels are distinguished by their formant fre- there must be a transition between each SVT quency patterns, as well as their duration, the shape. It became clear that encoded words are the timing between tongue and lip maneuvers, and perceptual units of speech, a consonant- vowel laryngeal phonation—detailed discussions can (CV) syllable being the minimal unit, that listen- be found in many texts including Fant ( 1960 ), ers then perceptually decode, taking account of Nearey (1978 ), Lieberman (1984), and Lieberman the constraints of speech production at some and Blumstein (1988). internal level (Liberman et al. 1967 ). The for- Movable type is often used as an analogy for mant frequency patterns of the phonemes that the discreteness of speech; we hypothetically make up words are always melded together. strung together discrete phonemes (sounds This fact that the phonemes postulated by approximated by the letters of the alphabet) to linguists don’t exist in the speech signal allows form words. The phonemes [t] [a] and [b], for humans to transmit information at a rate that example, could be rearranged to form the words exceeds that of any other acoustic signal by tab, bat, at, and ba. However, discrete phonemes means of a complex perceptual process. The are not present in the acoustic speech signal. In minimal units transmitted are encoded words, at the 1960s it was thought that talking machines a rate below 7 units per second. The perceived could work by simply isolating phonemes from speech signal then can be perceptually decoded tape recordings of carefully enunciated speech into sequences of phonemes. In a sense, pho- and then rearranging and stringing them together. nemes are abstract speech motor control and If phonemes really were beads on a string, there perception instruction sets. Chinese orthogra- should have been a segment of tape that con- phy, which codes words, is a better approxima- tained the sound [t], when a person spoke the tion of the speech signal than alphabetic word too , before a segment of tape that contained systems. Computerized speech recognition sys- the sound [u] (the vowel of the word too ). tems use algorithms that match the incoming However, the talking-machine project was a fail- signal to probable word templates. Dogs also ure; when the segment of the recording tape that can recognize words; some dogs can learn to was supposed to correspond to the “phoneme” [t] recognize hundreds of words with one trial, i.e., excised from the word too was isolated and immediately on hearing the word in association linked to the hypothetical discrete phoneme of with a referent—usually an object (Kaminski the vowel [i] excised from the word tea , the result et al. 2004). was incomprehensible. One of the problems encountered in speech The dynamics of speech production showed recognition systems is how to take into account that a talking-machine project based on isolating the effect of speakers of all ages and sizes, who and linking phonemes never could work. The have different SVT lengths. Since the length of positions assumed by the lips, tongue, jaw, and the SVT varies from person to person and for the larynx for the phoneme /t/ are affected by those same individual as he or she grows up, the abso- necessary to produce the vowel [u] in a different lute values of the formant frequency pattern vary manner than the vowel [i], yielding a different, for the same words. For example, the formant fre- overall, “encoded” formant frequency patterns quencies of an /i/ are 1.5 times higher for a child for the words too and tea. The articulatory ges- whose SVT length is 11.3 cm longer than for his tures that generate speech are always encoded father whose SVT is 17 cm long. Humans face 54 P. Lieberman the same problem, but the two different formant thought that the human tongue in some manner frequency patterns would both be perceived as facilitated speech communication. That supposi- examples of an /i/ owing to a speech-specifi c pro- tion is supported by computer-modeling studies cess of perceptual normalization which humans that calculate the range of formant frequencies that appear to share with other species—dogs can can be produced by both nonhuman and human also do it—probably horses and many other spe- SVTs (e.g., Lieberman et al. 1969 , 1972 ; Lieberman cies. Listeners internally estimate the length of a and Crelin 1971 ; Carre et al. 1995 ; De Boer 2010 ). speaker’s SVT and can estimate SVT length after The initial 1969 study calculated the formant hearing a short stretch of speech or “reverse- frequency patterns of the vowels that the SVT of engineering” a known phrase such as a person a rhesus macaque can produce. The tongue was saying hello , but Nearey (1978 ) showed that the positioned in the computer model of the mon- vowel /i/ (of the word see) was an optimal signal key’s mouth so as to best approximate the SVT for SVT normalization. The data of one of the confi gurations used by adult human speakers to fi rst studies aimed at developing automatic yield the vowels [i], [u], and [a]. These vowels speech recognition (Peterson and Barney 1952 ) delimit the range of vowels used in human lan- had pointed to the vowel /i/ being an optimal cue guages (Greenberg 1963 ). The computed mon- for SVT normalization. Two errors in 10,000 tri- key vowels did not include these vowels. als occurred for words that contained [i] where Newborn infants have SVTs that are similar to listeners had to identify monosyllabic words that those of nonhuman primates (Negus 1949 ; Crelin differed with respect to the vowel and speaker. 1969 ). Similar techniques were used to model the The words uttered by 10 different speakers were SVTs of chimpanzees and human newborn presented in quasi-random order listeners in this infants (Lieberman et al. 1972 ). These fi ndings experiment, and listeners had to immediately were replicated by Carre et al. (1995 ) and De adjust for different speakers’ voices. Hillenbrand Boer (2010 ). The morphology of the skull of the et al. ( 1995) using then state-of-the-art computer La Chapelle-aux- Saints Neanderthal fossil indi- analysis techniques reported similar results. cated that his SVT also was similar to that of a large human newborn (Crelin 1979). Similar computer modeling confi rmed that he had a Swallowing Versus Speaking restricted phonetic range that lacked quantal vowels (Lieberman and Crelin 1971 ). Unfortunately, the human tongue has created a Cineradiographic data of newborn infant cry species-specifi c hazard; compared to other ani- (Truby et al. 1965 ) guided the computer model- mals, we are more likely to choke on solid food ing of jaw, tongue, lip, and laryngeal maneuvers lodged in the larynx. Charles Darwin pointed out in these studies. the problem: At birth in humans, the tongue is largely posi- The strange fact that every particle of food and tioned in the mouth; its shape is fl at as is the case drink which we swallow has to pass over the orifi ce for other primates and most mammals (Negus of the trachea, with some risk of falling into the 1949 ). The proportion of the tongue in the oral lungs …. (1859 , p. 191) “horizontal” (SVTh) part of the infant oral cavity relative to the part of the tongue in the “vertical” Victor Negus’s comparative anatomy showed pharynx (SVTv), SVTh/SVTv, is 1.5, when the that the adult human larynx (and opening to the tra- larynx is positioned at its lowest point in the chea) was carried down into the pharynx because it “forceful” cries pictured in Truby et al. (1965 , “is closely apposed to the tongue” (Negus 1949 , pp. 75–78) which were the basis for the computer pp. 25–26). Choking on food, owing to the low models used by Lieberman and Crelin (1971 ) human larynx, remains the fourth leading cause of and Lieberman et al. (1972 ). It is not until age accidental death in the United States ( http://www. 6–8 years that the human tongue attains its adult nsc.org/library/report_injury_usa.htm). Negus 1:1 SVTh/SVTv proportions and almost circular 4 The Evolution of Language 55 posterior midsagittal shape. The growth process Neural Mechanisms by which the species-specifi c human vocal tract is formed is complex and sometimes takes as The neural basis for vocal tract normalization long as 10 years (Lieberman and McCarthy again appears to refl ect the mechanism fi rst pro- 1999 ; Lieberman et al. 2001). The length of the posed by Charles Darwin—an organ taking on a oral cavity is fi rst shortened by differential bone new role. Other species appear to estimate the growth that moves the hard palate back on the size of conspecifi cs and other species using the base of the skull (Lieberman 2011). The shape absolute values of their vocalizations’ formant and position of the tongue then gradually change frequencies. All other things being equal, larger from the newborn tongue, which is fl at and is animals have longer vocal tracts that produce positioned almost entirely in the oral cavity. The lower formant frequencies. Fitch and Reby human tongue descends down into the pharynx (2001) showed that deer lower their larynges so and achieves its posterior rounded contour, car- as to lengthen their SVTs, thus increasing the rying the larynx down with it. In contrast, the length of their vocal tract and producing lower nonhuman primate tongue throughout life is formant frequencies. The lowered formant fre- long, rectangular, and positioned primarily in the quencies serve to signal to conspecifi cs that an oral cavity. animal is larger than he actually is. However, the During fetal development and shortly after animals’ tongues remain anchored in their birth, the chimpanzee larynx drops slightly owing mouths. The larynx transiently descends by to an increase in the distance between the larynx increasing the distance between the hyoid bone and hyoid, whereas the human growth pattern and larynx. This maneuver does not change the involves the continual descent of the tongue shape of the SVT—its cross-sectional area func- (Nishimura 2005 ). Tongue shape and SVTh/SVTv tions as a function of distance. The cineradio- proportions in nonhuman primates remain almost graphs of other mammals vocalizing in Fitch constant from birth onwards. Darwin’s question (2000 ), contrary to Fitch’s claims, show that concerning why humans have a laryngeal position though transient larynx lowering occurs, the ani- that enhances the propensity to choke thus appears mals cannot produce quantal vowels because to be an adaptation that enhances the robustness of their tongues are still positioned in their mouths. speech communication. The dynamic tongue maneuvers discussed in Stevens ( 1972) independently showed that some detail in Fitch (2010, pp. 315–320) do not only the species-specifi c human SVT can pro- increase the phonetic range of animal vocaliza- duce the ten-to-one midpoint area function dis- tions. The “formant-dispersion” metric used by continuities that are necessary to produce the Fitch (2000 ) to infer vocal tract length from for- vowels [i], [u], and [a], which Stevens termed mants works only because the animal vocaliza- “quantal.” Stevens employed both computer tions always produce a vowel sound close to the modeling and physical models (wooden tubes neutral “schwa” vowel of human speech. Their that could be shifted to change the position of the tongues never generate SVT shapes that deviate 10:1 changes in SVT cross-sectional area). The from a slightly fl ared tube. quantal vowels are perceptually salient owing to Speech communication for language would be the convergence of two formant frequencies possible without the ability to produce quantal which yield spectral peaks. Their formant fre- vowels. Lieberman and Crelin pointed it out in quency patterns also do not shift when tongue their 1971 study which concluded that though position varies about one cm about the midpoint. Neanderthals lacked the capacity to produce Speakers thus can be sloppy and produce the quantal vowels, they could have produced the “same” vowel. They also can produce the vowel range of speech sounds other than quantal vowels [i] which facilitates the process of vocal tract nor- and some consonants. Neanderthals undoubtedly malization. The vowel [u] also has this property talked and had some form of language since their to a lesser degree—and it takes a human tongue stone-working technology, apparent in the and SVT to be able to produce [i] and [u]. archaeological record, could not have been trans- 56 P. Lieberman mitted in the absence of language. In light of the showed that language and speech were disrupted selective advantage speech’s information transfer only when subcortical brain damage was present. rate, Neanderthals undoubtedly talked, albeit Neuroimaging studies using CT scans and MRIs with higher error rates. have resolved the issue. Damage limited to the cortex including Broca’s area, sparing subcortical brain structures, never results in aphasia. The Neural Bases of Language Conversely, aphasia can occur when only subcor- and Cognition tical structures are damaged (e.g., Alexander et al. ( 1987)). The conclusion reached by neu- Studies comparing the vocal repertoires of non- rologists specializing in aphasia is that it never human primates with SVT modeling studies con- occurs, absent subcortical damage (Stuss and sistently show that they are unable to make full Benson 1986 ). use of their phonetic potential. They also could It has become apparent that circuits that link communicate to each other or mimic human activity in different parts of the brain regulate speech, albeit with reduced intelligibility since complex aspects of behavior in both animals and their SVTs would enable them to produce all humans. Converging evidence from the defi cits nonquantal vowels and most consonants. of “experiments in nature,” such as the stroke that However, they cannot talk. Current studies point destroyed parts of Paul Broca’s patients, neuro- to a link between the speech motor control capa- degenerative diseases and other insults to the bilities and the cognitive capacities that distin- brain, and neuroimaging techniques that monitor guish humans from other species deriving from activity in the brains of living subjects, points to neural circuits linking subcortical and cortical a class of circuits linking regions of the cortex structures of the human brain. with the subcortical basal ganglia playing critical roles in motor control, including speech, language, and a range of “higher” cognitive acts. Damage to The Broca-Wernicke Theory the basal ganglia or the pathways to it appears to be the basis for the language defi cits that charac- The traditional answer to the question of why terize aphasia (Lieberman 2000, 2002, 2006). only humans can talk is that Broca’s area of the Tracer studies of the brains of monkeys and cortex instantiates the “faculty of language.” In other animals that cannot be employed in human 1861 Paul Broca published his study of a stroke studies mapped out these circuits linking areas of victim whose speech was limited to a syllable the motor cortex with the subcortical basal gan- that sounded like tan . Broca’s postmortem obser- glia. Other cortical-basal ganglia circuits were vations were limited to the cortical surface of the noted that connected areas of the prefrontal cor- brain. Broca’s patient’s “tan” brain was preserved tex through the basal ganglia and other subcorti- in alcohol. An MRI of the preserved brain shows cal structures to temporal and parietal cortical that the cortical area usually thought to be Broca’s regions of the brain (e.g., Alexander et al. 1986 ). area (the left inferior gyrus) wasn’t damaged—an Noninvasive diffusion tensor imaging (DTI) in area anterior to it instead was damaged. Moreover, human subjects confi rms the presence of similar tan had massive damage to the basal ganglia, cortical to basal ganglia circuits (Lehericy et al. other subcortical structures, pathways connecting 2004 ). Studies of neurodegenerative diseases fi rst cortical and subcortical neural structures, and identifi ed some of the cognitive and linguistic other cortical areas (Dronkers et al. 2007 ). operations performed by these neural circuits. Nonetheless, in the decades that followed These circuits are disrupted in Parkinson disease Broca’s publication, “Broca’s area” has been (PD) owing to the depletion of the neurotransmit- taken to be the brain’s speech and language ter dopamine which degrades basal ganglia oper- “organ.” A few neurologists demurred, pointing ations (Jellinger 1990 ). The basal ganglia act as a out the fact that postmortem examinations sequencing engine, calling out motor control 4 The Evolution of Language 57 information stored in motor cortex (Marsden and revealing some of the roles played by different Obeso 1994). PD patients thus have diffi culty parts of the basal ganglia and cortex areas in sub- linking together and executing the submove- jects performing linguistic and cognitive tasks. ments of internally directed motor acts that are The Wisconsin Card Sorting Test (WCST) is a necessary to walk, talk, or perform manual motor standard instrument for measuring cognitive acts (Harrington and Haaland 1991 ; Speech fl exibility—being able to form and shift from Lieberman et al. 1992 ). As Marsden and Obeso one cognitive criterion to another. The usual pointed out, the basal ganglia in circuits linked to form of the WCST involves subjects sorting the prefrontal cortex constitute an engine that cards that each have images that differ with links cognitive acts and that can shift from one respect to shape or color or number. The fMRI criterion to another as circumstances dictate. data reported in Monchi et al. (2001 ) shows the Cognitive infl exibility and diffi culties occur in activation of a cortical-striatal loop involving the cognitive acts that require planning or selecting ventrolateral prefrontal cortex, the caudate criteria (e.g., Lange et al. 1992 ). nucleus of the basal ganglia, and the thalamus The primary cognitive defi cit of PD is being when subjects shift to the WCST criterion. A dif- unable to change the direction of a thought pro- ferent cortical-striatal loop that includes the pos- cess or action (Flowers and Robertson 1985 ). terior prefrontal cortex and the putamen of the Brain damage limited to the basal ganglia also basal ganglia is active when a sorting criterion can result in similar speech and cognitive defi cits. set shift is executed. The dorsolateral prefrontal Bilateral basal ganglia lesions in the subject stud- cortex was involved whenever subjects made any ied by Pickett et al. ( 1998) produced severe decision as they sorted cards, apparently moni- speech motor defi cits involving sequencing toring whether their responses were consistent laryngeal, lingual, and lung maneuvers. The sub- with the chosen criterion. Similar activation pat- ject had diffi culty comprehending distinctions in terns occurred when subjects were sorting words meaning conveyed by syntax and was unable to instead of images and had to match the words on change the criteria by which she had to sort cards the basis of semantic similarity, the similarity of on the “odd-man-out” test, which Flowers and the beginning of a syllable or rhyme (Simard Robertson (1985 ) devised to test PD patients’ et al. 2011 ). The neural circuits involved thus do cognitive fl exibility. not appear to be domain specifi c to either visual Functional magnetic resonance imaging or linguistic criteria. Studies ranging from (fMRI) which monitors local oxygen depletion recording electrical activity in basal ganglia neu- levels to track the level of neural activity in par- rons of mice and other animals as they learn ticular regions of the brain can show when a par- tasks (Graybiel 1995 ; Mirenowicz and Schultz ticular neural structure is active during a 1996 ; Jin and Costa 2010) to studies of PD cognitive task. Some of the local operations per- patients (Lange et al. 1992 , Monchi et al. 2007) formed in the structures that form human neural and birds (Brainard and Doupe 2002) also show circuits in humans thus can be discerned (e.g., that the basal ganglia in circuits that include cor- Duncan and Owen 2000 ; Petrides 2005 ). Dorsal tical areas play a critical role in associative learn- posterior (upper-back) areas of the motor cortex ing and in planning and executing motor acts control fi ne motor control. Ventrolateral (lower- during speech and, in birds, birdsongs. The start- side) prefrontal cortex is active during tasks that ing point may rest in reptiles which possess the involve actively selecting and retrieving infor- basal ganglia. Leaf and Powell (2011 ) show that mation stored in other regions of the brain. The the tropical arboreal lizard Anolis evermanni dorsolateral (upper-side) prefrontal cortex is exhibits cognitive fl exibility when faced with active while monitoring motor or cognitive problems fetching food. The question, as is the events during a task taking into account earlier case for most aspects of evolutionary biology, is events in working memory. fMRI studies, such the degree of cognitive fl exibility that a species as those by Monchi and his colleagues, are possesses. 58 P. Lieberman

ganglia, as well as through cortical to cortical Fully Human Language circuits. Prefrontal cortical areas appear to pull and Cognition memory traces of images, words, and probably other stored information out of information- The question then arises—why are humans so storing regions of the brain (e.g., Postle 2006 ; much better able to perform complex motor acts Badre and Wagner 2006 ; Miller and Wallis 2009 ). such as talking, command language, and exhibit Some studies using MRIs have claimed that the cognitive fl exibility that is a key element of humans have a disproportionately larger prefron- human creativity? When and how did these tal cortex than chimpanzees and other apes. human capacities evolve? However, as Semendeferi et al. ( 2002 ) point out, Comparative studies on the architecture of the MRIs inherently cannot show that humans have a frontal regions of the brains of monkeys and disproportionately larger prefrontal cortex. It is humans have been conducted over the course of impossible to differentiate prefrontal cortical more than a century (Brodmann 1908 –1912). areas from the motor regions of the frontal cortex Structural cortical differences do not appear to on an MRI; the total human frontal cortex, which account for nonhuman primates being unable to includes prefrontal as well as posterior areas talk or perform at human cognitive levels. involved in motor control, is not disproportion- Petrides ( 2005) in his review of studies of cortical ately larger than an ape’s. architecture concludes that “the basic architec- In short, human brains are bigger, which gives tonic organization (the distribution of neurons in humans the ability to store more information. the layers of the cortex) is the same in humans Memory traces are stored in the temporal cortex and monkeys.” Nor do humans possess the unique and the motor cortex and in parts of the brain that cortical to laryngeal neural circuit proposed by are also associated with processing visual images. Fitch (2012) that hypothetically enables speech Area V1, which is implicated in visual percep- motor activity (Lieberman 2012 ). tion, is activated when people think of an image A great deal of attention has been focused on (Kosslyn et al. 1999). It doesn’t seem to be the humans having big brains because brains require case that any single factor resulted in hominin lots of biological support; thus, they must be brains could have driven this process. Abrupt cli- doing something useful. Current studies show mate changes, alternating periods of glacial cold that human brains have about three times as many and heat, have been invoked as stressors that neurons as a chimpanzee brain (Herculano- resulted in hominin brains becoming larger. Houzel 2009 ). The posterior temporal cortex is However, there is no evidence for African glacial disproportionately larger than would be expected cold cycles. A different story has invoked alter- (Semendeferi et al. 1997 , 2002 ). Temporal cortex nating periods of drought and heavy rainfall is part of the brain’s long-term information stor- causing alternating desertlike or lush rain forests age system which may explain its larger size. in the Rift Valley of Africa drive hominin brain Working memory also may be enhanced by a size enlargement. However, archaic hominins proportionately larger temporal cortex. Current most likely would have moved away when the views on the neural basis of working memory, climate became cold or desertlike as was the case which involves keeping information in short- for other species. Another reoccurring theory term memory during a cognitive process, again posits group size in primates as the causal agent. suggest that it involves cortical to basal ganglia However, this theory cannot account for solitary circuits that access information by means of neu- orangutans having similar sized brains and gre- ral circuits linking prefrontal cortex to temporal garious chimpanzees. Given the common neural cortex and other structures (Badre and Wagner structures that constitute the neural circuits 2006; Postle 2006 ). The human prefrontal cortex known to be implicated in motor control, cogni- is active in a wide range of cognitive acts and is tion, and language, as well as emotional regula- linked to other neural structures by the basal tion, it is improbable that any “one” factor was 4 The Evolution of Language 59 responsible for increasing hominin brain size. sequences. Neanderthals and Denisovans, a Any aspect of behavior that would have led to hominin species that diverged from Neanderthals, more progeny surviving—the true test of success and humans all have this version of FOXP2 in the Darwinian “struggle for existence”—could (Krause et al. 2007 ; Meyer et al. 2012 ). have contributed to the increase in hominin brain Denisovans, who appear to have lived at different size evident in the fossil record. times across Eurasia, from Spain to Siberia, either were ancestral to Neanderthals or split off from Neanderthals about 200,000 years ago (Reich Transcriptional Factors et al. 2010 ). The hominin family tree is becoming very complex, and that question is open. A new avenue of inquiry has opened up that An additional mutation on a location in the focuses on transcriptional factors. Transcriptional gene (intron 8) which regulates the expression of factors essentially are “master” genes that affect FOXP2 in neurons is unique to humans (Maricic the way that other genes are activated to form bod- et al. 2012 ). This mutation occurred about ies and brains. The FOXP2 transcriptional factor 200,000 years ago, the period in which anatomi- came to light when the large, extended KE family cally modern human beings appeared and a in London was studied. Severe speech production, selective sweep occurred which resulted in sentence comprehension, and cognitive defi cits FOXP2human spreading throughout the human occurred in the family members who had only one population (Enard et al. 2002 ). Selective sweeps copy of the FOXP2human transcriptional factor, occur when a mutation provides a selective instead of the normal two (Fisher et al. 1998 ). The advantage that so enhances the survival of indi- genetic code in the double helix of DNA has to be viduals and their offspring that it spreads transcribed into single-stranded mRNA, which throughout a population over the course of rela- then is translated into proteins and the structures of tively few generations. For example, genes that living organisms. Transcription factors govern this conferred adult lactose tolerance, providing an process. The FOXP2 gene is one of many tran- additional food source, spread throughout differ- scription factors. The mouse form of Foxp2 con- ent human groups that herded goats, sheep, or trols the embryonic development of the lungs, the cows (Tishkoff et al. 2007 ). intestinal system, heart, and other muscles, as well Enhanced language, the medium by which as the spinal column (Shu et al. 2001 ). The struc- virtually all aspects of human technology and tures where FOXP2 and Foxp2 are expressed simi- culture are transmitted, along with enhanced cog- lar in the human and mouse brains include the nitive ability would account for the 200,000-year cortical-striatal- cortical circuits involved in motor intron 8 selective sweep. The mechanisms by control and cognition—the thalamus, caudate which transcriptional factors act are complex, nucleus, and putamen, the neural structures that and understanding them is a work in progress, but are anomalous in affl icted members of the KE it is clear that FOXP2 acts to enhance synaptic family (Lai et al. 2003 ). FOXP2 also acts on the plasticity in the cortical-basal ganglia circuits deep layers of the cortex. that are implicated in speech and cognition The focus on the role of FOXP2human (the (Lieberman 2013a , b). When FOXP2 human was superscript signifi es the human version of this knocked into mouse pups, the signifi cant neural gene) follows from its being one of the few genes difference was increased synaptic plasticity in that differ from their chimpanzee version. basal ganglia neurons, as well as increased den- Humans and chimpanzees share about 98 % of dritic lengths in the basal ganglia, thalamus, and their genes. The human version, FOXP2human , layer VI of the cortex (Enard et al. 2009 ; Reimers- evolved during the 6- or 7-million-year period of Kipping et al. 2011 ). The process by which we evolution that separates humans and chimpan- learn anything involves modifying synaptic zees. In that relative period, FOXP2 human under- “weights”—the degree to which synapses trans- went two amino acid substitutions in its DNA mit information to a neuron and code information 60 P. Lieberman

(Hebb 1949). Other “highly accelerated regions” The time line for the evolution of the unique (HARs) of the genome that are unique to humans human tongue suggests an early African origin also appear to be implicated in neural develop- for fully human cognitive and linguistic capabili- ment (Konopka et al. 2009 ; Somel et al. 2013 ). ties. As Darwin noted, the low position of the These genetic studies suggest that neural circuits human larynx, which follows from the migration humans share with other primates were, in effect, of the human tongue down into the neck, creates “supercharged” through the action of transcrip- a choking hazard. A set of acrobatic maneuvers tional genes (Lieberman 2013a, b ). must be carried out to avoid having food falling into and blocking the larynx. The hyoid bone, which supports the larynx, must be pulled for- Dating Fully Human Language ward and upward to move the larynx out of the food path. The epiglottis must simultaneously be The genetic evidence briefl y reviewed here, the fl ipped down to cover the larynx. The long necks archaeological record, and fossil evidence pro- that humans have compared to chimpanzees have vide a time line for the evolution of fully human a purpose. Human necks must be long so as to language. Artifacts, in themselves, cannot serve accommodate the portion of the human tongue in as markers of their maker’s cognitive or linguistic the neck (SVTv) that’s necessary to produce the abilities. Computer word processing involves a quantal vowel [i], plus the larynx attached to the technological base that is far more complex than tongue body. If the human neck were shorter, it sharpening the eighteenth-century quill pens, but would be impossible to swallow because the lar- it does not signify increased human cognitive or ynx would be positioned below it, in the chest literary capabilities. However, there are periods where it would be blocked by the collarbone and extending over millions of years in which the it would be impossible to swallow. Oldowan tools attributed to Homo habilis are vir- Robert McCarthy measured the cervical verte- tually identical. “Hand axes” attributed to Homo brae of the necks of 73 specimens of modern erectus are more complex; it is diffi cult to see humans from populations distributed around the how the technique necessary to make them could globe, as well as a sample of Neanderthals and have been transmitted without some form of lan- early modern humans. The long mouths of guage, but they conform to the same pattern over Neanderthals and earlier extinct hominin species 100,000 years. Neanderthal stone tools are still preclude their having the human tongue propor- more complex and Neanderthals survived in a tions noted above that can produce quantal vow- cold diffi cult environment. Their brains were as els. Mouth lengths can be determined with large as humans and they undoubtedly talked. certainty from boney landmarks on fossil skulls But one element that characterizes all human cul- (Lieberman and McCarthy in press). If the oral tures was missing—there is almost no trace of the cavity is long, the part of the tongue in the neck creative impulse that leads humans to produce must match its length to produce quantal vowels. art—artifacts that are useless. Nor is there evi- Neck lengths can be determined from surviving dence of the behavioral “variability”—the pat- cervical vertebrae. tern of innovation, change, and multiple solutions Tongues that could have produced quantal to a problem that marks human culture (Shea vowels become evident in Upper Paleolithic fossil 2011 ). In contrast, in Africa where modern hominins unearthed in Europe who lived about humans fi rst appeared, a paint “factory” using a 40,000 years ago. Since natural selection can act complex process that took days to bind pigments only on overt behavior, we can be certain that the to mediums was operating 100,000 years ago hominins who had such tongues also had brains (Hensilwood and d’Errico 2011). Stone tools that enabled them to talk, else there would have with inscribed decorative lines that date back been only the negative consequence of an 75,000 years were found in the same cave. Shell increased propensity for choking to death that beads dated back 82,000 years were made in comes from having a human tongue. Thus, we can North Africa (Bouzouggar et al. 2007 ). infer the presence of the brain mechanisms that 4 The Evolution of Language 61 regulate the voluntary, rapid, complex, internally of the cortex with the basal ganglia regulate guided motor acts that underlie human speech. internally specifi ed motor acts including speech Brains and body coevolved to make human and a range of cognitive functions, including the speech possible. But we can infer more than the ability to change the direction of an act or ability to talk because the cortical-basal ganglia thought process—cognitive fl exibility. Synaptic circuits that allow humans to learn and carry out connectivity and malleability in the basal gan- the complex speech motor acts also are part of the glia have been enhanced by a series of mutations human cognitive system. on the FOXP2 transcriptional factor. Selective The stone tools and art of the European Upper sweeps point to these mutations playing an Paleolithic have been interpreted as evidence for important role in the Darwinian “struggle for a “cultural revolution” (Klein 1999 ), which per- existence”—the survival of offspring. Other haps led Chomsky (2013) to claim that language genes unique to humans also appear to have suddenly came into being then. However, there enhanced human cognitive/communicative abil- would have been no selective advantage for the ity. There is no basis for the sudden appearance retention of the peculiar human tongue whose of human language 50,000–100,000 years ago only useful attribute is enhancing the robustness without the agency of natural selection. of speech, unless speech was already in place, probably at human levels 200,000 years ago in Africa. Hominin tongues may have initially References moved down into the neck as skulls restructured, shortening the mouth and palate (Lieberman Alexander, G. E., DeLong, M. R., & Strick, P. L. (1986). 2012), but the neck length of the Skhul V fossil, Parallel organization of segregated circuits linking basal ganglia and cortex. Annual Review of which has a short mouth (Lieberman and Neuroscience, 9 , 357–381. McCarthy 2007 , in press ), was not long enough Alexander, M. P., Naeser, M. A., & Palumbo, C. L. (1987). to support a fully human tongue, pointing to an Correlations of subcortical CT lesion sites and aphasia increase in neck length leading to the evolution of profi les. Brain, 110 , 961–991. Armstrong, L. E., & Ward, I. C. (1926). Handbook of human speech capabilities. Relative neck length English intonation . Leipzig/Berlin: Teubner. also increases in children between birth and age Badre, D., & Wagner, A. D. (2006). Computational and 6–8 years, as their tongues restructure and move neurobiological mechanisms underlying cognitive down into the neck (Mahajan and Bharucha fl exibility. PNAS, 103 , 7186–7190. Bouhuys, A. (1974). Breathing . New York: Grune and 1994). Stratton. Bouzouggar, A., Barton, N., Vanhaeren, M., D’errico, F., et al. (2007). 82,000-year-old shell beads from North Conclusion Africa and implications for the origins of modern human behavior. PNAS, 104 , 9964–9969. Brainard, M. S., & Doupe, A. J. (2000). Interruption of a Some of the biological bases of human language basal-ganglia-forebrain circuit prevents plasticity of have a long evolutionary history that can be learned vocalizations. Nature, 404 , 762–766. traced to the ancestors of living nonhuman spe- Broca, P. (1861). Nouvelle observation d’aphémie produite par une lésion de la moitié postérieure des deuxième et cies. Some of the phonetic elements of human troisième circonvolutions frontales (Bulletin Societé language also are used by other species for differ- Anatomie, 2nd series, Vol. 6, pp. 398–407). ent ends, such as signaling their size. Chimpanzees Brodmann, K. (1908). Beiträge zur histologischen have the capacity to acquire and use words and Lokalisation der Grosshirnrinde. VII. Mitteilung: Die cytoarchitektonische Cortexgleiderung der Halbaffen simple syntax. But human capabilities are quali- (Lemuriden). Journal für Psychologie und Neurologie, tatively superior. 10 , 287–334. The neural bases that govern complex aspects Carre, R., Lindblom, B., & MacNeilage, P. (1995). of behavior are circuits that link activity in dif- Acoustic factors in the evolution of the human vocal tract. C. R (Academie des Sciences Paris, t320(Serie ferent parts of the brain. Circuits linking regions IIb), Vol. t320, pp. 471–476). 62 P. Lieberman

Cheney, D. L., & Seyfarth, R. M. (1990). How monkeys Greenberg, J. (1963). Universals of language . Cambridge: see the world: Inside the mind of another species . MIT Press. Chicago: University of Chicago Press. Harrington, D. L., & Haaland, K. Y. (1991). Sequencing in Chomsky, N. (1972). Language and mind . San Diego: Parkinson’s disease: Abnormalities programming and Harcourt Brace Jovanovich. controlling movement. Brain, 114 , 99–115. Chomsky, N. (2012). The science of language (Interviews Henshilwood, C. S., d’Errico, F., van Niekerk, K. L., with James McGilvray). Cambridge: Cambridge Coquinot, Y., Jacobs, Z., Lauritzen, S. E., Menu, M., University Press. García-Moreno, R., et al. (2011). A 100,000 year old Crelin, E. S. (1969). Anatomy of the newborn; An atlas . ochre processing workshop at Blombos Cave, South Philadelphia: Lea and Febiger. Africa. Science, 334 , 219–221. Cummings, J. L. (1993). Frontal-subcortical circuits and Herculano-Houzel, S. (2009). The human brain in num- human behavior. Archives of Neurology, 50 , 873–880. bers: A linearly scaled-up primate brain. Frontiers in Darwin, C. (1859/1964). On the origin of species . Human Neuroscience, 3 , 1–11. Cambridge, MA: Harvard University Press Hewes, G. W. (1973). Primate communication and the De Boer, B. (2010). Modeling vocal anatomy’s signifi cant gestural origin of language. Current Anthropology, 14 , effect on speech. Journal of Evolutionary Psychology, 5–24. 8 , 351–366. Hillenbrand, J. L., Getty, A., Clark, M. J., & Wheeler, K. Dronkers, N. F., Plaisant, O., Iba-Zizain, M. T., & Cananis, (1995). Acoustic characteristics of American English E. A. (2007). Paul Broca’s historic cases: High resolu- vowels. Journal of the Acoustical Society of America, tion MR imaging of the brains of Leborgne and 97 , 3099–3111. Lelong. Brain, 130 , 1432–1441. Jellinger, K. (1990). New developments in the pathology Duncan, J., & Owen, A. M. (2000). Common regions of of Parkinson’s disease. In M. B. Streifl er, A. D. the human frontal lobe recruited by diverse cognitive Korezyn, J. Melamed, & M. B. H. Youdim (Eds.), demands. TINS, 10 , 475–483. Parkinson’s disease: Anatomy, pathology and therapy Enard, W., Przeworski, M., Fisher, S. E., Lai, C. S. L., (Advances in neurology, Vol. 53, pp. 1–15). New York: Wiebe, V., et al. (2002). Molecular evolution of Raven. FOXP2, a gene involved in speech and language. Jin, X., & Costa, R. M. (2010). Start/stop signals emerge Nature, 41 , 869–872. in nigrostriatal circuits during sequence learning. Enard, W., Gehre, S., Hammerschmidt, K., Hölter, S. M., Nature, 466 , 457–462. Blass, T., Somel, M., et al. (2009). A humanized ver- Klein, R. G. (1999). The human career (2nd ed.). Chicago: sion of Foxp2 affects cortico-basal ganglia circuits in Chicago University. mice. Cell, 137 , 961–971. Konopka, G., Bomar, J., Winden, K., Coppola, G., Fant, G. (1960). Acoustic theory of speech production . Jonsson, Z. O., Gao, F., Peng, S., & Preuss, T. M. The Hague: Mouton. (2009). CNS development genes by FOXP2. Nature, Fernald, A., Taeschner, T., Dunn, J., Papousek, M., de 462 , 213–217. Boysson-Bardies, B., & Fukui, I. (1989). A cross- Krause, J., Lalueza-Fox, C., Orlando, L., Enard, W., language study of prosodic modifi cations in mothers’ Green, R. E., Burbano, H. A., et al. (2007). The and fathers’ speech to preverbal infants. Journal of derived FOXP2 variant of modern humans was shared Child Language, 16 , 477–501. with Neandertals. Current Biology, 17 , 1908–1912. Finlay, B., & Darlington, R. (1995). Linked regularities in Lai, C. S., Gerrelli, D., Monaco, A. P., Fisher, S. E., & the development and evolution of mammalian brains. Copp, A. J. (2003). FOXP2 expression during brain Science, 268 , 1578–1584. development coincides with adult sites of pathology in Fisher, S. E., Vargha-Khadem, F., Watkins, K. E., Monaco, a severe speech and language disorder. Brain, 126 , A. P., & Pembrey, M. E. (1998). Localization of a gene 2455–2462. implicated in a severe speech and language disorder. Lange, K. W., Robbins, T. W., Marsden, C. D., James, M., Nature Genetics, 18 , 168–170. Owen, A., & Paul, G. M. (1992). L-dopa withdrawal Fitch, W. T. (2000). Skull dimensions in relation to body in Parkinson’s disease selectively impairs cognitive size in nonhuman mammals: The causal bases for performance in tests sensitive to frontal lobe dysfunc- acoustic allometry. Zoology, 103 , 40–58. tion. Psychopharmacology, 107 , 394–404. Fitch, W. T. The evolution of language . New York: Leaf, M., & Powell, B. J. (2011). Behavioral fl exibility Cambridge University Press. and problem-solving in a tropical lizard. Biology Flowers, K. A., & Robertson, C. (1985). The effects of Letters, 8 , 28–30. Parkinson’s disease on the ability to maintain a mental Lehericy, S., et al. (2004). Diffusion tensor fi ber tracking set. Journal of Neurology, Neurosurgery, and shows distinct corticostriatal circuits in humans. Psychiatry, 48 , 517–529. Annals of Neurology, 55 , 522–527. Gardner, R. A., & Gardner, B. T. (1969). Teaching sign Liberman, A. M., Cooper, F. S., Shankweiler, D. P., & language to a chimpanzee. Science, 165 , 664–672. Studdert-Kennedy, M. (1967). Perception of the Graybiel, A. M. (1995). Building action repertoires: speech code. Psychological Review, 74 , 431–461. Memory and learning functions of the basal ganglia. Lieberman, P. (1967). Intonation, perception and lan- Current Opinion in Neurobiology, 5 , 733–741. guage . Cambridge, MA: MIT Press. 4 The Evolution of Language 63

Lieberman, D. E. (2011). The evolution of the human Mirenowicz, J., & Schultz, W. (1996). Preferential activa- head . Cambridge, MA: Harvard University Press. tion of midbrain dopamine neurons by appetitive Lieberman, P. (2012). Vocal tract anatomy and the neural rather than aversive stimuli. Nature, 379 , 449–451. bases of talking. Journal of Phonetics, 40 , 608–622. Monchi, O., Petrides, M., Petre, K., Worsley, K. J., & Lieberman, P. (2013a). The unpredictable species: What Dagher, A. (2001). Wisconsin card sorting revisited; makes humans unique . Princeton: Princeton University Distinct neural circuits participating in different stages Press. of the test as evidenced by event-related functional Lieberman, P. (2013b). Synapses, language and being magnetic resonance imaging. The Journal of human. Science, 342 , 944–945. Neuroscience, 21 , 7739–7741. Lieberman, P., & Crelin, E. S. (1971). On the speech of Nearey, T. (1978). Phonetic features for vowels . Neanderthal man. Linguistic Inquiry, 2 , 203–222. Bloomington: Indiana University Linguistics Club. Lieberman, M. R., & Lieberman, P. (1973). Olson’s “pro- Negus, V. E. (1949). The comparative anatomy and physi- jective verse” and the use of breath control as a struc- ology of the larynx . New York: Hafner. tural element. Language and Style, 5 , 287–298. Nishimura, T. (2005). Developmental changes in the Lieberman, D. E., & McCarthy, R. C. (1999). The ontog- shape of the supralaryngeal vocal tract in chimpan- eny of cranial base angulation in humans and chim- zees. American Journal of Physical Anthropology, panzees and its implications for reconstructing 126 , 193–204. pharyngeal dimensions. Journal of Human Evolution, Peterson, G. E., & Barney, H. L. (1952). Control methods 36 , 487–517. used in a study of the vowels. Journal of the Acoustical Lieberman, P., & McCarthy, R. M. (2007). Tracking the Society of America, 24 , 175–184. evolution of language and speech. Expedition, 49 , Petrides, M. (2005). Lateral prefrontal cortex: 15–20. Architectonic and functional organization. Lieberman, P., & McCarthy, R. M. (in press). The evolu- Philosophical Transactions of the Royal Society B: tion of speech and language. In W. Henke & I. Tattersall Biological Sciences, 360 , 781–795. (Eds.), Handbook of paleoanthropology (2nd ed.). Pickett, E. R., Kuniholm, E., Protopapas, A., Friedman, J., Lieberman, P., Klatt, D. H., & Wilson, W. H. (1969). Vocal & Lieberman, P. (1998). Selective speech motor, syntax tract limitations on the vowel repertoires of rhesus and cognitive defi cits associated with bilateral damage monkey and other nonhuman primates. Science, 164 , to the putamen and the head of the caudate nucleus: A 1185–1187. case study. Neuropsychologia, 36 , 173–188. Lieberman, P., Crelin, E. S., & Klatt, D. H. (1972). Pike, K. E. (1945). The intonation of American English . Phonetic ability and related anatomy of the newborn, Ann Arbor: University of Michigan Press. adult human, Neanderthal man, and the chimpanzee. Postle, B. R. (2006). Working memory as an emergent American Anthropologist, 74 , 287–307. property of the mind and brain. Neuroscience, 139 , Lieberman, P., Kako, E. T., Friedman, J., Tajchman, G., 23–38. Feldman, L. S., & Jiminez, E. B. (1992). Speech pro- Reich, D., Green, R. E., Kircher, M., Krause, J., Patterson, duction, syntax comprehension, and cognitive defi cits N., Durand, E. Y., et al. (2010). Genetic history of an in Parkinson’s disease. Brain and Language, 43 , archaic hominin group from Denisova cave in Siberia. 169–189. Nature, 468 , 1053–1060. Lieberman, D. E., McCarthy, R. C., Hiiemae, K. M., & Reimers-Kipping, S., Hevers, S., Paabo, S., & Enard, W. Palmer, J. B. (2001). Ontogeny of postnatal hyoid and (2011). Humanized Foxp2 specifi cally affects cortico- laryngeal descent: Implications for deglutition and basal ganglia circuits. Neuroscience, 175 , 75–84. vocalization. Archives of Oral Biology, 46 , 117–128. Savage-Rumbaugh, S., Rumbaugh, D., & McDonald, K. MacLean, P. D., & Newman, J. D. (1988). Role of midline (1985). Language learning in two species of apes. frontolimbic cortex in the production of the isolation call Neuroscience and Biobehavioral Reviews, 9 , of squirrel monkeys. Brain Research, 450 , 111–123. 653–665. Maricic, T., Gunther, V., Georgiev, O., Gehre, S., et al. Semendeferi, K., Damasio, H., Frank, R., & Van Hoesen, (2012). A recent evolutionary change affects a regula- G. W. (1997). The evolution of the frontal lobes: A volu- tory element in the human FOXP2 gene. Molecular metric analysis based on three-dimensional reconstruc- Biology and Evolution, 30 , 844–852. tions of magnetic resonance scans of human and ape McCarthy, R. C., & Strait, D. S. (2005). Morphological brains. Journal of Human Evolution, 32, 375–378. constraints on hominin speech production. Semendeferi, K., Lu, A., Schenker, N., & Damasio, H. Paleoanthropology, 3 , AO2. (2002). Humans and apes share a large frontal cortex. Meyer, M., et al. (2012). A high coverage genome Nature Neuroscience, 5 , 272–276. sequence from an archaic Denisovan individual. Shea, J. (2011). Homo sapiens is as Homo sapiens was: Science, 338 . doi: 10.1126/science.1224344 . Behavioral variability versus “behavioral modernity”: Miller, E. K., & Wallis, J. D. (2009). Executive function In paleolithic archaeology. Current Anthropology, 52 , and higher order cognition: Defi nition and neural sub- 1–35. strates. In L. Squire (Ed.), Encyclopedia of neurosci- Shu, W., Yang, H., Zhang, L., Ju, M. M., & Morrisey, E. E. ence (Vol. 4, pp. 99–104). Oxford: Academic. (2001). Characterization of a new subfamily of winged- 64 P. Lieberman

helix/forkhead (Fox) genes that are expressed in the Stuss, D. T., & Benson, D. F. (1986). The frontal lobes . lungs and act as transcriptional repressors. Journal of New York: Raven. Biological Chemistry, 276, 27488–277497. Tishkoff, S. A., Reed, F. A., Ranciaro, A., Voight, B. F., Simard, F., Joanette, Y., Petrides, M., Jubault, T., Madjar, et al. (2007). Convergent adaptation of human lactose C., & Monchi, O. (2011). Fronto-striatal contributions persistence in Africa and Europe. Nature Genetics, 39 , to lexical set-shifting. Cerebral Cortex, 21 , 31–39. 1084–1093. Truby, H. L., Bosma, J. F., & Lind, J. (1965). Newborn Somel, M., Liu, X., & Khaitovich, P. (2013). Human brain infant cry . Upsalla: Almquist and Wiksell. evolution; transcripts, metabolites and their regulators. Tseng, C. Y. (1981). An acoustic study of tones in Nature Reviews Neuroscience, 14 , 112–127. Mandarin . PhD dissertation, Brown University. Stevens, K. N. (1972). Quantal nature of speech. In E. E. Tyson, E. (1699). Orang-outang, sive, Homo sylvestris; David Jr. & P. B. Denes (Eds.), Human communica- or, The anatomy of a pygmie compared with that of a tion: A uniﬁ ed view (pp. 51–66). New York: McGraw monkey, an ape, and a man. London: Printed for Hill. Thomas Bennet and Daniel Brown. Evolution of the Human Brain: From Matter to Mind 5

Michel A. Hofman

and thus generate new selection forces that can Introduction modify the structures involved. Mayr ( 1982 , p. 612) even argues that “many if not most acqui- Organisms are faced during their lives with an sitions of new structures in the course of evolu- immense variety of problems, ranging from tion can be ascribed to selection forces by newly purely physical ones, such as changes in climate acquired behaviors.” or geomorphic disturbances, to organism- This suggests that in highly complex organisms, specifi c problems related to food supply, preda- such as primates, behavior rather than environmen- tion, homeostasis, and reproduction. In order to tal change may be the major driving force for evo- enhance their chances of survival, organisms lution at the organismal level. However, this does have to fi nd adequate solutions for the problems not detract from the fact that all organisms, whether with which they are confronted, for any of them they are simple refl ex automata or active and could easily be fatal. Problem solving, in other complex explorers, are above all concerned with words, is an essential dynamic survival mecha- keeping track of their local spatiotemporal environ- nism, evolved to cope with disturbances in the ment, as part of their struggle for existence. Since ecological equilibrium. It can therefore be sensory information processing and the ability to looked upon as an adaptive capacity enabling model reality (or certain parts of it) are essential organisms to adjust themselves to one another components in this process, our idea of problem and to their physical environment (see, e.g., solving seems to correspond reasonably well to the Hodos and Campbell 1990; Macphail and notion of biological intelligence (Hofman 2003 ). In Bolhuis 2001 ; Roth and Wullimann 2001 ; fact, with the evolution of sensory systems as adap- Reader et al. 2011 ; Shettleworth 2012a ). The tations to specialized environments, the capacity to organism’s adaptability, however, is but one process large amounts of sensory information aspect of fi tness. Free- moving organisms, for increased and, with that, the power to create more example, can actively explore their environment complex physical realities. In this chapter, some of the organizational principles and operational modes will be explored that underlie the information-processing capacity * M. A. Hofman , Ph.D. ( ) of the human brain, and it will be argued that the Netherlands Institute for Neuroscience , Meibergdreef 47 , 1105 BA Amsterdam , The Netherlands complexity of the cortical network circuitry is a e-mail: [email protected] measure of intelligence.

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 65 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_5, © Springer Science+Business Media New York 2015 66 M.A. Hofman

clinical and experimental evidence suggesting an Evolution of the Cerebral Cortex important role of the cerebellum in cognitive and affective functions, in close connection with cor- If we assume that biological intelligence in tical associative areas (reviewed by Schmahmann higher organisms is the product of processes of 2010). Although the cerebral cortex is not the complex sensory information processing and only brain structure which was selected for in mental faculties, responsible for the planning, evolution for greater growth, as a result of grow- execution, and evaluation of intelligent behavior, ing environmental pressure for more sophisti- variations among species in intelligence must in cated cognitive abilities, it has played a key role principle be observable in the neural substrate. In in the evolution of intelligence. higher organisms, especially in primates, the complexity of the neural circuitry of the cerebral cortex is considered to be the neural correlate of Scaling of the Primate Cerebral the brain’s coherence and predictive power and, Cortex thus, a measure of intelligence. The evolutionary expansion of the cerebral During the past decades, considerable progress cortex, indeed, is among the most distinctive has been made in explaining the evolution of the morphological features of mammalian brains. cerebral cortex in terms of physical and adaptive Particularly in species with large brains, and most principles (see, e.g., Macphail and Bolhuis 2001 ; notably in great apes and marine mammals, the Hofman 2003 ; Lefebvre et al. 2004 ; Lefebvre brain becomes disproportionately composed of 2012 ; Roth and Dicke 2005 , 2012 ). In addition, a this cortical structure (Welker 1990 ; Nieuwenhuys quantitative approach to the comparative mor- 1994a , b ; Northcutt and Kaas 1995 ; Striedter phology of the brain has made it possible to iden- 2004 ; Hofman and Falk 2012 ; Fig. 5.1 ). The vol- tify and formalize empirical regularities in the ume of cortical gray matter, for example, diversity of brain design, especially in the geom- expressed as a percentage of total brain volume etry of the cortex (e.g., Hofman 1989 , 2012 ; increases from about 25 % for insectivores to Changizi 2001 , 2007 ; Clark et al. 2001 ). 50 % for humans (Frahm et al. 1982 ; Hofman Analysis of the cerebral cortex in anthropoid 1988), whereas the relative size of the entire cere- primates, for example, revealed that the volume bral cortex (including white matter) goes from of the neocortex is highly predictable from abso- 40 % in mice to about 80 % in humans (Hofman lute brain size (Hofman 1989 , 2007 ; Finlay and 1988 , 2012; Azevedo et al. 2009 ; Herculano- Darlington 1995 ; Zhang and Sejnowski 2000 ; Houzel 2009 , 2012 ). Finlay et al. 2001 ; for recent reviews see Hofman On the other hand, the relative size of the cer- and Falk 2012 ). The volume of the cortical gray ebellum remains constant across phylogenetic matter, containing local networks of neurons that groups, occupying about 10–15 % of the total are wired by dendrites and mostly unmyelinated brain mass in different orders (Hofman 1988 ). axons, is basically a linear function of brain vol- Comparative studies among four mammalian ume, whereas the mass of long-range axons, orders, including primates, have recently revealed forming the underlying white matter volume, that the absolute neuronal composition in the cor- increases disproportionately with brain size tex covaries signiﬁ cantly with that of the cerebel- (Fig. 5.2 ). As a result, the volume of gray matter lum (Herculano-Houzel et al. 2008 ; Lent et al. expressed as a percentage of total brain volume is 2012 ), showing that these two brain structures about the same for all anthropoid primates. display coordinated growth during phylogenesis The relative white matter volume, on the other in mammals (for reviews, see Herculano-Houzel hand, increases with brain size, from 9 % in 2012 ; Lent et al. 2012 ). pygmy marmosets ( Cebuella pygmaea) to about Such a coordinated evolution of the cerebral 35 % in humans, the highest value in primates cortex and cerebellum ﬁ ts well with the recent (Hofman 1989 ). The nonlinear nature of this 5 Evolution of the Human Brain: From Matter to Mind 67

Fig. 5.1 Lateral views of the brains of some mammals to of the motor cortex. In man, these anterior and posterior show the evolutionary development of the neocortex association areas are strongly developed. A primary audi- (green ). In the hedgehog, almost the entire neocortex is tory cortex, AS association cortex, Ent entorhinal cortex, I occupied by sensory ( blue ) and motor ( red ) areas. In the insula, M primary motor cortex, PF prefrontal cortex, PM prosimian Galago, the sensory cortical areas are separated premotor cortex, S primary somatosensory cortex, V pri- by an area occupied by association cortex (outlined in yel- mary visual cortex (Reproduced with permission from low). A second area of association cortex is found in front Nieuwenhuys 1994b )

process is further emphasized by plotting the relative to brain volume for a hypothetical pri- relative volume of white matter as a function of mate. The model, for example, predicts a white brain size (Fig. 5.3 ). The high correlation between matter volume of about 1,470 cm3 for an anthro- both variables ensures that the curve, and its con- poid primate with a brain volume of 3,000 cm 3 ﬁ dence limits, can be used for predictive pur- (Hofman 2001b , 2012 ). In other words, in such a poses to estimate the volume of white matter large- brained primate, white matter would 68 M.A. Hofman

changes in the frontal (but not non-frontal) white matter volume are associated with changes in other parts of the brain, including the basal ganglia, a group of subcortical nuclei functionally linked to executive control (Smaers et al. 2010 ). These comparative analyses indicate that the evolutionary process of neocorticalization in primates is mainly due to the progressive expansion of the axonal mass that implement global communication, rather than to the increase in the number of cortical neurons and the importance of high neural connectivity in the evolution of brain size in anthropoid primates. Fig. 5.2 Volumes of cerebral gray and white matter as a Wen and Chklovskii ( 2005 ) have shown that function of brain volume in anthropoid primates, includ- the competing requirements for high connectivity ing humans. Logarithmic scale. The slopes of the regres- and short conduction delay may lead naturally to sion lines are 0.985 ± 0.009 ( gray matter ) and 1.241 ± 0.020 the observed architecture of the mammalian neo- ( white matter ). Note the difference in the rate of change between gray matter (“neural elements”) and white matter cortex. Obviously, the brain functionally beneﬁ ts (“neural connections”) as brain size increases (Reproduced from high synaptic connectivity and short con- with permission from Hofman 2001b ) duction delays. A magnetic resonance imaging study, furthermore, focusing speciﬁ cally on the prefrontal cortex, has shown that the volume of the white matter underlying prefrontal areas is disproportionately larger in humans than in other primates (Schoenemann et al. 2005 ). It suggests that the connectional elaboration of the prefrontal cortex, which mediates such important behavioral domains as planning, aspects of language, attention, and social and temporal information processing, has played a key role in human brain evolution.

Design Principles of Neural Fig. 5.3 Relative white matter volume as a function of Organization brain volume in anthropoid primates. Semilogarithmic scale. The proportion of white matter increases with brain Evolutionary changes in the cerebral cortex have size, from 22 % in a monkey brain of 100 cm 3 to about 65 % in a hypothetical primate with a brain size of occurred mainly parallel to the cortical surface 10,000 cm 3 (Modifi ed with permission from Hofman (tangentially) and have been sharply constrained 2001b ) in the vertical (radial) dimension, which makes it especially well suited for the elaboration of mul- comprise about half of the entire brain volume, tiple projections and mapping systems. A mosaic compared to one-third in modern man. of functionally specialized areas has indeed been Volumetric measurements of gray and white found in the mammalian cortex, some of the matter in the neocortex of anthropoid primates functions being remarkably diverse (Kaas 1993 , have shown that the “universal scaling law” of 2008 , 2012 ; Krubitzer 1995 , 2007 ; Schoenemann neocortical gray to white matter applies sepa- 2006 ). At the lower processing levels of the cor- rately for frontal and non-frontal lobes and that tex, these maps bear a fairly simple topographical 5 Evolution of the Human Brain: From Matter to Mind 69 relationship to the world, but in higher areas, become evident that these cortical circuits inte- precise topography is sacrifi ced for the mapping grate at higher levels of information processing, of more abstract functions. Here, selected aspects as a result of the hierarchical organization of the of the sensory input are combined in ways that brain, thus enabling the system to combine dis- are likely to be relevant to the animal. similar views of the world. It implies that if we Using modern anatomical tracing methods, seek the neural basis of biological intelligence, physiological recordings, and mapping studies, it including mind-like properties and conscious- has been established that each sensory modality ness, we can hardly localize it in a specifi c region is mapped several times in different areas, with of the brain, but must suppose it to involve all about a dozen representations of the visual world those regions through whose activity an organism and half a dozen each of auditory inputs and is able to construct an adequate model of its somatosensory sensations. In fact, the maps dif- external world, perhaps it may even encompass fer in the attributes of the stimulus represented, in the entire neo- and subcortical network. how the fi eld is emphasized, and in the types of It is evident that these neocortical columns are computations performed. Clearly, the specifi ca- functional and morphological units whose archi- tions of all these representations mean that func- tecture may have been under selective evolution- tional maps can no longer be considered simply ary pressure in different mammalian lineages in as hard-wired neural networks. They are much response to encephalization and specializations more fl exible than previously thought and are of cognitive abilities. We are beginning to under- continually modifi ed by feedback and lateral stand some of the geometric, biophysical, and interactions. These dynamic changes in maps, energy constraints that have governed the evolu- which seem likely to result from local interaction of these neural networks (e.g., Felleman and tions and modulations in the cortical circuits, Van Essen 1991 ; Chklovskii et al. 2002 ; 2004 ; provide the plasticity necessary for adaptive Klyachko and Stevens 2003 ; Laughlin and behavior and learning. Although species vary in Sejnowski 2003 ; Rockland 2010 ; Casanova et al. the number of cortical areas they possess and in 2011 ). To operate effi ciently within these con- the patterns of connections within and between straints, nature has optimized the structure and areas, the structural organization of the primate function of these processing units with design neocortex is remarkably similar. principles similar to those used in electronic The tremendous increase in the cortical sur- devices and communication networks. In fact, the face without a comparable increase in its thick- basic structural uniformity of the cerebral cortex ness during mammalian evolution has been suggests that there are general architectural prin- explained in the context of the radial-unit hypoth- ciples governing its growth and evolutionary esis of cortical development (for reviews, see development (Cherniak 1995 , 2012 ; Hofman Rakic 2007 , 2009). According to this model, neo- 1996 , 2001a , 2007 ; Rakic 2009 ; Bullmore and cortical expansion is the result of changes in pro- Sporns 2012 ). liferation kinetics that increase the number of Comparative studies furthermore indicate that radial columnar units without changing the num- variability in subtle subcomponents of the colum- ber of neurons within each unit signifi cantly. nar organization in human and nonhuman pri- Therefore, the evolutionary expansion of the neo- mates, such as the composition of the interneuron cortex in primates is mainly the result of an subtypes, are a primary source of interspecifi c increase in the number of radial columns. differences in minicolumn morphology among The widespread occurrence of these neocorti- species (Raghanti et al. 2010 ). Humans deviate cal columns, furthermore, qualifi es them to be from other primates in having a greater width of considered as fundamental building blocks in minicolumns in specifi c cortical areas, especially neural evolution (for reviews see Mountcastle in the prefrontal cortex, owing to constituents of 1997 ; Buxhoeveden and Casanova 2002b ; the peripheral neuropil space (Buxhoeveden Rockland 2010 ; Buxhoeveden 2012). It has and Casanova 2002a ; Semendeferi et al. 2011 ). 70 M.A. Hofman

These fi ndings support the idea (Semendeferi brain is such that it may perform a great number et al. 2002 ; Allen 2009 ; Teffer and Semendeferi of complex functions with a minimum expendi- 2012 ) that human evolution, after the split from ture of energy and material both in the perfor- the common ancestor with chimpanzees, was mance of the functions and in the construction of accompanied by discrete modifi cations in local the system. In general, there will be a number of circuitry and interconnectivity of selected parts adequate designs for an object, which, for practi- of the brain. The differences in columnar diame- cal purposes, will all be equivalent. ter among primates, however, are only minor Recently, we have shown that in species with compared to the dramatic variation in overall cor- convoluted brains, the fraction of mass devoted to tex size. Thus, it seems that the main cortical wiring seems to increase much slower than that change during evolution has presumably been an needed to maintain a high degree of connectivity increase in the number rather than the size of between the neural networks (Hofman 2003 , these neural circuits. 2007 ). These fi ndings are in line with a model of neuronal connectivity (Deacon 1990 ; Ringo 1991) which says that as brain size increases, Neural Network Wiring there must be a corresponding fall in the fraction of neurons with which any neuron communicates Although the details of the interpretation of the directly. The reason for this is that if a fi xed per- columnar organization of the neocortex are still centage of interconnections is to be maintained in controversial (for recent reviews, see Da Costa the face of increased neuron number, then a large and Martin 2010 ; Rockland 2010 ), it is evident fraction of any brain size increase would be spent that the potential for brain evolution results not maintaining such degree of wiring, while the from the unorganized aggregation of neurons but increasing axon length would reduce neural com- from cooperative association by the self-similar putational speed (Ringo et al. 1994 ). The human compartmentalization and hierarchical organiza- brain, for example, has an estimated interconnection of neural circuits and the invention of fractal tivity of the order of 10 3 , based on data about the folding, which reduces the interconnective axo- number of modular units and myelinated nerve nal distances. fi bers (Hofman 2012 ). This implies that each cor- Recent network studies, using diffusion tensor tical module is connected to a thousand other imaging (DTI), have demonstrated that the neu- modules and that the mean number of processing rons in the neocortex are structurally and func- steps, or synapses, in the path interconnecting tionally highly organized and that this also holds these modules is about two. for the wiring of the brain (Van den Heuvel and Herculano-Houzel et al. (2010 ) have shown Sporns 2011 ; Wedeen et al. 2012 ). The intercon- that in primates the mass of the white matter necting white matter axonal pathways are not a scales linearly across species with its number of mass of tangled wires, as thought for a long time, nonneuronal cells, which is expected to be pro- but they form a rectilinear three-dimensional grid portional to the total length of myelinated axons continuous with the three principal axes of devel- in the white matter. Decreased connectivity in the opment. The topology of the brain’s long-range brain is compatible with previous suggestions communication network looks like a 3-D chess- that neurons in the cerebral cortex are connected board with a number of highly connected neocor- as a small-world network and should slow down tical and subcortical hub regions. the increase in global conduction delay in corti- The competing requirements for high connec- ces with larger numbers of neurons (Sporns et al. tivity and short conduction delay may lead natu- 2004 , 2007 ; Wang et al. 2008 ; Fig. 5.4 ). rally to the observed architecture of the human Once the brain has grown to a point where the neocortex. Obviously, the brain functionally ben- bulk of its mass is in the form of connections, then efi ts from high synaptic connectivity and short further increases (as long as the same ratio in inter- conduction delays. The design of the primate connectivity is maintained) will be unproductive. 5 Evolution of the Human Brain: From Matter to Mind 71

Fig. 5.4 Organizational principles of random, small- cantly higher degrees of local clustering. Functional corti- world, and scale-free networks. Structural cortical net- cal networks, on the other hand, exhibit both scale-free works are neither completely connected with each other attributes with power law degree distributions as well as nor randomly linked; instead, their connections have small-world attributes (Modiﬁ ed with permission from small-world attributes with path lengths that are close to Sporns et al. 2004 ) those of equivalent random networks but with signiﬁ -

Increases in number of units will be balanced by more adequately it will be able to respond to decreased performance of those units due to the environmental challenges and the better will be increased conduction time. This implies that large its chances of survival (Hofman 2003 ). The limit brains may tend to show more specialization in to any intelligent system therefore lies in its abili- order to maintain processing capacity. Indeed, an ties to process and integrate large amounts of increase in the number of distinct cortical areas sensory information and to compare these signals with increasing brain size has been reported with as many memory states as possible and all (Welker 1990 ; Kaas 2000 , 2012 ; Striedter 2004 ). It that in a minimum of time. It implies that the may even explain why large-brained species may functional capacity of a neuronal structure is develop some degree of brain lateralization as a inherently limited by its neural architecture and direct consequence of size. If there is evolutionary signal processing time (see, e.g., Hofman 2001a ; pressure on certain functions that require a high Laughlin and Sejnowski 2003 ; Changizi and degree of local processing and sequential control, Shimojo 2005 ). such as linguistic communication in human brains, The processing or transfer of information these will have a strong tendency to develop in one across cortical regions, rather than within regions, hemisphere (Ringo et al. 1994 ; Aboitiz et al. in large-brained primates can only be achieved by 2003 ). reducing the length and number of the interconnective axons in order to set limits to the axonal mass (Fig. 5.5 ). The number of interconnective Biological Limits to Information ﬁ bers can be reduced, as we have seen, by com- Processing partmentalization of neurons into modular circuits in which each module, containing a large The primate brain, as we have seen, has evolved number of neurons, is connected to its neural from a set of underlying structures that constrain environment by a small number of axons. The its size and the amount of information it can store length of the interconnective ﬁ bers can be reduced and process. If the ability of an organism to by folding the cortical surface and thus shortening process information about its environment is a the radial and tangential distances between brain driving force behind evolution, then the more regions. Local wiring—preferential connectivity information a system, such as the brain, receives between nearby areas of the cortex—is a simple and the faster it can process this information, the strategy that helps keep cortical connections short. 72 M.A. Hofman

the brain, suggesting that in select sets of communicating fi bers, large brains reduce transmission delays and metabolic fi ring costs at the expense of increased volume. Delays and potential imprecision in cross-brain conduction times are especially great in unmyelinated axons, which may transmit information via fi ring rate rather than precise spike timing. In the neocortex, axon size distributions can account for the scaling of per-volume metabolic rate and suggest a maximum supportable fi ring rate, averaged across all axons, of 7 ± 2 Hz. Clearly, the white matter architecture must follow a limited energy budget to Fig. 5.5 The number of connections (C ), cortical pro- optimize both volume and conduction time. cessing units ( U ), and level of interconnectivity ( I ) in the Another way to keep the aggregate length of primate neocortex as a function of brain size. Semilogarithmic scale. Values are normalized to one at a axonal and dendritic wiring low, and with that the brain volume of 100 cm 3 , the size of a monkey brain. Note conduction time and metabolic costs, is to that the number of myelinated axons increases much increase the degree of cortical folding. A major faster than the number of cortical processing units (see disadvantage of this evolutionary strategy, how- also Fig. 5.3). The human cerebrum, for example, contains six times more myelinated axons than that of a rhe- ever, is that an increase in the relative number of sus monkey, whereas the number of cortical processing gyri can only be achieved by reducing the gyral units is only three times larger. Dashed lines show the width. At the limit, the neurons in the gyri would potential evolutionary pathway of these neural network be isolated from the remainder of the nervous elements in primates with very large brains, that is, beyond the hypothetical upper limit of the brain’s process- system, since there would no longer be any opening power (see text and Fig. 5.6 ). Note that a further expo- ing for direct contact with the underlying white nential growth in the number of cortical processing units, matter. Prothero and Sundsten ( 1984 ) therefore without an increase in the number of connections, will introduced the concept of the gyral “window,” lead to a decrease in connectivity between these units and thus to more local wiring (Reprinted with permission which represents the hypothetical plane between from Hofman 2012 ) a gyrus and the underlying white matter through which nerve fi bers running to and from the gyral folds must pass. According to this hypothesis, In principle, effi cient cortical folding could fur- there would be a brain size where the gyral “win- ther reduce connection length, in turn reducing dow” area has an absolute maximum. A further white matter volume and conduction times increase in the size of the brain beyond that point, (Young 1993 ; Scannell et al. 1995 ; Chklovskii that is, at 2,800 cm3 , would increase the cortical et al. 2004 ). Thus, the development of the cortex surface area, but the “window” would decrease, does seem to coordinate folding with connectivity leading to a lower degree of neuronal integration in a way that could produce smaller and faster and an increase in response time. brains. The remarkably high correlation between gray Recently, Wang et al. (2008 ) have shown that matter, white matter, and brain size in anthropoid there are functional trade-offs in white matter primates ensures that the proposed model can be axonal scaling in mammals. They found that the used for predictive purposes to estimate the vol- composition of white matter shifts from compact, ume of white matter relative to brain volume for a slow-conducting, and energetically expensive hypothetical primate (Hofman 2001b ). Model unmyelinated axons to large, fast-conducting, and studies of the growth of the neocortex at different energetically inexpensive myelinated axons. The brain sizes, using a conservative scenario, revealed fastest axons have conduction times of 1–5 ms that with a brain size of about 3,500 cm3 , the total across the neocortex and <1 ms from the eye to volume of the subcortical areas (i.e., cerebellum, 5 Evolution of the Human Brain: From Matter to Mind 73

umes as inferred from fossil cranial endocasts (Klein 2009). It shows that ancestral primates living between 35 and 20 million years ago— arboricole and quadruped—did not have more than 20 billion neurons in their brains. In the Pliocene period, between 5.3 and 2.5 million years ago, neuronal numbers may have increased to about 40 billion in Australopithecus, just above the estimated 30 billion neurons of chimpanzees. These hominins became bipedal and produced the fi rst fl aked stone tools. Another increase took place in the early Pleistocene, about 2.5 million years ago, with the appearance of the genus Homo . Fig. 5.6 Relative subcortical volume as a function of brain The number of neurons in the brain grew to about volume. The predicted subcortical volume (i.e., brain volume—predicted neocortex volume) must be zero at zero 50 billion in Homo habilis, reaching about 70 bil- brain size. Likewise, the subcortical volume will be zero lion in Homo erectus , and fi nally about 90 billion when the brain is exclusively composed of cortical gray in modern man. With such a large number of neu- 3 and white matter. At a brain size of 3,575 cm , the subcorti- rons, bipedal locomotion consolidated, and hand- cal volume has a maximum (see also Fig. 5.5 ). The maximum simulated value for the subcortical volume (366 cm 3 ) fi nger movements acquired sophisticated abilities, is then taken as 100 %. The larger the brain grows beyond which allowed Homo to produce more and more this critical size, the less effi cient it will become. Assuming elaborate tools, dominate fi re, and improve social constant design, it follows that this model predicts an upper interactions. It means that over the past 2–2.5 mil- limit to the brain’s processing power. Modern humans are at about two-thirds of that maximum (Modifi ed with permis- lion years, more than a doubling in the number of sion from Hofman 2001b ) neurons has taken place, leading to one of the most complex and effi cient structures in the ani- mated universe, the human brain. brain stem, diencephalon, etc.) reaches a maximum In view of the central importance placed on value (Fig. 5.6 ). Increasing the size of the brain brain evolution in explaining the success of our beyond that point, following the same design species, one may wonder whether there are phys- principle, would lead to a further increase in the ical limits that constrain its processing power and size of the neocortex, but to a reduction of the evolutionary potential. The human brain has subcortical volume. Consequently, primates with evolved from a set of underlying structures that very large brains (e.g., over 5 kg) may have a constrain its size and the amount of information it declining capability for neuronal integration can store and process. In fact, there are a number despite their larger number of cortical neurons. of related factors that interact to limit brain size, factors that can be divided into two categories: (1) energetic constraints and (2) neural process- Limits to Human Brain Evolution ing constraints (see, e.g., Wang et al. 2008 ; Herculano-Houzel 2009 ). A progressive enlargement of the hominid brain started by about 2–2.5 million years ago, probably from a bipedal australopithecine form with a brain Energetic Limits size comparable to that of a modern chimpanzee (see, e.g., Falk 2004 , 2007 , 2012; Robson and The human brain generates about 15 watts (W) Wood 2008 ; De Sousa and Cunha 2012 ). The lin- in a well-insulated cavity of about 1,500 cm3 . ear scaling law determined for primates allowed From an engineering point of view, the removal Lent et al. (2012 ) to estimate the number of neu- of suffi cient heat to prevent thermal overload rons in the brains of hominins, using brain vol- could be a signifi cant problem. But the brain is 74 M.A. Hofman actively cooled by blood and not simply by heat to primate brain size (Hofman 2001b ; Fig. 5.6 ). conduction from the surface of the head. So the Without a radical change in the macroscopic limiting factor is how fast the heat can be organization of the brain, however, this hypothet- removed from the brain by blood fl ow. It has ical limit will never be approached, since at that been suggested by Falk (1990 ) and others that point (ca. 8,750 cm 3 ), the brain would consist the evolution of a “cranial radiator” in homi- entirely of cortical neurons and their interconnec- nids helped provide additional cooling to deli- tions, leaving no space for any other brain cate and metabolically expensive parts of the structure. brain, such as the cerebral cortex. This vascular Cochrane and his colleagues (1995 ) looked at cooling mechanism would have served as a the different ways in which the brain could evolve “prime releaser” that permitted the brain size to to process more information or work more effi - increase dramatically during human evolution. ciently. They argue that the human brain has So to increase cooling effi ciency in a larger (almost) reached the limits of information pro- brain, either the blood must be cooler when it cessing that a neuron-based system allows and fi rst enters the structure or the fl ow rate must be that our evolutionary potential is constrained by increased above current levels. the delicate balance maintained between conduc- Another factor related to blood fl ow has to tion speed, pulse width, synaptic processing time, do with the increasing energy requirements of a and neuron density. By modeling the information- larger brain, a problem that is exacerbated by processing capability per unit time of a human- the high metabolic cost of this organ. It is type brain as a function of interconnectivity unlikely, however, that the rate of blood fl ow or and axonal conduction speed, they found that the increasing volume used by the blood vessels the human brain lies about 20–30 % below the in the brain—in humans about 4 %—constrain optimal, with the optimal processing ability its potential size. A bigger brain is metaboli- corresponding to a brain about twice the cur- cally possible because our cardiovascular sys- rent volume. Any further enhancement of tem could evolve to transport more blood at human brainpower would require a simultaneous greater pressure to meet the increased demand. improvement of neural organization, signal pro- This should not be taken to imply that thermal cessing, and thermodynamics. Such a scenario, and metabolic mechanisms play no role at all in however, is an unrealistic biological option and setting limits to brain size. Ultimately, ener- must be discarded because of the trade-off that getic considerations will dictate and restrict the exists between these factors. size of any neuron-based system, but as theo- Of course, extrapolations based on brain mod- retical analyses indicate, thermal and metabolic els, such as the ones used in the present study, factors alone are unlikely to constrain the implicitly assume a continuation of brain devel- potential size of our brain until it has increased opments that are on a par with growth rates in the to at least ten times its present size (Cochrane past. One cannot exclude the possibility of new et al. 1995 ). structures evolving in the brain, or a higher degree of specialization of existing brain areas, but within the limits of the existing “Bauplan,” Neural Processing Limits there does not seem to be an incremental improvement path available to the human brain. At a The limit to any neural system lies in its ability to brain size of about 3,500 cm3 , corresponding to a process and integrate large amounts of informa- brain volume two to three times that of modern tion in a minimum of time, and therefore, its man, the brain seems to reach its maximum pro- functional capacity is inherently limited by its cessing capacity. The larger the brain grows neural architecture and signal processing time. beyond this critical size, the less effi cient it will The scaling model of the geometry of the neocor- become, thus limiting any improvement in cogni- tex, for example, predicts an absolute upper limit tive power. 5 Evolution of the Human Brain: From Matter to Mind 75

The conscious representation of the world is Neural Correlates of Consciousness likely to be widely distributed over many areas of the cerebral cortex and possibly over certain sub- Consciousness and affective experience may cortical structures as well (Baars 1997 ). Crick have arisen concurrently in the evolution of the and Koch (1998 ) postulated that only some types nervous system, as a way to elaborate and extend of specifi c neurons will express the neural the potential reach of instinctual urges, while new correlate(s) of consciousness and that these neu- levels of cortical information processing and cog- rons will probably be fairly close together and nition promoted the ability of organisms to effi - will all project roughly to the same place. An ciently pursue goals essential to survival. In fact, alternative hypothesis is that the neural correlate affective experience, being an intrinsic brain of consciousness is necessarily global (Greenfi eld function, cannot exist independent of conscious- 1995 ). In its most extreme form, this would mean ness, since in essence it is something that exists that at one time or another, any neuron in the cor- as part and parcel of conscious perception tex and associated structures could be part of the (Zeman 2001 ; Shettleworth 2012a ). neural correlate of consciousness. In approaching the problem of consciousness, The neural correlate of consciousness is Crick and Koch made the tentative assumption defi ned as the minimal set of neuronal events that that all the different aspects of consciousness gives rise to a specifi c aspect of a conscious per- employ a basic mechanism or perhaps a few such cept (Crick and Koch 2003 ). The cerebral cortex mechanisms (Crick and Koch 1990 , 1998 ). In the is probably the most suited part of the brain to case of visual consciousness, for example, they look for this neural substrate, as it has very highly have suggested that its biological usefulness in and specifi cally interconnected neuronal net- humans is to produce a single but complex works, many types of excitatory and inhibitory interpretation of the visual scene in the light of interneurons, and acts by forming transient coali- past experience, either of ourselves or of our tions of neurons, that is, assemblies of nerve ancestors (embodied in our genes), and to make cells, the members of which support one another. this interpretation directly available, for a suffi - The dynamics of coalitions are not simple, as cient time, to parts of the brain that make a choice Crick and Koch ( 1990 , 2003) have pointed out. In among many different but possible plans of general, at any moment the winning coalition is action (Crick and Koch 1995 ). Exactly how this somewhat sustained and embodies what an ani- works in detail is unclear. mal is conscious of. On the basis of experimental To be aware of an object or event, Crick and results in the macaque, Desimone and Duncan Koch (1995 ) have argued that the brain has to con- (1995 ) suggest that selective attention biases the struct a multilevel, explicit, symbolic interpretation competition among competing cell assemblies, of parts of the visual scene. It means that there are but they do not explicitly relate this idea to specifi c groups of neurons at all levels of the visual consciousness. hierarchy which employ coarse coding to represent Coalitions can vary both in size and in charac- some aspect of the visual scene. In the case of a ter. For example, a coalition produced by visual particular face, all of these neurons can fi re to imagination (with one’s eyes closed) may be less somewhat face-like objects (Young and Yamane widespread than a coalition produced by a vivid 1992). Notice that while the information needed to and sustained visual input from the environment. represent a face is contained in the fi ring of the gan- These cortical neuronal networks (at least for glion cells in the retina, there is no explicit repre- perception) can be thought of as having nodes. sentation of the face there. A representation of an Each node is needed to express one aspect of one object or an event will usually consist of represen- percept or another. An aspect cannot become tations of many of the relevant aspects of it, and conscious unless there is an essential node for it. these are likely to be distributed, to some degree, For consciousness, there may be other necessary over different parts of the visual system. conditions, such as projecting to the frontal 76 M.A. Hofman

cortical areas. Thus, a particular coalition is an and spatial complexity of the species’ niche, active network, consisting of the relevant set of without referring, however, to the kinds of situa- interacting nodes that temporarily sustain itself tions organisms encounter in everyday life (Crick and Koch 2003 ). The smallest useful node (Reader et al. 2011 ; Roth and Dicke 2012 ). It is, may be a cortical column (Mountcastle 1997 ) or, in fact, a measure of capacity, independent of the perhaps, a portion of a cortical column. The fea- way the capacity is used, and it may be treated as ture which that node represents is (broadly) its a trait for “anagenetic” rather than “cladistic” columnar property. Edelman and Tononi ( 2000 ) analysis (Gould 1976 ; Jerison 1985 ). It implies presented a theory of consciousness, based on the that when distantly related species are compara- idea of a “dynamic core,” which resembles the ble in their problem-solving capacity, we should coalition concept to a large extent. The dynamic consider the species to be comparable in biologi- core hypothesis, however, rejects the idea that cal intelligence. Yet the near equality in intelli- there is a special subset of neurons that alone gence may be based upon radically different expresses the neural correlate of consciousness, a adaptations. Since neural mechanisms and action view which is also defended in this chapter. patterns evolve in the contexts of the environ- Most of the theories of consciousness have the ments in which they are effective, and since spe- idea of competing assemblies of neurons in com- cies never occupy identical niches, many and mon. Consciousness depends on certain coali- various intelligences (in the plural) must have tions that rest on the properties of very elaborate evolved in conjunction with evolving environ- neuronal networks. It is suggested that attention ments (Jerison 1985 ). consists of mechanisms that bias the competition In theory, each ecological niche requires its among coalitions, especially during their own degree of biological intelligence. That means formation. Furthermore, the idea that the spatio- that specifi c neural and sensorimotor adaptations temporal dimensions of these nodes represent the always occur in relation to particular environ- neural correlates of mind is most appealing, as it ments. A striking example is the mammalian suggests that consciousness, being an integral brain, where the evolutionary changes in the bal- part of the species’ problem-solving capacity, ance of the sensory systems are the result of the correlates to some extent with the degree of com- adaptive radiation of species into many different plexity of a nervous system. Therefore, the search ecological niches (Pirlot 1987 ; Macphail 1993 ; for the neural correlates of consciousness should Macphail and Bolhuis 2001 ; for reviews, see Roth be complemented by a search for its computa- and Wullimann 2001 ). These sensory systems, tional correlates (see, e.g., Atkinson et al. 2000 ; like any other biological feature, could evolve as a Zeman 2001 ). result of natural selection, because any subject that forms inadequate representations of outside reality will be doomed by natural selection. Evolutionary Models of the Mind In this view, cognitive systems and emotional phenomena can also be considered to be the Considering biological intelligence as the result of interactions between genetic aptitude problem-solving capacity of an organism makes and natural environment, as they have a number it possible to speak of degrees of intelligence and of biologically useful functions: one is to keep of its evolution from amoeba to man (Hofman track of the individual’s whereabouts in the world 2003 ). But what does it mean precisely when one by constructing a schematic model of reality says that species differ in intelligence or that ver- (Popper 1982 ; Churchland and Churchland 2002 ; tebrates are in general more intelligent than Premack 2007 ). It is evident that the mind, as an invertebrates? It means that there are differences emergent property of suffi ciently complex living in the abilities of organisms to perceive and inter- systems, has its evolutionary history like any pret the physical world. Biological intelligence other trait that increases adaptation to the can thus be conceived as to refl ect the temporal environment and that its functions have increased 5 Evolution of the Human Brain: From Matter to Mind 77 with the evolution from lower to higher organ- learning mechanisms in the cerebral cortex (see isms (Popper 1982 ). It might explain the dramatic Panksepp and Panksepp 2000 ; Adolphs 2009 ). It evolutionary expansion of the human neocortex, does not mean that there are no special-purpose being the region where both perception and learning systems in the brain, such as fear learn- instruction take place, where the external world is ing, but the human cerebral cortex includes much interpreted and modeled, where concepts are more than a conglomeration of special-purpose formed and hypotheses tested, in short, where the learning mechanisms. It contains a neural archi- physical world interacts with the mind. tecture that can generate fl exible features which Evolutionary psychology seeks to explain may be best conceptualized as rewritable. these evolved functional characteristics of the Cultural evolution, however, being the emer- human mind through the lens of an explanatory gent result of the evolution of mind, cannot dis- framework where special adaptive mechanisms pense with biological preconditions; it builds on are postulated to have been critical for hominid biological facts and faculties. Though cultural survival and reproductive success (see Panksepp evolution indeed presupposes biological evolu- et al. 2002 ). These “adaptive modules” are theo- tion, it is not fully explicable in terms of theories retical constructs unique to the hominid lineage and methods of the latter. In fact, cultural evolu- and should be clearly distinguished from the spa- tion has transgressed organic evolution and tiotemporally defi ned neural processing units (or shows a certain autonomy (see Hofman 2003 ). modules) of the cerebral cortex discussed in the The special status of cultural heredity can be previous sections (for a review, see Shettleworth derived from the fact that most cultural innova- 2012b ). The existence of a variety of genetically tions are devised precisely in order to meet the inherited “adaptive modules” is dubious at best environmental challenges or to improve our mod- when considered simultaneously with our current els of reality, whereas biological evolution has a understanding of mammalian brain organization. mindless random character. It is appropriate, Indeed, the organization of the cerebral cortex, therefore, to distinguish adaptations to the envi- which is commonly assumed to be a prime ana- ronment due to cultural selection from those that tomical substrate for unique cognitive functions, take place by the selection of genotypes. Cultural exhibits no robust signs of localized anatomical inheritance, furthermore, is an infi nitely faster specialization above and beyond specifi c sensory process than genetic inheritance, since it is based and motor connections and their polymodal on the transmission of information through direct interactions. communication and through books, the arts, and Although the adaptation of an organism to its the media, which makes that a new scientifi c dis- environment is the chief process directing bio- covery, or technical achievement can be transmit- logical evolution, with the evolution of intelli- ted to the whole of mankind in less than one gence, organisms became more and more generation (Ayala 1986 ). independent of their environments, by modifying the environments according to their needs. This process culminated in the evolution of mankind, Human Language and Intelligence which can be understood only as a result of the interaction of two kinds of evolution: the biologi- It is evident that the role of human language in cal and the cultural (Ayala 1986 ; Herrmann et al. the transmission of knowledge is extremely 2007 ; Premack 2007 ). Such considerations have important, even so prominent and pervasive that led various authors to argue that the human brain it is hardly possible to estimate human general can acquire a large variety of epigenetically intellectual capacity independent of linguistic derived functions via interactions of a limited capacity (Macphail 1982 ; Schoenemann 2012 ). number of evolutionary conserved affective/ Its manifestations and, in particular, that of its motivational systems (situated largely in subcor- newly acquired functions—description and argu- tical areas) with a set of plastic general-purpose mentation—are the most peculiar phenomena in 78 M.A. Hofman human problem solving. While animals can com- it is surely not enough: the degree of intelligence municate by expressing their inner state by means reached by a species does not determine the proof their behavior and by signaling to conspecifi cs pensity of its reproductive success. This may be (e.g., in case of danger), man is the only creature inferred from the indiscriminate elimination of that is able to make true and false statements and millions of species through the eras, from ammo- to produce valid and invalid arguments. nites to australopithecines. It means that though The progressive accumulation of interactions adaptability increases with the evolution of bio- between environment (both physical and social), logical intelligence, environmental catastrophes “conserved” subcortical systems, and the can always be fatal to a species. But not only “general-purpose” cerebral cortex gave rise to a external factors can threaten the existence of qualitatively different shade of mind—one that organisms; Homo sapiens, despite its impressive could communicate not merely with signs, but in intellectual capacities, might in the end become symbolic terms. On the other hand, we have seen the victim of its own mind by, paradoxically, cre- that a language system—of the type found in ating problems that it is then unable to solve. humans—is not essential for consciousness. It is plausible that organisms that do not possess a sophisticated language system are aware of the Concluding Remarks external world. This is not to say, however, that language does not enrich consciousness or that it All organisms are constantly engaged in solving does not contribute to our model of reality. problems and must therefore have fi tting and rel- If we assume that part of the basis of human evant models of their specifi c environments in speech is inherited in the DNA and that language order to enhance their chances of survival. is as much a biological as a cultural adaptation, Consequently, the problem-solving capacity of a then changes in the brain that permit the advanta- species is assumed to refl ect the temporal and geous supplement of language acquisition to per- spatial complexity of its ecological niche. ception and communication would have had The thesis presented here is that biological obvious selective advantages throughout the intelligence can be considered to be a correlate of period of hominid evolution (Deacon 1998 ; the problem-solving capacity of a species, mani- Schoenemann 2012 ). We may conceive human festing itself in the complexity of the species’ language, therefore, as a superorganic form of model of reality. With the evolution of sensory adaptation, having evolved not only as a cogni- systems as adaptations to specialized environ- tive adaptation contributing to the knowledge of ments, the capacity to process large amounts of reality of each individual but also as a means of sensory information increased and, with that, the sharing and, even more importantly, infl uencing power to create more complex physical realities. states of mind among conspecifi cs. Indeed, The processing of large amounts of information because of language, human beings are not only originating from the various sense organs and the able to construct individual representations of the construction of complex models of reality require external world, but they can also contribute to a neural system that selects, integrates, stores, and and learn from collective models of reality, that models—in other words, a system with mind-like is, the cumulative experience of the whole of properties that enables the organism to make mankind. With its cognitive and linguistic skills, sense of an otherwise chaotic world. But once we Homo sapiens tries to know its world and even allow mind-like properties to come in, such as exerts itself to the utmost to control it. motivation, emotion, preference, and anticipation, It is obvious that by virtue of language, human we must allow that it is not only the hostile envi- beings tend to have highly organized informa- ronment which plays an organizing or designing tional states of mind and, consequently, are excel- role in the evolution of biological intelligence but lent problem solvers. But although knowledge of also the active search of an organism for a new reality may be a necessary condition for survival, ecological niche, a new mode of living. 5 Evolution of the Human Brain: From Matter to Mind 79

Since the mind, prehuman and human, takes a Buxhoeveden, D. P., & Casanova, M. F. (2002a). The most active part in evolution and especially in its minicolumn hypothesis in neuroscience. Brain, 125 , 935–951. own evolution, hominization and the evolution of Buxhoeveden, D. P., & Casanova, M. F. (2002b). The our linguistic world may have begun as a cultural minicolumn and the evolution of the brain. Brain, adaptation to new ecological niches. The process Behavior and Evolution, 60 , 125–151. probably started at the time of hominid diver- Casanova, M. F., El-Baz, A., & Switala, A. (2011). Laws of conservation as related to brain growth, aging, and gence a few million years ago, as part of the cog- evolution: Symmetry of the minicolumn. Frontiers in nitive and manipulative adaptation to what was in Neuroanatomy, 5 , 66. doi: 10.3389/fnana.2011.00066 . essence a more complex physical reality. In other Changizi, M. A. (2001). Principles underlying mamma- words, some of the seemingly unique higher lian neocortical scaling. Biological Cybernetics, 84 , 207–215. functions of the human brain, such as language Changizi, M. A. (2007). Scaling the brain and its connec- and other neuro-symbolic capacities, were not tions. In J. H. Kaas (Ed.), Evolution of nervous sys- necessarily due to genetic selection and may have tems (Vol. 3, pp. 167–180). New York: Academic. emerged epigenetically through learning and cul- Changizi, M. A., & Shimojo, S. (2005). Parcellation and area-area connectivity as a function of neocortex size. tural experiences because of the dramatic expan- Brain, Behavior and Evolution, 66 , 88–98. sion of the neocortex and its increased tendency Cherniak, C. (1995). Neural component placement. to neural plasticity. It seems that the time is Trends in Neurosciences, 18 , 522–527. ﬁ nally ripe to begin to building an evolutionary Cherniak, C. (2012). Neural wiring optimization. Progress in Brain Research, 195 , 361–371. viewpoint of the mind based on comparative con- Chklovskii, D. B., Schikorski, T., & Stevens, C. F. (2002). cepts that incorporate the intrinsic systems found Wiring optimization in cortical circuits. Neuron, 34 , in all primate brains. 341–347. Chklovskii, D. B., Mel, B. W., & Svoboda, K. (2004). Cortical rewiring and information storage. Nature, 431 , 782–788. Churchland, P. S., & Churchland, P. M. (2002). Neural References worlds and real worlds. Nature Reviews Neuroscience, 3 , 903–907. Aboitiz, F., Lopez, J., & Mortiel, J. (2003). Long distance Clark, D. A., Mitra, P. P., & Wang, S. S.-H. (2001). communication in the human brain: Timing con- Scalable architecture in mammalian brains. Nature, straints for inter-hemispheric synchrony and the origin 411 , 189–192. of brain lateralization. Biological Research, 36 , 89–99. Cochrane, P., Winter, C. S., Hardwick, A. (1995). Adolphs, R. (2009). The social brain: Neural basis of Biological limits to information processing in the social knowledge. Annual Review of Psychology, 60 , human brain. Retrieved from: http://www.cochrane. 693–716. org.uk/opinion/archive/articles.phd Allen, J. S. (2009). The lives of the brain: Human evolu- Crick, F., & Koch, C. (1990). Towards a neurobiological tion and the organ of mind . Cambridge, MA: Belknap. theory of consciousness. Seminars in Neuroscience, 2 , Atkinson, A. P., Thomas, M. S. C., & Cleeremans, A. 263–275. (2000). Consciousness: Mapping the theoretical land- Crick, F., & Koch, C. (1995). Are we aware of neural scape. Trends in Cognitive Sciences, 4 , 372–382. activity in primary visual cortex? Nature, 375 , Ayala, F. J. (1986). Booknotes. Biology and Philosophy, 1 , 121–123. 249–262. Crick, F., & Koch, C. (1998). Consciousness and neuro- Azevedo, F. A. C., Carvalho, L. R. B., Grinberg, L. T., science. Cerebral Cortex, 8 , 97–107. Farfel, J. M., Ferretti, R. E. I., Leite, R. E. P., Filho, Crick, F., & Koch, C. (2003). A framework of conscious- W. J., Lent, R., & Herculano-Houzel, S. (2009). Equal ness. Nature Neuroscience, 6 , 119–126. numbers of neuronal and nonneuronal cells make the Da Costa, N. M., & Martin, K. A. C. (2010). Whose corti- human brain an isometrically scaled-up primate brain. cal column would that be? Frontiers in Neuroanatomy, The Journal of Comparative Neurology, 513 , 4 , 16. doi: 10.3389/fnana.2010.00016 . 532–541. De Sousa, A., & Cunha, E. (2012). Hominins and the Baars, B. J. (1997). In the theater of consciousness . emergence of the modern human brain. Progress in New York: Oxford University Press. Brain Research, 195 , 293–322. Bullmore, E., & Sporns, O. (2012). The economy of brain Deacon, T. W. (1990). Rethinking mammalian brain evo- network organization. Nature Reviews Neuroscience, lution. American Zoologist, 30 , 629–705. 13 , 336–349. Deacon, T. W. (1998). The symbolic species: The Buxhoeveden, D. P. (2012). Minicolumn size and human co- evolution of language and the brain . New York: cortex. Progress in Brain Research, 195 , 219–235. Norton and Company. 80 M.A. Hofman

Desimone, R., & Duncan, J. (1995). Neural mechanisms R. P. Kesner & D. S. Olton (Eds.), Neurobiology of of selective visual attention. Annual Review of comparative cognition (pp. 1–20). Hillsdale: Erlbaum. Neuroscience, 18 , 193–222. Hofman, M. A. (1988). Size and shape of the cerebral cor- Edelman, G. M., & Tononi, G. A. (2000). A universe of tex in mammals. Part II. The cortical volume. Brain, consciousness . New York: Basic Books. Behavior and Evolution, 32 , 17–26. Falk, D. (1990). Brain evolution in homo: The “radiator” Hofman, M. A. (1989). On the evolution and geometry of theory. Behavioral and Brain Sciences, 13 , 333–381. the brain in mammals. Progress in Neurobiology, 32 , Falk, D. (2004). Braindance: New discoveries about 137–158. human origins and brain evolution. Revised and Hofman, M. A. (1996). Evolution of the human brain: expanded edition. Gainesville: University Press of Design without a designer. In N. Elsner & Florida. H.-U. Schnizler (Eds.), Brain and evolution (pp. 141– Falk, D. (2007). Evolution of the primate brain. In 170). Stuttgart: Thieme Verlag. W. Henke & I. Tattersall (Eds.), Handbook of palaeo- Hofman, M. A. (2001a). Evolution and complexity of the anthropology (Vol. 2, pp. 1133–1162). New York: human brain: Some organizing principles. In G. Roth Springer. & M. F. Wullimann (Eds.), Brain evolution and cogni- Falk, D. (2012). Hominin paleoneurology: Where are we tion (pp. 501–521). New York: Wiley. now? Progress in Brain Research, 195 , 253–270. Hofman, M. A. (2001b). Brain evolution in hominids: Are Felleman, D. J., & Van Essen, D. C. (1991). Distributed we at the end of the road. In D. Falk & K. R. Gibson hierarchical processing in the primate cerebral cortex. (Eds.), Evolutionary anatomy of the primate cerebral Cerebral Cortex, 1 , 1–47. cortex (pp. 113–127). Cambridge: Cambridge Finlay, B. L., & Darlington, D. B. (1995). Linked regulari- University Press. ties in the development and evolution of mammalian Hofman, M. A. (2003). Of brains and minds: A neurobio- brains. Science, 268 , 1578–1584. logical treatise on the nature of intelligence. Evolution Finlay, B. L., Darlington, D. B., & Nicastro, N. (2001). and Cognition, 9 , 178–188. Developmental structure in brain evolution. Behavioral Hofman, M. A. (2007). Brain evolution and intelligence in Brain Sciences, 24 , 263–278. primates. In S. Watanabe & M. A. Hofman (Eds.), Frahm, H. D., Stephan, H., & Stephan, M. (1982). Integration of comparative neuroanatomy and cogni- Comparison of brain structure volumes in insectivora tion (pp. 33–53). Tokyo: Keio University Press. and primates. Part I. Neocortex. Journal für Hofman, M. A. (2012). Design principles of the human Hirnforschung, 23 , 375–389. brain: An evolutionary perspective. Progress in Brain Gould, S. J. (1976). Grades and clades revisited. In R. B. Research, 195 , 373–390. Masterton, W. Hodos, & H. J. Jerison (Eds.), Evolution, Hofman, M. A., & Falk, D. (Eds.). (2012). Evolution of brain and behavior: Persistent problems (pp. 115– the brain in primates. From neuron to behavior . 122). Hillsdale: Erlbaum. Amsterdam: Elsevier. Greenfi eld, S. A. (1995). Journey to the centers of the Jerison, H. J. (1985). Animal intelligence as encephaliza- mind . New York: Freeman. tion. Philosophical Transactions of the Royal Society Herculano-Houzel, S. (2009). The human brain in num- London B, 308 , 21–35. bers: A linearly scaled-up primate brain. Frontiers in Kaas, J. H. (1993). Evolution of multiple areas and mod- Human Neuroscience, 3, 31. doi: 10.3389/ ules within neocortex. Perspectives in Developmental neuro.09.031.2009 . Neurobiology, 1 , 101–107. Herculano-Houzel, S. (2012). Neuronal scaling rules for Kaas, J. H. (2000). Why is brain size so important: Design primate brains: The primate advantage. Progress in problems and solutions as neocortex gets bigger or Brain Research, 195 , 325–340. smaller. Brain and Mind, 1 , 7–23. Herculano-Houzel, S., Collins, C.E., Wong, P., Kaas, J.H. Kaas, J. H. (2008). The evolution of the complex sensory & Lent, R. (2008). The basic nonuniformity of the and motor systems of the human brain. Brain Research cerebral cortex. Proceedings of the National Academy Bulletin, 75 , 384–390. of Sciences of the United States of America, 105 , Kaas, J. H. (2012). The evolution of neocortex in pri- 12593–12598. mates. Progress in Brain Research, 195 , 91–102. Herculano-Houzel, S., Mota, B., Wong, P., & Kaas, J. H. Klein, R. G. (2009). The human career (3rd ed.). Chicago: (2010). Connectivity-driven white matter scaling and Chicago University Press. folding in primate cerebral cortex. Proceedings of the Klyachko, V. A., & Stevens, C. F. (2003). Connectivity National Academy of Sciences of the United States of optimization and the positioning of cortical areas. America, 107 , 19008–19013. Proceedings of the National Academy of Sciences of Herrmann, E., Call, J., Hernandez-Lloreda, M. V., Hare, the United States of America, 100 , 7937–7941. B., & Tomasello, M. (2007). Humans have evolved Krubitzer, L. (1995). The organization of neocortex in specialized skills of social cognition: The cultural mammals: Are species differences really so different? intelligence hypothesis. Science, 317 , 1360–1366. Trends in Neurosciences, 18 , 408–417. Hodos, W., & Campbell, C. B. G. (1990). Evolutionary Krubitzer, L. (2007). The magnifi cent compromise: Cortical scales and comparative studies of animal cognition. In fi eld evolution in mammals. Neuron, 56 , 201–208. 5 Evolution of the Human Brain: From Matter to Mind 81

Laughlin, S. B., & Sejnowski, T. J. (2003). Communication in the neocortex. Frontiers in Neuronanatomy, 4 , 3. in neural networks. Science, 301 , 1870–1874. doi: 10.3389/neuro.05.003.2010 . Lefebvre, L. (2012). Primate encephalization. Progress in Rakic, P. (2007). The radial ediﬁ ce of cortical architec- Brain Research, 195 , 393–412. ture: From neuronal silhouettes to genetic engineer- Lefebvre, L., Reader, S. M., & Sol, D. (2004). Brains, ing. Brain Research Reviews, 55 , 204–219. innovations and evolution in birds and primates. Brain, Rakic, P. (2009). Evolution of the neocortex: A perspec- Behavior and Evolution, 63 , 233–246. tive from developmental biology. Nature Reviews Lent, R., Azevedo, F. A. C., Andrade-Moraes, C. H., & Neuroscience, 10 , 724–735. Pinto, A. V. O. (2012). How many neurons do you Reader, S. M., Hager, Y., & Laland, K. N. (2011). The have? Some dogmas of quantitative neuroscience evolution of primate general and cultural intelligence. under revision. European Journal of Neuroscience, 35 , Philosophical Transactions of the Royal Society 1–9. London B, 366 , 1017–1027. Macphail, E. M. (1982). Brain and intelligence in verte- Ringo, J. L. (1991). Neuronal interconnection as a func- brates . Oxford: Clarendon. tion of brain size. Brain, Behavior and Evolution, 38 , Macphail, E. M. (1993). The neuroscience of animal intel- 1–6. ligence: From the seahare to the seahorse . New York: Ringo, J. L., Doty, R. W., Demeter, S., & Simard, P. Y. Columbia University Press. (1994). Time is of the essence: A conjecture that hemi- Macphail, E. M., & Bolhuis, J. J. (2001). The evolution of spheric specialization arises from interhemispheric intelligence: Adaptive specializations versus general conduction delay. Cerebral Cortex, 4 , 331–343. process. Biological Reviews, 76 , 341–364. Robson, S. L., & Wood, B. (2008). Hominin life history: Mayr, E. (1982). The growth of biological thought. Reconstruction and evolution. Journal of Anatomy, Diversity, evolution and inheritance. Cambridge, MA: 212 , 394–425. Belknap. Rockland, K. S. (2010). Five points on columns. Frontiers Mountcastle, V. B. (1997). The columnar organization of in Neuroanatomy, 4 , 22. doi: 10.3389/ the brain. Brain, 120 , 701–722. fnana.2010.00022 . Nieuwenhuys, R. (1994a). The neocortex: An overview of Roth, G., & Dicke, U. (2005). Evolution of the brain and its evolutionary development, structural organization intelligence. Trends in Cognitive Sciences, 9 , and synaptology. Anatomy and Embryology, 190 , 250–257. 307–337. Roth, G., & Dicke, U. (2012). Evolution of the brain and Nieuwenhuys, R. (1994b). The human brain: An introduc- intelligence in primates. Progress in Brain Research, tory survey. Medica Mundi, 39 , 64–79. 195 , 413–430. Northcutt, R. G., & Kaas, J. H. (1995). The emergence Roth, G., & Wullimann, M. F. (Eds.). (2001). Brain evolu- and evolution of mammalian neocortex. Trends in tion and cognition . New York: Wiley. Neurosciences, 18 , 373–379. Scannell, J. W., Blakemore, C. J., & Young, M. P. (1995). Panksepp, J., & Panksepp, J. B. (2000). The seven sins of Analysis of connectivity in the cat cerebral cortex. The evolutionary psychology. Evolution and Cognition, 6 , Journal of Neuroscience, 15 , 1463–1483. 108–131. Schmahmann, J. D. (2010). The role of the cerebellum in Panksepp, J., Moskal, J. R., Panksepp, J. B., & Kroes, cognition and emotion: Personal reﬂ ections since R. A. (2002). Comparative approaches in evolutionary 1982 on the dysmetria of thought hypothesis, and its psychology: Molecular neuroscience meets the mind. historical evolution from theory to therapy. Neuroendocrinology Letters, 23(Special issue, Suppl Neuropsychology Reviews, 20 , 236–260. 4), 105–115. Schoenemann, P. T. (2006). Evolution of the size and Pirlot, P. (1987). Contemporary brain morphology in eco- functional areas of the human brain. Annual Review of logical and ethological perspectives. Journal für Anthropology, 35 , 379–406. Hirnforschung, 28 , 145–211. Schoenemann, P. T. (2012). Evolution of brain and intel- Popper, K. R. (1982). The place of mind in nature. In ligence. Progress in Brain Research, 195 , 443–459. R. Q. Elvee (Ed.), Mind in nature (pp. 31–59). San Schoenemann, P. T., Sheehan, M. J., & Glotzer, I. D. Francisco: Harper and Row. (2005). Prefrontal white matter volume is dispropor- Premack, D. (2007). Human and animal cognition: tionately larger in humans than in other primates. Continuity and discontinuity. Proceedings of the Nature Neuroscience, 8 , 242–252. National Academy of Sciences of the United States of Semendeferi, K., Lu, A., Schenker, N., & Damasio, H. America, 104 , 13861–13867. (2002). Humans and great apes share a large frontal Prothero, J. W., & Sundsten, J. W. (1984). Folding of the cortex. Nature Neuroscience, 5 , 272–276. cerebral cortex in mammals: A scaling model. Brain, Semendeferi, K., Teffer, K., Buxhoeveden, D. P., Park, Behavior and Evolution, 24 , 152–167. M. S., Bludau, S., Amunts, K., Travis, K., & Raghanti, M. A., Spocter, M. A., Butti, C., Hof, P. R., & Buckwalter, J. (2011). Spatial organization of neurons Sherwood, C. C. (2010). A comparative perspective on in the frontal pole sets humans apart from great apes. minicolumns and inhibitory GABAergic interneurons Cerebral Cortex, 21 , 1485–1497. 82 M.A. Hofman

Shettleworth, S. J. (2012a). Fundamentals of comparative Wyatt, K. D. (2008). Functional trade-offs in white cognition . New York: Oxford University Press. matter axonal scaling. The Journal of Neuroscience, Shettleworth, S. J. (2012b). Modularity, comparative cog- 28 , 4047–4056. nition and human uniqueness. Philosophical Wedeen, V. J., Rosene, D. L., Wang, R., Dai, G., Mortazavi, Transactions of the Royal Society London B, 367 , F., Hagmann, P., Kaas, J. H., & Tseng, W. Y. (2012). 2794–2801. The geometric structure of the brain fi ber pathway. Smaers, J. B., Schleicher, A., Zilles, K., & Vinicius, L. Science, 335 , 1628–1638. (2010). Frontal white matter volume in anthropoid pri- Welker, W. (1990). Why does cerebral cortex fi ssure and mates. PLoS One, 5, e9123. doi: 10.1371/journal. fold? A review of determinants of gyri and sulci. In pone.0009123 . E. G. Jones & A. Peters (Eds.), Cerebral cortex (Vol. Sporns, O., Chilavo, D. R., Kaiser, M., & Hilgetag, C. C. 8B, pp. 23–136). New York: Plenum Press. (2004). Organization, development and function of Wen, Q., & Chklovskii, D. B. (2005). Segregation of the complex brain networks. Trends in Cognitive Sciences, brain into gray and white matter: A design minimizing 8 , 418–425. conduction delays. PLoS Computational Biology, 1 , Sporns, O., Honey, C. J., & Kötter, R. (2007). 617–630. Identifi cation and classifi cation of hubs in brain net- Young, M. P. (1993). The organization of neural systems works. PLoS One, 2, e1049. doi: 10.1371/journal. in the primate cerebral cortex. Proceedings of the pone.0001049 . Royal Society of London B, 252 , 13–18. Striedter, G. F. (2004). Principles of brain evolution . Young, M. P., & Yamane, S. (1992). Sparse population Sunderland: Sinauer Associates. coding of faces in the inferotemporal cortex. Science, Teffer, K., & Semendeferi, K. (2012). Human prefrontal 256 , 1327–1331. cortex: Evolution, development, and pathology. Zeman, A. Z. J. (2001). Consciousness. Brain, 124 , Progress in Brain Research, 195 , 191–218. 1263–1289. Van den Heuvel, M. P., & Sporns, O. (2011). Rich-club Zhang, K., & Sejnowski, T. J. (2000). A universal scaling organization of the human connectome. The Journal of law between gray matter and white matter of cerebral Neuroscience, 31 , 15775–15786. cortex. Proceedings of the National Academy of Wang, S. S.-H., Shultz, J. R., Burish, M. J., Harrison, Sciences of the United States of America, 97 , K. H., Hof, P. R., Towns, L. C., Wagers, M. W., & 5621–5626. Intelligence as a Conceptual Construct: The Philosophy 6 of Plato and Pascal

Dana Princiotta and Sam Goldstein

and Pascal 39. During their respective lives, the Introduction to Philosophers legacies they bestowed upon the world have of Different Times impacted the realms of science, religion, morality, mathematics and beyond. Philosophers seek truth and meaning in life. In the Republic, Plato states, “Then you see Philosophers engage in “the study of ideas about that this knowledge may be truly called knowledge, truth, the nature and meaning of life” necessary, necessitating as it clearly does the (Merriam-Webster Dictionary 2013 ). Such an use of the pure intelligence in the attainment of endeavor also entails deﬁ ning a set of ideas about pure truth” (cited in Plato and Jowett 2013 , knowledge, truth, the nature and meaning of life, p. 6998). In writing this chapter, we hope to how to do something, or how to live. enhance the reader’s understanding of Plato and This chapter presents the lives and ideas of Pascal and their roles in our current conceptual- whose work sets an integral foundation for our ization of intelligence. current conceptualizations of intelligence. Plato and Pascal’s speculation for the love of wisdom offered hundreds of years ago—the former P l a t o originating in France and the latter in Greece— remains vibrant today. The two philosophers Plato Enters the World did not restrict themselves to particular knowledge acquisition. Plato lived to the age of 80 Plato was born circa 428 B.C. in Athens, Greece. Due to the time period of Plato’s life, few primary sources exist. Born to parents Ariston and Perictione, it is suggested Plato was named after D. Princiotta , Ph.D. (*) his grandfather, Aristocles, and acquired the 1845 West Orange Grove Road , Tucson , nickname of Plato, based upon the width of his AZ 85704 , USA shoulders (Mastin 2008 ). Ariston passed away e-mail: [email protected] and Perictione remarried her uncle. Scholars S. Goldstein , Ph.D. speculate that Plato had two brothers, Adeimantus University of Utah , 230 South 500 East, Suite 100 , Salt Lake City , UT 84102 , USA and Glaucon; a sister, Potone; and a half brother, e-mail: [email protected] Antiphon (Mastin 2008 ).

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 83 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_6, © Springer Science+Business Media New York 2015 84 D. Princiotta and S. Goldstein

Beginnings of a Philosopher genuine knowledge in either of these ways one must engage in a strategy of examination much Plato’s education commenced in Athens. Plato like the systematic strategy of discovering on reportedly learned under the guidance of Socrates one’s own” (Benson 2005 ). beginning at age 20 (Mastin 2008 ). The two con- templated philosophy in concert for approximately 8 years (408 to 400 B.C.). Plato’s studies First Efforts of Defi ning Human with Socrates were terminated when Socrates Intelligence was placed on trial for impiety (e.g., introducing new deities to his society) and the “corruption of In what he called the “fi rst efforts of human intel- youth.” Plato would pay homage to his mentor ligence,” Plato provides his view as a philosopher and friend through his Socratic dialogues: (Plato and Jowett 2013 ): Apology, Crito, Euthyphro, and Phaedo. They were not, like being or essence, mere vacant Following Socrates’ execution in 399 B.C., abstractions, but admitted of progress and growth, Plato’s education ferried him abroad to Italy and while at the same time they confi rmed a higher sen- Egypt in his studies of geometry, astronomy, and timent of the mind, that there was order in the universe. And so there began to be a real sympathy geology as a disciple of Socrates. This travel between the world within and the world without. reportedly took place during 399–387 B.C., The numbers and fi gures which were present to the approximately. During this period, Plato pro- mind’s eye became visible to the eye of sense; the duced the Apology of Socrates among other writ- truth of nature was mathematics; the other properties of objects seemed to reappear only in the light ten works. Soon thereafter (385 B.C.), Plato of number. Law and morality also found a natural opened the doors to the Academy of Athens expression in number and fi gure. Instruments of (credited as the fi rst university) and educated oth- such power and elasticity could not fail to be a ers in philosophy for approximately 40 years. ‘most gracious assistance’ to the fi rst efforts of human intelligence. (Plato and Jowett 2013 , p. 745) The Academy endured until 539 A.D. The peda- gogics of the Academy of Athens emphasized Such efforts are again recognized in Plato’s mathematics, politics, philosophy, biology, and Republic: astronomy. These areas of study came to be And do you not know, I said, that all mere opinions known as Platonism (Mastin 2008 ). The are bad, and the best of them blind? You would not Byzantine Emperor Justinian I closed the deny that those who have any true notion without Academy of Athens. The Emperor believed that intelligence are only like blind men who feel their way along the road? (Plato and Jowett 2013 , the Academy was “a threat to the propagation of p. 6700) Christianity” (Mastin 2008 ). In his early years, Plato pursued his desire to grasp knowledge, contending that knowledge can either be acquired from a source that already Vehicles in Plato’s Philosophy comprehended this knowledge or by acquiring this knowledge independently (Benson 2005 ). The gestation of Plato’s philosophy relied upon Primarily, the two major outlets in which one three major factors—reason, will, and appetite could acquire knowledge could be sought through (Blaise Pascal 2013 ). He believed that humans dreams, oracles, and “other divination,” or were in a state of confl ict between appetites and through systematic strategies (Benson 2005 ). emotions (Hergenhahn 2009 ); thus, there is a Plato did not believe that knowledge could sim- battle for resources in discussing the needs of the ply be transferred from one individual to the next, human body and the rational pursuits. “The as might a virus. Plato also did not believe that supreme goal in life, according to Plato, should knowledge was acquired in a passive fashion of be to free the soul as much as possible from the “consuming” or listening to a lecture (Benson adulteration of the fl esh” (Hergenhahn 2009 , 2005 ). Plato believed that “in order to acquire p. 48). 6 Intelligence as a Conceptual Construct: The Philosophy of Plato and Pascal 85

Although Plato believed that intelligence was searching into things, which are upon the utmost a trait, he stated: limit of human intelligence, and then all of a sudden we fall rather heavily to the earth. There are no By reason of these affections the soul is at fi rst intermediate steps which lead from one to the without intelligence, but as time goes on the stream other. (Plato and Jowett 2013 , p. 807) of nutriment abates, and the courses of the soul regain their proper motion, and apprehend the Plato explained God’s greater plan through same and the other rightly, and become rational. The soul of him who has education is whole and Timaeus: perfect and escapes the worst disease, but, if a Timaeus: Why did the Creator make the world? He man’s education be neglected, he walks lamely was good, and therefore not jealous, and being free through life and returns good for nothing to the from jealousy he desired that all things should be world below. (Plato and Jowett 2013 , p. 308) like himself. Wherefore he set in order the visible world, which he found in disorder. Now he who is Plato’s dialogues allowed him to create and uti- best could only create the fairest; and refl ecting lize the Socratic method of questioning and answer- that of visible things the intelligent is superior to ing (Mastin 2008 ; Plato 2013a, b ). One of the major the unintelligent, he put intelligence in soul and goals in the use of dialogues in his teaching was an soul in body, and framed the universe to be the best and fairest work in the order of nature, and the attempt to instigate thought within his students. He world became a living soul through the providence also aspired to promote the independent thought of God. (Plato and Jowett 2013 , p. 206) and independent answering of questions (Mastin 2008 ). Perhaps the most famous dialogues include In other words, the world had become a living the Socratic dialogues, the Symposium, Republic, entity because the “God” instilled intelligence Timaeus, Critias, and the Sophist, though this list is within the creatures (Taylor 1928 ). However, this dependent upon individual interests (Hergenhahn intelligence was to be used for a purpose. “The 2009). In this chapter, we attempt to summarize a ancient physical philosophers have been charged selection of dialogues as they relate to the construc- by Dr. Whewell and others with wasting their fi ne tion of intelligence. intelligences in wrong methods of enquiry; and their progress in moral and political philosophy has been sometimes contrasted with their sup- Timaeus and Intelligence posed failure in physical investigations” (Plato and Jowett 2013 , p. 774). Although an accurate chronology of the dialogues does not exist, Timaeus and the Republic share common themes surrounding Plato’s Republic and Intelligence the construction of intelligence, orderliness, and society. Timaeus is one of Plato’s dialogues In Plato’s most famous dialogues, Plato’s including Socrates, Timaeus of Locri, Critias, Republic, Plato paints the image of a utopian and Hermocrates. The characters or philoso- society. “The Republic is considered one of the phers in this dialogue discuss the physical single most infl uential works in the whole of world and the human beings residing in nature Western Philosophy, although his account of (Taylor 1928). Socrates’ trial in the ‘Apology’ may be the most Plato highlights intelligence as being housed read” (McClain 1977 ). Plato’s motivation to cre- in the soul rather than the body (Carpenter 2008 ). ate a utopian society may have been infl uenced With the soul separated from the body by his by the traumatic execution of his mentor, through accounts, this lends to suggest that intelligence is democratic means, nonetheless (Taylor 1928 ). viewed as a trait (Carpenter 2008 ). Within his society exist three different types of individuals—those prescribing to appetite (e.g., There is further diffi culty in explaining this part of the Timaeus—the natural order of thought is workers, slaves), those prescribing to emotion inverted. We begin with the most abstract, and (e.g., soldiers), and those prescribing to reason proceed from the abstract to the concrete. We are (philosopher-kings) (Hergenhahn 2009 ). 86 D. Princiotta and S. Goldstein

Plato believed that the Republic necessitated do not often grow together, and that persons who leadership by the wise philosophers (Hergenhahn possess them and are at the same time high-spirited and magnanimous are not so constituted by nature 2009 ). He explains this by arguing: as to live orderly and in a peaceful and settled man- …And so we get four names—two for intellect, ner; they are driven any way by their impulses, and and two for opinion—reason or mind, understand- all solid principle goes out of them. (Plato and ing, faith, perception of shadows—which make a Jowett 2013 , p. 6646) proportion—being: becoming:: intellect: opinion—and science: belief:: understanding: perceptions of shadows…And would you have the future rulers of your ideal State intelligent beings, or stu- The Allegory of the Cave pid posts? Certainly not the latter. Then you must train them in dialect, which will teach them to ask Within Plato’s Republic, we are introduced to his and answer questions, and is the coping-stone of the sciences. (Plato and Jowett 2013 , p. 1710) allegory of the cave. In his allegory of the cave, Plato discusses our method of acquiring informa- Plato further describes this society: tion, namely, the education of society. In this alle- The lower portion of the lower or visible sphere gory, Plato paints an image of prisoners confi ned will consist of shadows and refl ections, and its to a cave in which they can only look forward upper and smaller portion will contain real objects (Plato and Jowett 2013 ). Prisoners are able to in the world of nature or of art. The sphere of the view shadows of individuals passing behind them intelligible will also have two divisions—one of mathematics, in which there is no ascent but all is through light from a fi re (Hergenhahn 2009 ). descent; no inquiring into premises, but only Plato depicts the prisoners’ understanding of drawing inferences. In this division the mind their world as a lower form of understanding works with fi gures and numbers, the image of from the viewpoint of the Divided of Line previ- which are taken not from the shadows, but from the objects, although the truth of them is seen only ously discussed (Hergenhahn 2009 ). The shad- with the mind’s eye; and they are used as hypoth- ows provide the current reality for the prisoners. eses without being analyzed…. (Plato and Jowett After providing various scenarios including 2013, p. 1453) the release of the prisoners from their confi nes, Plato concludes that the reality of the prisoners He continues by stating, “When the sun shines would certainly be shifted by the introduction of the eye sees, and in the intellectual world where the sun and their new reality. Furthermore, “if truth is, there is sight and light. Now that which is they could lay hands on the man who was trying the sun of intelligent natures, is the idea of good, to set them free and lead them up, they would kill the cause of knowledge and truth, yet other and him” (Trotter 1986, p. 257; Hergenhahn 2009 ). fairer than they are, and standing in the same This is an attempt to create an analogy in which relation to them in which the sun stands to light” educators are leading individuals into the light (Plato and Jowett 2013 , p. 1419). Within these we identify as education. Leading the prisoner spheres, Plato delineates four divisions or assign- into the light is seen as leading an individual into ments for knowledge (Plato and Jowett 2013 , the world of reason rather than simple sensory p. 1435): (1) pure intelligence, (2) active intelli- experiences (Hergenhahn 2009 ). In other words, gence, (3) faith, and (4) perception of shadows. those Plato describes as ignorant would not wish “And the clearness of the several faculties will be to be enlightened by reason, by his account. in the same ratio as the truth of objects to which they are related” (Plato and Jowett 2013, p. 1435). With much effort, Plato attempted to create an Pythagorean Concepts order in his society for the philosopher-kings to lead the society; however, Plato admitted to the It is commonly stated that the second half of shortcomings of this Republic: Plato’s career was heavily infl uenced by the You are aware, I replied, that quick intelligence, Pythagoreans (Taylor 1928 ). With the assistance memory, sagacity, cleverness, and similar qualities, of the Pythagoreans, Plato argued that experiences 6 Intelligence as a Conceptual Construct: The Philosophy of Plato and Pascal 87 of the mind were superior to those experienced Aristotle and the Future by the fl esh (Hergenhahn 2009 ). “Plato borrowed of Plato’s Work much from the Pythagoreans. It was through the Platonic philosophy that elements of the Aristotle was one of Plato’s best-known students Dionysiac-Orphic religion became part of (Taylor 1928 ). For Aristotle, it was challenging to the heritage of Western civilization” (Hergenhahn follow Plato and his teachings at times. 2009 , p. 35). Plato’s departure from the simpler realism of his One of the prominent concepts Plato adopted master, as noted by Aristotle, towards that “intelli- from the Pythagoreans in the second half of his gible world” opposed by him so constantly to the academic “career” entailed the Theory of Forms. visible world, into which many fi nd it so hard to follow him at all, and in which the “ideas” become Plato proposes a “dualistic universe consisting of veritable persons. To speak, to think, to feel, about abstract forms and matter” (Hergenhahn 2009 , abstract ideas as if they were living persons; that is p. 665). This concept states, “Everything in the the second stage of Plato’s speculative ascent. empirical world is a manifestation of a pure form (Plato and Jowett 2013 , p. 1774) (idea) that exists in the abstract. Thus chairs, chariots, rocks, cats, dogs, and people are inferior Others carried on Plato’s teaching. They manifestations of pure forms” (Hergenhahn were known as Neoplatonists (Mastin 2008 ). 2009 , p. 46). Stated differently, most of what we As Western civilization continued to develop, “experience” is merely a manifestation of Plato’s original writings were lost but resur- thoughts, ideas, or forms of the actual “object” faced through the help of Greek Neoplatonists (Hergenhahn 2009 , p. 46). Furthermore, “The (McClain 1977 ). only true knowledge is that of the forms, a knowledge that can be gained only by refl ecting on the innate contents of the soul. Sensory experience Legacy interferes with the attainment of knowledge and should be avoided” (Hergenhahn 2009 , p. 64). In Plato died in approximately 348 B.C. in Athens, other words, knowledge of intelligence relies on Greece, in his early 80s (Blaise Pascal 2013 ). He “intuitive” grasping of the world, not the grasp- provided foundations for the future of mathemat- ing of pure matter or pure form—“the material ics, reason, morals, equality, science, and nature. world as it seems to us is not the real world, but Western society has inherited Plato’s teachings of only a shadow or a poor copy of the real world” the physical body and the immortal soul, which (Mastin 2008 ). has provided support for the prominent world A second important Pythagorean concept, religions including Jewish, Christian, and Islamic Reminiscence Theory of Knowledge, states that studies. Plato’s teachings are imperative to our “all knowledge is innate and can be attained only modern construction of intelligence, making him through introspection, which is the searching of a prominent fi gure for educators, historians, and one’s inner experiences” (Hergenhahn 2009 , scientists to study. p. 48). Plato was a fi rm believer in reincarnation and reason being a function of the immortal soul (Hergenhahn 2009 ). Furthermore, Plato believed Pascal “All nature is akin, and the soul has learned everything, so that when a man has recalled a Pascal Enters the World single piece of knowledge…there is no reason why he should not fi nd out all the rest, if he Blaise Pascal was born on June 19, 1623, in keeps a stout heart and does not grow weary of Clermont-Ferrand, France, to Etienne and the search, for seeking and learning are in fact Antoinette Pascal (O’Connor and Robertson nothing, but recollection” (Hamilton and Cairns 1996 ). He was a brother to sisters Jacqueline and 1961, p. 364). Gilberte (Lataste 1911 ). His father assisted in the 88 D. Princiotta and S. Goldstein courts of Clermont as an advocate and was also dubbed the Pascaline. He later invented the known to be a tax collector and mathematician. hydraulic press and syringe (O’Connor and The family moved to Paris for educational oppor- Robertson 1996 ), the roulette machine (1655), tunities for the children following Antoinette’s and the cycloid (1659), an instrument able to cal- passing (Blaise Pascal 2013 ). culate solid volumes (O’Connor and Robertson 1996 , 2007). He was infi nitely interested in atmospheric pressure including measurements of A Child Prodigy Is Recognized atmospheric pressure through estimations of weight (Blaise Pascal 2013 ). In Paris, Blaise would learn Latin, Greek, and Spanish (Lataste 1911 ) from his father. Although his father wished to delay Blaise’s acquisition of Pascal’s Ailing Health mathematics, Blaise taught himself mathematical principles (Blaise Pascal 2013 ). He understood By 1646, Pascal’s ailing health impeded his that the sum of the angles of a triangle is always work. Pascal’s physician recommended he rest 180° at a very young age (O’Connor and for a period of time (Lataste 1911 ). At this point Robertson 1996 ). Blaise’s development contin- in time, his beliefs and behaviors were altered: ued to climb sharply, and his father recognized There are two vectors of experience in Pascal’s Blaise as a child prodigy. Contributing to the world, vectors that for inattentive people would fi elds of mathematics and science at an early age have remained unconnected but that for a creative following homeschooling in religion, philosophy, mind might well come together in new and interesting ways. First of all, in this time of his life, math, and physics (O’Connor and Robertson Pascal was a man of the world, living up to the 1996), Blaise was inducted into the Academie expectations of his class living the life of an intel- Libre, a group focusing on the development of ligent gentleman. This meant parties; this meant science and mathematics (Blaise Pascal 2013 ). diversion; this meant gambling. It is unlikely, given his health that he engaged in any dangerous liai- At the young age of 16, Blaise proposed and sons nor would his conscience have allowed it. So defended mathematical theorems. Shortly there- instead, he invented the new mathematics of prob- after, he published his Essay on Conic Sections ability, the calculation of expectations and a math- (Lataste 1911 ). ematics for probing the mysterious heart of Fortune. (Connor 2006 , p. 3679) As stated, these life experiences afforded him Major Academic Contributions opportunities to develop the probability theorem and Inventions with the help of his colleague Pierre de Fermat (Rogalsky 2006 ; O’Connor and Robertson 1996 ). In his short 39 years, the Essay on Conic Sections Pascal apprehended that there was indeed a fi xed was merely the beginning of important theorems likelihood of a particular outcome when rolling a relevant today. Geometry and probability were dice in 1654 (Blaise Pascal 2013 ). This probabil- two particularly important fi elds of study and con- ity theorem was recorded in Pensées or tribution for Pascal (Lataste 1911 ). The “mystic “Thoughts” and contributed to Pascal’s Wager, hexagram” was one of his earlier theorems, devel- discussed below. oped in adolescence as well (O’Connor and Robertson 1996 ), otherwise known as Pascal’s theorem. His theorems in mathematics would The Infl uence of a Near-Death later infl uence Sir Isaac Newton in his own work Experience relating to general binomial theorem for fractional and negative powers (Blaise Pascal 2013 ). Pascal’s work was channeled in a different man- Pascal’s inventions included an arithmetic ner when he had a frightening near-death experi- machine (i.e., a calculator) by the age of 18, ence in the year 1654 (O’Connor and Robertson 6 Intelligence as a Conceptual Construct: The Philosophy of Plato and Pascal 89

1996). This experience fueled two of Pascal’s you can defend neither of the propositions” greatest legacies—Lettres Provinciales and (Pascal 1660 ; Pascal & Trotter 1910). Pensées (Lataste 1911 ). Pascal described himself This argument is one in which Pascal remains in the following manner: a famous philosopher three and a half centuries What the gentlemen found after a time was a later. His words resonate with us today: young man whose intelligence far outstripped their —Yes; but you must wager. It is not optional. You own, who was able to see further and think more are embarked. Which will you choose then; Let us clearly, who had a pungent wit, who could be cruel see. Since you must choose, let us see which inter- one minute and solicitous the next. They came to ests you least. You have two things to lose, the true know a terribly unhappy young man who wanted and the good; and two things at stake, your reason to be a saint and yet love life, who berated himself and your will, your knowledge and your happi- for his worldliness one minute and then laughed at ness; and your nature has two things to shun, error some raucous joke the next. If they had been told and misery. Your reason is no more shocked in that this Pascal fellow would soon have a mystical choosing one rather than the other, since you must experience of God and withdraw from the world, of necessity choose…. In truth, there is infi nity they would not have been surprised, for he was between the certainty of gain and the certainty of halfway there anyway. If they had been told that he loss. But the uncertainty of the gain is propor- would soon after write one of the greatest and nas- tioned to the certainty of the stake according to the tiest works of the modern age, the Provincial proportion of the chances of gain and loss. (Pascal Letter, a book of satire that Voltaire kept by his 1660; Pascal & Trotter 1910) bedside every night and used as a model, and that these letters would be directed against the all- This is in accord with other “thoughts on mind powerful Jesuits, they would not have been sur- and on style” when Pascal states, “People are prised by that, either. (Connor 2006 , p. 354) generally better persuaded by the reasons which they have themselves discovered than by those which have come into the mind of others” (Pascal Pascal’s Wager 1660 ; Hughes 2011 ). This is true in Pascal’s own life experiences with a near-death experience and Previously known as a man of science, Pascal religious enlightenment. became known as a man of God—though in reality he was both (Lataste 1911 ). He proposed Pascal’s Wager, in that God is rational—“If God Pensées does not exist, one will lose nothing by believing in him, while if he does not, one will lose every- Pascal reveals his thoughts on intelligence and thing by not believing” (Pascal 1660 ; O’Connor reason in Pensées. He applied this to science, but and Robertson 1996 ). In general, Pascal insisted in later years of life, to religion (Lataste 1911 ). As that mankind must gamble (O’Connor and a philosopher, Pascal initiated important discus- Robertson 1996 ). Simply it stated: “If I wager for sions regarding intelligence and the acquisition of and God is-infi nite gain; if I wager for and God is knowledge. In Pensées (Pascal 1910), Pascal dis- not-no loss. If I wager against and God is-infi nite cusses the realm of mathematics and the acquisi- loss; if I wager against and God is not-neither tion of principles, distinguishing the process of loss nor gain” (Pascal 1660 ; Lataste 1911 ). acquisition among individuals. Pascal stated: Pascal’s insistence on wagering extends The reason, therefore, that some intuitive minds beyond the religious domain. When discussing are not mathematical is that they cannot at all turn religion, he states that in choosing whether God their attention to the principles of mathematics. But the reason that mathematicians are not intui- exists or not, “A game is being played at the tive is that they do not see what is before them, and extremity of this infi nite distance where heads or that, accustomed to the exact and plain principles tails will turn up. What will you wager? According of mathematics, and not reasoning till they have to reason, you can do neither the one thing nor the well inspected and arranged their principles, they are lost in matters of intuition where the principles other; according to reason, you can defend neither do not allow of such arrangement. (Pascal 1660 ; the one thing nor the other; according to reason, Pascal & Trotter 1910 , p. 2) 90 D. Princiotta and S. Goldstein

He continues by stating: feelings, seeking principles and being unable to And thus it is rare that mathematicians are intuitive see at a glance” (Pascal 1660 ; Rogalsky 2006 ). and that men of intuition are mathematicians, Ultimately, Pascal melds the two with Pensées because mathematicians wish to treat matters of 172: “The way of God who disposes all things intuition mathematically and make themselves with gentleness, is to instill religion into our ridiculous, wishing to begin with deﬁ nitions and then with axioms, which is not the way to proceed minds with reasoned arguments and into our in this kind of reasoning. Not that the mind does hearts with grace, but attempting to instill it into not do so, but it does it tacitly, naturally, and with- hearts and minds with force and threat is to out technical rules; for the expression of it is instill not religion but terror” (Pascal 1660 ; beyond all men, and only a few can feel it. (Pascal 1660 ; Pascal & Trotter 1910 , p. 2) Rogalsky 2006 ).

Furthermore, “Mathematicians who are only mathematicians have exact minds, provided all Final Thoughts on Pascal’s Philosophy things are explained to them by means of defi nitions and axioms; otherwise they are inaccurate Pascal believed that philosophy was a process, as and insufferable, for they are only right when the most philosophers or scientists might also con- principles are quite clear” (Pascal 1660 ; Pascal & clude. He argued the following: “Skepticism. Trotter 1910 , p. 2). Excess, like defect of intellect is accused of mad- In other words, Hughes ( 2011 ) interprets ness. Nothing is good but mediocrity” (Pascal Pascal’s position as the difference between 1660; Hughes 2011). Further, “All the principles intuitive and mathematical minds laying within of philosophers are true: the skeptics, the stoics, their perspective—“All mathematicians would the atheists and so on. But their conclusions are then be intuitive if they had clear sight, for they false, because the opposite principles are also true” do not reason incorrectly from principles known (Pascal 1660 ; Hughes 2011 ). He argues that phi- to them; and intuitive minds would be mathemat- losophers like Descartes probe science too deeply, ical if they could turn their eyes to the principles describing him as “useless and uncertain” (Pascal of mathematics to which they are unused” (Pascal 1660 ; Rogalsky 2006 ). Pascal challenged the likes 1660; Hughes 2011). In conclusion, “The greater of Descartes and the “Cartesian Monad” alike: intellect one has, the more originality one fi nds in There are different kinds of right understanding; men. Ordinary persons fi nd no difference between some have right understanding in a certain order of men” (Pascal & Trotter 1910 , p. 3). Pascal further things, and not in others, where they go astray. elaborates in perhaps a judgmental tone, “Do you Some draw conclusions well from a few premises, and this displays an acute judgment. For example, wish people to believe good of you? Don’t speak” the former easily learn hydrostatics, where the (Pascal 1660 ; Hughes 2011 ). premises are few, but the conclusions are so fi ne that only the greatest acuteness can reach them. And in spite of that these persons would perhaps not be great mathematicians, because mathematics Intelligence and Religion contain a great number of premises, and there is perhaps a kind of intellect that can search with ease On the fl ip side to the previous examples from a few premises to the bottom: and cannot in the Pensées, an example is illustrated in Pensées least penetrate those matters in which there are many premises. Others draw conclusions well number 423—“the heart has its reasons of which where there are many premises. (Pascal 1660 ; reason knows nothing” (Pascal 1660 ; Pascal & Pascal & Trotter 1910 ) Trotter 1910 , p. 172; Rogalsky 2006 ). Similarly, “Those who are accustomed to judge by feeling do not understand the process of reasoning, for they would understand at fi rst sight and are not Pascal’s Legacy used to seek for principles. And others, on the contrary, who are accustomed to reason from Blaise Pascal struggled with dyspepsia (i.e., principles, do not at all understand matters of digestive disorder) as well as insomnia for half of 6 Intelligence as a Conceptual Construct: The Philosophy of Plato and Pascal 91 his life. He died on August 19, 1662, of a malignant Certainly Plato laid the framework for Pascal, stomach tumor in Paris at the age of 39 (Lataste but as we have seen, Pascal was an independent 1911 ). He died two months after his 39th birth- thinker and contributed a wealth of innovative day. As a fi nal gesture, he invited a Parisian fam- ideas. Pascal infl uenced many fi elds of study, ily to live in his home and left the estate to this including physics, computer science, and geom- homeless family when he passed. etry. Three hundred years beyond Pascal’s death, During his last year of life, he also focused on in the 1970s, a unit of measurement was named helping the economically less fortunate travel after him related to atmospheric pressure, the through Paris by developing and crafting a plan Pascal (Pa). We continue to benefi t from Pascal’s for individuals to travel with horse-drawn car- logic. He is best remembered as a French Roman riages for a modest fare (Rogalsky 2006 ). Catholic philosopher (Rogalsky 2006 ). Awards According to sources, the profi ts were then are given in France in Pascal’s honor, known as donated to charity. At the time, this was the fi rst the Blaise Pascal Chairs for international scien- mass-transit system in Paris (Rogalsky 2006 ). tists in the Ile-de-France region. A university in his hometown of Clermont-Ferrand is named after Pascal. A programming language for computer scientists has also been named after Pascal. Concluding Remarks Finally, Pascal’s law refers to a principle in hydrostatics that is still utilized (O’Connor and Although it is diffi cult to locate all of Plato’s Robertson 1996 ). primary sources, most of his 36 dialogues and 13 Pensées was considered to be one of Pascal’s letters are available through different avenues on greatest contributions. Unfortunately, his work the World Wide Web. The very nature of informa- was incomplete at his time of death (O’Connor tion science can be traced back to Plato and and Robertson 1996 ). He accomplished much in Socrates’ philosophy regarding intelligence and his 39 years. Pascal has challenged us to better the acquisition of knowledge. We might credit understand and weight the merits of both reason Plato for teaching us how to think. Notice the and religion (Rogalsky 2006 ). He accomplished statement “teaching us how to think” rather than this though burdened by illness for much of his “teaching us what to think.” adult life. He has challenged and motivated three Plato instituted the fi rst institute of higher edu- hundred years of philosophers and thinkers cation in the Western world, an amenity that (Rogalsky 2006 ). T. S. Eliot paid homage to many continue to derive educational wealth from. Pascal by describing him as “a man of the world Although political causes attempted to thwart among ascetics, and an ascetic among men of the Plato’s drive as a philosopher (the fi rst being the world” (Broome 1965 ). According to documents, execution of his mentor, Socrates, early in life), Pascal stated even on his deathbed, “Sickness is Plato continued as a philosopher, and the acad- the natural state of Christians” (Broome 1965 ). emy prevailed for hundreds of years under the The legacies of Plato and Pascal are undeni- guidance of continuing scholars (Carpenter able. The memory of these men, though diffi cult 2008 ). The academy continued for centuries fol- to locate primary sources, is alive within the lowing Plato’s death, under the guidance of his works of philosophers and scientists existing nephew, Speusippus, with Aristotle on staff as a between the time periods of their lives into the teacher for 20 of those years (Carpenter 2008 ). twenty-fi rst century. Both men produced impor- The hope is as depicted in Plato’s Republic, tant contributions and philosophies to the con- “And so we arrive at the result, that the pleasure cepts of intelligence. of the intelligent part of the soul is the pleasantest We close with words from Pascal. Pascal of the three, and that he of us in whom this is the wrote: ruling principle has the pleasantest life” (Plato True eloquence makes light of eloquence, true and Jowett 2013 , p. 7935). morality makes light of morality; that is to say, the 92 D. Princiotta and S. Goldstein

morality of the judgment, which has no rules, Mastin, L. (2008). Socratic—by historical period—the makes light of the morality of the intellect. For it is basics of philosophy. Web. 01 Oct 2013. to judgment that perception belongs, as science McClain, E. G. (1977). The Pythagorean Plato: Prelude belongs to intellect. Intuition is the part of to the song itself . Stony Brook/Boulder: N. Hays. judgment, mathematics of intellect. To make light Merriam-Webster. (2013). Merriam-Webster.com. of philosophy is to be a true philosopher. (Pascal Retrieved 1 Oct 2013, from http://www.merriam- 1660 ; Jowett 2013) webster.com/dictionary/philosophy O’Connor, J. J., & Robertson, E. F. (1996). Blaise Pascal. Retrieved 22 July 2013, from http://www- history.mcs.st-andrews.ac.uk/Biographies/Pascal. html R e f e r e n c e s O’Connor, J. J., & Robertson, E. F. (2007). A history of calculus . University of St Andrews. Consultado el, Benson, H. H. (2005). Plato’s rationalistic method. In A. 07–08. Nelson (Ed.) A companion to rationalism . Oxford, UK: Pascal, B. (1910). Pascal’s Pensées (trans: Trotter, W. F.). Blackwell Publishing Ltd. doi: 10.1002/9780470996904. Pascal, B. (1660). Pensées (trans: Trotter, W. F.). ch5 . New York: Collier & Son, 1910. Blaise Pascal. (2013). The biography channel website . Plato. (1986). (W. F. Trotter, Trans.). Roslyn, NY: Black. Retrieved 22 July 2013, from http://www.biography. Plato. In Hamilton, E., & Cairns, H. (1961). The collected com/people/blaise-pascal-9434176 dialogues of Plato, including the letters. New York: Broome, J. H. (1965). Pascal. London: Edward Arnold. Pantheon Books. Carpenter, A. D. (2008). Embodying intelligence: Animals Plato. (2013a). The biography channel website . Retrieved and us in Plato’s Timaeus. In Dillon, J., & Zovko, J. 30 Sept 2013, from http://www.biography.com/ (Eds.). Berlin: Academie Verlag. people/plato-9442588 Connor, J. A. (2006). Pascal’s wager: The man who played Plato. (2013b). Essential dialogues of Plato (trans: Jowett, dice with God. San Francisco: HarperSanFrancisco. B.). Digireads.com Publishing. Hergenhahn, B. R. (2009). An introduction to the history of Plato, & Jowett, B. (2013). Plato’s the republic . Plato psychology (6th ed.). Wadsworth: Cengage Learning. (trans: Jowett, B.). ISBN 9781775420514: Amazon. Hughes. (2011). Blaise Pascal thoughts (Pensees). com: Books. Web. 01 Oct 2013. Retrieved 16 Sept 2013 from http://sqapo.com/pascal. Rogalsky, T. (2006). Blaise Pascal: Mathematician, mys- htm tic, disciple. Journal of the Association of Christians Lataste, J. (1911). Blaise Pascal. In The Catholic in Mathematical Sciences. Retrieved 02 Oct 2013 encyclopedia . New York: Robert Appleton Company. from http://www.acmsonline.org/journal2006.htm . Retrieved 26 Sept 2013 from New Advent: http:// Taylor, A. E. (1928). A commentary on Plato’s Timaeus. www.newadvent.org/cathen/11511a.htm Oxford: Clarendon Press. The Life and Evolution of Early Intelligence Theorists: Darwin, 7 Galton, and Charcot

Jordan Rigby

animals, plants, and humans. With their assistance, Charles Robert Darwin Darwin was later able to produce theories on the transformation of animals and humans that would Charles Robert Darwin was born on February 12, forever change science and infl uence psychology’s 1809, to a world with evolutionary theory still in its views of intelligence. infancy. Thinkers before him, such as Plato and Darwin was the fi fth of six children born to Aristotle, did not particularly believe in evolution Robert and Susan Darwin. He was described as a (Hergenhahn 2009 ). This did not prevent theorists “dreamy, grey-eyed, thickset child, intent on his from building on what was available. In fact, the own thoughts behind a shock of brown hair, but year Darwin was born Jean Lamarck published a warm-hearted and loving for all” (Browne 1995 , book suggesting an inheritance of acquired traits p. 10). Even from an early age, he was an avid col- relating to the passing down of muscular character- lector of antiques and natural objects. The plan for istics from adult to offspring. Contributions such as Darwin was to attend Shrewsbury School, located this identifi ed what was missing from early theo- at the center of Shrewsbury, Shropshire, England, ries, and a transformation took place. Another where he grew up. Darwin did not receive this contributor was Herbert Spencer. He specifi cally schooling well, stating, “Nothing could have been applied these ideas to the human mind and society. worse from the development of my mind” (Darwin He suggested that the nervous system began as 1892/2000, p. 9). Despite being busy with subjects something simple and homogeneous but trans- such as Greek and Roman history and literature, formed into something differentiated and very which were then considered subjects essential to a complex (Hergenhahn 2009 ). He was credited with gentlemanly education, Darwin still chose to use the introduction of the term intelligence into psy- his free time collecting specimens (Stefoff 1996 ). chology (Guilford 1967 ). These theories made it He was a mediocre student (Berra 2009 ). into scientifi c literature only after naturalists and Because Darwin was drawn to natural history evolutionists, such as Jean Lamarck and Herbert and science from a very early age, he and his Spencer, began discussing the altering of species of brother, Erasmus, set up a lab to carry out experiments (Browne 1995). Darwin continued these experiments even after his brother left for Cambridge University (Browne 1995 ). After * J. Rigby , Psy.D. ( ) leaving Shrewsbury at the age of 16 (Browne Neurology, Learning and Behavior Center , 230 South 500 East, Suite 100, Salt Lake City , UT 84102 , USA 1995), Darwin met up with Erasmus in Edinburgh, e-mail: [email protected] Scotland, fully intending on becoming a doctor

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 93 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_7, © Springer Science+Business Media New York 2015 94 J. Rigby like his father (Browne 1995 ). He was the third differently than those who fi rst instructed him generation of Darwin to study medicine. It was (Browne 1995 ). Initially, he thought he might not long before medicine began to look less write a book on the geology of the places he was attractive. When Darwin returned to Scotland to visiting (Berra 2009 ). He spent weeks at South begin his second year of medical school, he found American destinations such as Brazil, Argentina, the lectures boring and the horror of surgery and Chile. Much of his work was self-imposed without anesthesia too much for him to manage and diffi cult, but he wished for nothing more than (Berra 2009 ). He once again turned to natural the opportunity to complete it. He enjoyed this history and all but ignored his medical lectures to work and it was a source of great pleasure and attend lectures on taxidermy and presentations at pride to be involved in it (Browne 1995 ). A key to scientifi c societies (Berra 2009 ). Darwin then the excellence of this research was his meticulous transferred to Christ’s College at Cambridge to approach. He acquired a huge collection of speci- study for the church. He was trained to become mens. He was assigned an assistant, Syms an Anglican clergyman and graduated 3 years Covington, to help skin and clean birds and mam- after beginning his studies (Hergenhahn 2009 ). mals. They would also sort and pack shells, Darwin maintained interests in subjects other plants, bones, rocks, and fossils for shipment than the clergy while at Cambridge, such as ento- back to Henslow (Berra 2009 ). mology. This opened the door for networking Darwin became interested in the application with professionals in other fi elds. One such con- of transformation to humans and the perceived tact would send Darwin down a path that led to ability of human beings to change. This was fun- the foundation of his theories and infl uenced his damental to his later conversion to evolutionary career more than any other (Berra 2009 ). John theory. He was intent on identifying human Henslow was a botanist who was impressed by beings’ place in the natural world (Browne 1995 ). Darwin. Henslow was everything Darwin hoped After departing from South America, the for in a scientifi c mentor, even to the point of Beagle headed west toward the Galápagos teaching Darwin to use technical instruments Islands. Darwin visited four of the 16 major (Browne 1995 ). This mentor-mentee relationship islands (Berra 2009 ). He studied huge tortoises, eventually led to a recommendation by Henslow lizards, sea lions, and 13 species of fi nch for Darwin to be invited to participate as an (Hergenhahn 2009 ). He was especially fascinated unpaid naturalist aboard the ship Beagle (Browne with identifying subgroups between specimens 1995 ). Against his father’s objections and an from island to island. It was not until the Beagle actual list of reasons stating why it was not a sailed to the Galápagos Islands that he was able good idea, he could not refuse (Berra 2009 ). to confi rm his belief that most animals and plants Despite becoming very ill upon setting sail, spread only as far as geographical barriers Darwin considered his voyage on the Beagle as allowed. With geographical barriers in place, “my second life.” The setting off of the Beagle was Darwin was able to separate subfamilies of ani- “a birthday for the rest of my life” (Darwin mals based on physical features (Browne 1995 ). 1892 /2000, p. 120). This trip was important It was during this voyage that Darwin’s idea of because it gave him a chance to sharpen his skills becoming a clergyman disappeared. He did not of observation and description of people and reject God, at least at fi rst, by still giving a spiri- places. Darwin was under the impression that he tual meaning to the natural wonders he saw on would be gone for no more than 3 years depending the Beagle (Browne 1995 ). The Beagle continued on the weather (Browne 1995 ). The mission was to on to Tahiti, New Zealand, and Australia survey South America, take longitude readings, (Hergenhahn 2009 ). The trip ended up being 5 and make geophysical measurements (Berra 2009 ). years long. Darwin was eager to return home. Based on detailed logbooks and journals When the Beagle docked in Falmouth, England, Darwin kept throughout the entire journey, it is Darwin was one of the fi rst to get off (Browne evident that it was his nature to think big and 1995 ). 7 The Life and Evolution of Early Intelligence Theorists: Darwin, Galton, and Charcot 95

Due to interest among London’s naturalists, breadth of ideas and information (Browne 1995 ). Darwin believed that his primary scientifi c focus, Darwin delayed publication of his theory of evo- following his time on the Beagle, should concern lution for more than 20 years. There is some evi- the fossils he located. It was not until later that dence to suggest that Darwin even intended on he focused on his animal specimens (Browne having his work published after his death. He 1995 ). Perhaps a turning point in his research wrote his wife Emma a letter that discussed a was when he began asking questions regarding preference to be dead rather than suffer the con- the distribution of species among the Galápagos. troversy of his writings, which he thought would For Darwin, the possibility of a common ances- lead him and Emma to social exclusion (Browne try based on specimen’s physical resemblances 1995 ). An external motivator could have been drew his attention from the fossils. He eventually responsible for him publishing his theory concluded that everything, including mankind, (Hergenhahn 2009 ). was linked to one ancestral chain through trans- Darwin received a letter from Alfred Wallace, mutation (Browne 1995 ). who was also infl uenced by Malthus’s essay. In Darwin’s fi rst attempts at evolutionally theory the letter, he described a theory of evolution needed refi ning. His earliest evolutionary ideas almost identical to Darwin’s. Yet, because of the were disorderly, scattered, and seemed to ramble. abundance and meticulousness of data collected They primarily circled around the question of by Darwin, the theory of evolution is attributed to how transmutation might actually work. Despite him, not Wallace (Browne 1995 ). being more of a question than a theory at this Prior to receiving the letter from Wallace, point, transmutation engulfed Darwin’s time and Darwin had already written a quarter of a million efforts (Browne 1995 ). At this time, his ideas words in his “big book,” which he expected were different than would be put into print years would fi ll three volumes. After the letter, Darwin later. He possessed the pivotal idea of change in ceased work on the “big book” and focused on living beings and of real links to the ancestors of what he called an “abstract” of his species theory animals and mankind. However, he had yet to (Berra 2009 ). This would eventually become On include one piece that could tie it all together, the Origin of Species ( 1859 ). Darwin proposed natural selection. that natural selection operated on a living being This needed principle to tie everything as if it were a statistical necessity, a law of nature together came from An Essay on the Principle of stripped of divine infl uences, relentlessly honing Population (1798/ 1914) by Thomas Malthus. An animals, plants, and humans in the struggle for economist, Malthus observed that food supply existence. Organisms either adapted or died increased arithmetically, while the human popu- (Browne 2002 ). Darwin was not able to use the lation tended to increase geometrically term “evolution,” at that time. The term was (Hergenhahn 2009 ). Furthermore, from Malthus’s mostly used to describe the embryological pro- essay, Darwin recognized that differential death cess of a gradual unfolding of hidden structures. rates gave him a mechanism for explaining the It was the ensuing debate around his published preservation of adaptations, which until then was work that changed the word to its modern mean- a mysterious quantity (Browne 1995 ). ing (Browne 2002). In On the Origin of Species , One thing was clear, the Darwin who set sail Darwin referred to “descent with modifi cation” on the Beagle years earlier would likely have had ( 1859, p. 236). He suggested that a struggle for a diffi cult time formulating these ideas. The col- survival of all living organisms is related to the lection of his personal ambitions, social back- reproductive capacity being greater than the envi- ground, behavior, desires, and shortcomings ronment can sustain. Among offspring there are contributed to the development of his ideas as infi nite individual variations, some of which are much as solid scientifi c reasoning and research. more conducive to survival than others. This The range of his intellectual experience was results in the survival of the fi ttest; survival most impressive. His notebooks demonstrated a results in a natural selection among the offspring 96 J. Rigby of a species. The individual characteristics that It was in The Descent that Darwin laid out are naturally selected are theorized to be more how evolutionary theory explains the intelligence adaptive and have more of a survival value than of mankind. He suggested that behaviors com- those that are not passed down through offspring. monly associated with modern intelligence Fitness is measured by a species’ ability to sur- developed from the primal instincts of nonhuman vive and reproduce. Those organisms possessing ancestors. This led to human intelligence and ani- adaptive features are fi t, while those that do not mal intelligence being differentiated by degree, are unfi t. Darwin believed that this process just not by type. Darwin believed that the intellectual happens without direction or purpose involved. abilities of man, and their variations, were inher- This process is dependent on the environment in ited. It was to this intellectual ability that man- which the species exists (Browne 2002 ). kind owed for its key position in the world On the Origin of Species was met with both (Darwin 1871 /1896). Darwin offered examples criticism and praise. Darwin’s theory was attrib- to support the theory that humans and animals uted to the creation of a new science and philoso- share many cognitive attributes like wonder, curi- phy. Not until this time had a new branch of osity, long-term memory, the ability to pay atten- human knowledge been due to the fruits of one tion, to imitate the behavior of others, and to researcher (Berra 2009 ). However, Darwin’s the- reason (Darwin 1871/1896). He suggested that ory in On the Origin of Species was criticized, civilized nations would have a tendency to even to the point that its publication ended some increase in the number and in the standard of the of Darwin’s friendships (Berra 2009 ). It has been intellectually able. There is an innate tendency suggested by some that Darwin’ s theories con- toward continued development in mind and body. tributed to Captain Fitzroy, captain of the Beagle , This leads to the more intelligent genes within taking his own life 6 years after the publication of the same community being able to succeed in the Darwin’s theory because he felt at least partially long run in leaving more offspring, resulting in responsible for its development (Hergenhahn the inferior genes being counted out of the next 2009 ). Despite a mixed welcome, Darwin’s pub- generation. This is a form of natural selection lisher, John Murray, received 1,500 orders when (Darwin 1871 /1896). it was released (Browne 2002 ). On the Origin of Not all of Darwin’s contributions to intelli- Species went through six editions and has been gence theory were positive. He entertained a translated into 29 languages (Berra 2009 ). number of mistaken beliefs by today’s standards, After the publication of On the Origin of one of which involved the intellectual ability of Species, Darwin continued to publish in spite of a men compared to women. He believed that chronic physical ailment. He turned his attention women were intellectually subordinate to men to botany for a time. It was not until Darwin’s (Alland 1985 ). He also suggested that contempo- book The Descent of Man, and Selection in rary or primitive people were the link between Relation to Sex ( 1871 /1896) that Darwin began to primates and modern humans and were, there- build a foundation that humans are also the prod- fore, inferior (Hergenhahn 2009 ). uct of evolution. He suggested that both humans Late in the year of 1881, Darwin began to suf- and apes descended from a common ancestor. fer from chest pains which progressively got This was deliberately omitted from On the Origin worse. Despite the best efforts of various doctors, of Species . Also, in The Descent , Darwin used the he died on April 19, 1882, following a heart term “evolution” for the fi rst time in the modern attack the night before where he briefl y lost con- sense (Berra 2009 ). He was interested in “human sciousness. He was 73 years old. Primarily he variation, geographic diversity, facial expres- was to be buried in a churchyard next to family sions, moral sensibilities, inheritance, reproduc- members near Downe, the small village just out- tive behavior, and sexual selection” or the entire side London where he and Emma had been living natural history of mankind (Browne 2002 , for the past 40 years; however, at the request of p. 307). the Parliament, a funeral was held at Westminster 7 The Life and Evolution of Early Intelligence Theorists: Darwin, Galton, and Charcot 97

Abbey. Notable men from many scientifi c, natu- the founders of the Quaker religion. The family ralist, government, and philosophic circles were primarily made its money manufacturing guns, in attendance (Browne 2002 ). All of his surviving but later through investment banking. His father children attended, as well, but not Emma. His also joined the Church of England (Fancher grave is a few feet from the resting place of Isaac 1985 ). Newton and Charles Lyell. Being buried next to Despite this lineage and a fl are for academics, others who had such an impact on science is Galton was not really prepared with professional appropriate since Darwin’s theory was just as or substantial education, despite his sister’s best impactful. His ideas are attributed to changing efforts. His natural inclinations for curiosity and the history of philosophy and psychology by intellectual restlessness were considered dis- infl uencing traditional views of human nature. tracting and negative qualities when he entered He inspired others to not just look at the mind but formal British education. For much of his aca- behavior for individual differences. This had a demic career, he was considered mediocre. This direct infl uence on the introduction of new did not stop him from viewing his education as a schools of thought such as functionalism and competition among classmates. After fi nishing behaviorism, where individual differences and his second year of medical training at 18 years measurements were emphasized (Hergenhahn old, he wholeheartedly believed that he was 2009 ). going to fi nish fi rst on the fi nal medical examination. He only won a Certifi cate of Honor instead of the top prize (Fancher 1985 ). Perhaps too Francis Galton painful to admit the defeat, Galton incorrectly remembered fi nishing fi rst on this examination Francis Galton was born February 16, 1822, to a in his autobiography written 70 years later wealthy and distinguished English family. He (Galton 1908 ). Throughout his career at was the youngest of seven children. It was not Cambridge, he did just well enough to keep his long before Galton began to demonstrate an hopes alive to win honors in mathematics. It was intellectual brilliance. By 18 months, he was during this time he became interested in the ben- familiar with the alphabet. He began reading efi ts of examinations themselves. The idea of an books by 2 years old. At age 5, he was able to instrument that was able to distinguish students recite a Sir Walter Scott poem (Gillham 2001 ). at the top from the rest of the group was an Galton would impress visitors by reading untapped resource as far as Galton was con- Shakespeare as early as 6 years old (Fancher cerned (Fancher 1985 ). A measurement that 1985 ). He was instructed by his older sister, 12 would identify students who stood above others years his senior, in French, Greek, and Latin. academically would be useful. His desire that Galton greatly enjoyed school. Before long he exams would place him in a superior category in was doing so well that he made head boy, despite the area of mathematics was never realized. The being younger than most of the other children. collection of his work earned him the equivalent One teacher reported, “the young gentleman is of a B + average at one of the most competitive always found studying the abstruse sciences” universities in the world (Fancher 1985 ). This (Pearson 2011 , p. 67). Galton developed a sense was not the outcome that Galton desired or antic- of brightness while young. He was motivated to ipated. This crushed his dream to be a top excel in academic competition. Wrangler , a title reserved for students at The subject of science, among other things, Cambridge who earned the top scores on each was important to the family. Galton’s mother was year’s mathematics honors examination. Shortly the daughter of Erasmus Darwin, who made his following this realization, he withdrew from name as a poet, physician, and early evolutionary future exams and studying mathematics alto- theorist. Charles Darwin was the older half gether. This was followed by what some con- cousin to Galton. Galton’s father descended from sider an emotional breakdown (Fancher 1985 ). 98 J. Rigby

The last year of Galton’s time at Cambridge intellectual capacity and physical capacity. He was fi lled with dinner parties, balls, and wine argued that with physical training, physical parties. By this time he was an attractive young improvement could be expected. However, there man with blond hair, high forehead, pale blue is a genetic limit or ceiling when improvement eyes, and v-shaped mouth (Gillham 2001 ). The stops, no matter the time or effort devoted to end of Galton’s academic career was not what he training. Galton said the same about intellectual nor his disappointed family had envisioned. He ability: was only able to earn a “poll” or ordinary degree, This is precisely analogous to the experience that much like his cousin Darwin. With his formal every student has had of the working of his mental academic career over, he failed to prove himself power. The eager boy, when he fi rst goes to school the genius both he and his family envisioned. and confronts intellectual diffi culties, is astonished at his progress. He glories in his newly developed Galton earned a substantial inheritance when mental grip and growing capacity for application, his father passed away in 1844. After several and, it may be, fondly believes it to be within his years drifting and living on his inheritance, he reach to become one of the heroes who have left consulted a professional phrenologist for a per- their mark upon the history of the world. The years go by: he competes in the examinations of school sonality profi le that would steer him in the right and college, over and over again with his fellows, direction (Gillham 2001 ). The professional and soon fi nds his place among them. He knows assessed that Galton’s skull, with other informa- that he can beat such and such of his competitors; tion, suggested men with his type of intellect do that there are some with whom he runs on equal terms, and others whose intellectual feats he cannot distinguish themselves in universities. Galton not even approach. (p. 13) turned his attention to Africa. He organized a major expedition through South-West Africa that In Galton’s history and experience, he took him away for a couple of years. This would observed the differences between those who had lead to the creation of one of the fi rst maps of the similar environments. This appeared to contrib- area. Similar to Darwin’s voyage on the Beagle , ute to his not completely eliminating the infl u- Galton’s travels greatly contributed to the forma- ence of environment on intellectual development, tion of his theories (Fancher 1985 ). but heavily favoring the role of innate abilities Upon returning home to England from Africa, (Fancher 1985 ). Galton was different. He was matured and expe- Galton was also infl uenced, like much of the rienced by adversity (Gillham 2001 ). This change scientifi c world, by Darwin’s On the Origin of motivated him to work on what he referred to as Species. Despite having no real interest in biol- the “human side of geography” while also work- ogy, at least not to the degree of his cousin, he ing on informative books for travelers (Pearson was very interested in evolutionary theory as it 2011 , p. 129). He was impressed by the diversity applied to the diversity of humans. This was the between different African groups. This aware- root of his own ideas that took on great signifi - ness of ethnic diversity produced several major cance and occupied the majority of his life’s ideas which today are at the heart of the nature- work (Fancher 1985 ). Darwin’s writing “made a nurture and IQ controversies (Fancher 1985 ). marked epoch” in how Galton understood the Despite his successes, a question continued to mind (Galton 1908 , p. 287). plague Galton. Coming from a family of privi- Galton was excited by Darwin’s idea that while lege and having so many academic opportunities, unfavorable variations in breeding populations he had trouble explaining how his environment were not passed on, favorable variations would could not produce the genius most thought he increase in frequency over generations. Darwin would become. He was told by the phrenologist applied this theory to physical characteristics. that despite these advantages, it was insuffi cient Using this same theory, Galton thought that per- innate ability that he lacked. He would later agree haps psychological differences were inheritable with this theory in his book Hereditary Genius and based on small variations in the brain and ner- (1869 ), when he drew a resemblance between vous system. Adding to his already established 7 The Life and Evolution of Early Intelligence Theorists: Darwin, Galton, and Charcot 99 belief that ability was innate, the idea of inherit- sons, and brothers. Talents were also not evenly ability took on a whole new meaning with practi- distributed among the sample. There was a pro- cal applications (Fancher 1985 ). At the time, pensity for an unusual talent to run in a family readers were just beginning to consider the uncon- (Galton 1869). Again, expanding on Darwin, ventional thought of physical and some mental Galton determined that these heredity similarities traits being inherited from generation to genera- could be put into the same selection process as tion in animals. Many were unprepared to con- physical similarities. His reported results, sider that theory being applied to humans (Gillham although very general and not conclusive, were 2001). What Galton was proposing was that consistent with his hypothesis that differences in human talent and character changed similar to the ability are inherited. Galton understood the limi- domestic animals and cultivated plants discussed tations to his research alluding to there being no by Darwin (Gillham 2001 ). Humans were subject advantage as benefi cial as a good education to natural selection. Galton began to see the (Fancher 1985 ). He suggested behind heredity, advantages in this process by enhancing qualities families shared environmental circumstances. through selective breeding. Galton fi rst made Given his own academic experience, it is under- these thoughts public in a two-part article pub- standable how he would minimize the impor- lished in Macmillan’s Magazine in 1865. These tance of education. series of papers were the precursors to Galton’s In an effort to establish empirical support for 1869 book Hereditary Genius. Darwin tended to his ideas, Galton presented two research tech- agree with Galton stating, “I am inclined to agree niques still used today. First, he utilized the role with Francis Galton in believing that education of adoptive families who have shared environ- and environment produce only a small effect on ments but not genetics. Critics at the time had the mind of anyone, and that most of our equali- issues related to his sample size being truly repre- ties are innate” (Gillham 2001 , p. 155). sentative. Despite these limitations, adoptive In Hereditary Genius Galton compiled a sam- families were useful in identifying hereditary ple of those whose reputation suggested they had similarities (Fancher 1985 ). The other technique shown unusual talent in their lives. He did not employed involved twins. He understood the include those who were considered brilliant due potential both dizygotic twins and monozygotic to ancestry, like hereditary aristocracy (Fancher twins have in understanding the environment’s 1985 ). He used two systems to classify the men’s versus nature’s infl uence on development. With ability. First he classifi ed men by reputation, then his data in hand, he applied the bell curve as an by their performance on examinations. However, evaluative technique. he was unable to apply the examination scores to With the ground work laid by Hereditary a normal distribution. So this was abandoned, Genius, Galton could move forward with his and the fi rst classifi cation method was the only eugenics program. He was passionate about this. inclusion criteria used (Gillham 2001 ). He would It became a matter of moral and civic obligation. use men’s reputations to measure heritable abil- He made efforts to improve the intellectual stan- ity. By utilizing statistical means to analyze the dard of humans, not through educational or envi- data, a somewhat novel approach for the subject ronmental reform but genetics (Fancher 1985 ). at the time, he determined that his sample repre- His mission to accomplish this was twofold: fi rst sented a very small percentage of the normal the spread of intellectually and psychologically population, about one person in 4,000 (Galton superior humans through reproduction of off- 1869 ). After establishing the infrequency associ- spring; second the establishment of laws and culated with his sample in the general population, he tural customs to encourage superior humans to cross-checked their lineage and found 10 % of fl ourish at a faster rate than those deemed infe- them had at least one close relative in the sample. rior. In doing so, these favorable genes would be Immediate family was much more common than more numerous in society and would increase in distant relatives. The sample was full of fathers, quality through natural selection. Just before his 100 J. Rigby death he fi nished a novel, Kantsaywhere, about a the various mental and bodily powers” (Gillham eugenic utopia. Although in name only, Galton’s 2001 , p. 211). Galton even wrote to the leading legacy of eugenics will be forever tied to such psychologists in England to allow him to display programs as the Nazis encouraging racially their names to what he had in mind. However, appropriate women to have larger families to much of the experimental psychology and sen- increase the number of desirable individuals. sory perception work was being done in Germany This cruel practice of Nazi policies, ultimately by men like Herman Helmholtz and Wilhelm resulting in the “fi nal solution” and the holocaust, Wundt (Gillham 2001 ). Demographics were led to a total revolt counter to eugenics after gathered, and participants in the lab took various World War II (Gillham 2001 ). Although Galton tests in sequential fashion over the period of created the term “eugenics” and campaigned for about a half hour. They were measured with vary- it, he was simply generalizing Darwin’s theory of ing devices assessing “Keenness of Sight and of evolution through natural selection to mankind. Hearing; Colour Sense, Judgment of Eye; He would have likely been shocked to hear that Breathing Power, Reaction Time; Strength of only 20 years after his death, the name of eugen- Pull and Squeeze; Force of Blow; Span of Arms; ics was being used for forcible sterilization and Height, both standing and sitting; and Weight” murder (Gillham 2001 ). (Gillham 2001 , p. 213). To measure the strength It was Galton’s intention to identify prominent of a blow, for example, Galton used a rod made of or eminent individuals so that they could pass on fi r protruding out of a tub so that it could move their superior genes to offspring. These individu- freely. A buffer to prevent the participants from als were prominent due to their numerous posi- hurting his or her hand after smashing their fi st tive contributions to society. However, the timing down was attached. Following an impact, a of this societal status was customarily specifi ed pointer attached to the rod measured the distance after prime childbearing age. This was a problem the rod traveled in the tube (Gillham 2001 ). for Galton. How was he going to identify those Primarily, Galton avoided measuring head size. prominent individuals before this prime age He felt that women’s heads were too diffi cult to passed? A measure was needed to recognize a measure “on account of their bonnets, and the young adult’s ability. After being measured, this bulk of their hair” (Gillham 2001 , p. 213). This natural ability could be used as a predictor of practice was later included in the battery of who would make eminent contributions later in measurements. life. Those deemed favorable to make these types The Anthropometric Laboratory was a big of contributions would then be encouraged to success despite many of the tests never really intermarry and produce offspring (Fancher 1985 ). working as intended. Many failed to correlate Galton was holding on to his basic belief that with any independent signs of accomplishment there is a measurable difference between indi- or intelligence (Fancher 1985). But, by its end in vidual’s brains and nervous systems that can be 1885, statistics had been collected on 9,337 peo- differentiated. He went on to design a series of ple, each measured in 17 different ways (Gillham measures for sensory perception, reaction time, 2001). Each person was paid three pence to par- and physical energy (Gillham 2001 ). The goal ticipate in the experience (Fancher 1985 ). was to identify differences in neurological effi - Participants even returned for repeated measure- ciency or what he would refer to as natural ments. Before its closing Galton began collecting intelligence. fi ngerprints. This was a newly acquired interest In order to gather data, he set up, equipped, that turned into a major branch of his work. He and maintained the Anthropometric Laboratory also studied mental imagery as a quantifi able at his own expense. It was Galton’s desire “to measure of the human mind. Based on one’s compile a list of instruments suitable for the out- “imagery ability,” they could be compared to fi t of an Anthropometric Laboratory especially others. From this came the fi rst word associa- those for testing and measuring the effi ciency of tion tests. It is not known whether or not this 7 The Life and Evolution of Early Intelligence Theorists: Darwin, Galton, and Charcot 101 infl uenced Sigmund Freud, but Galton was with long black hair tossed back and a small utilizing concepts of psychoanalysis (Hergenhahn black mustache who did not get out to socialize 2009). Galton collected an enormous amount of much. His program was fi lled with students from data that required analyzing. In doing so, he across the globe: British, American, German, and developed his concept of regressing, invented the Austrian. This program involved a fi xed yearly correlation coeffi cient, and proposed one of his curriculum, forfeiting traditional grades for a most important and infl uential books, Natural series of year-end tests to graduate. Clinical com- Inheritance (Galton 1894 ). The Anthropometric petency or practical skills in medicine were not Laboratory had also stirred up an interest in test- weighed heavily on these fi nal examinations. ing ability as a new area of research. As a result, Theoretically, it was possible to pass all medical mental heredity began to become commonly examinations without actually interacting with a known and accepted (Fancher 1985 ). patient. Hospitals were used for teaching and part Sir Francis Galton died on January 17, 1911, of the curriculum, but much of what was offered after approximately 5 years of failing physically was only taken advantage of by the more asser- but remaining mentally whole. Throughout his tive students. One year of clinical rotations was life, he was involved in exploration and travel required the year Charcot began his training writing. He was largely responsible for the devel- (Goetz et al. 1995 ). opment of fi ngerprinting as a forensic method. This was not suffi cient for Charcot. He was Galton’s contributions to mental imagery infl u- interested in pursuing an internship. At the time, enced psychology, along with methods in pedi- this position was optional and highly competi- gree analysis and twin studies. He introduced the tive. Only the elite students were selected for this statistical method of correlation and regression, additional clinical experience. It took Charcot 2 resulting in the use of quantitative methods into years before he was accepted into this special social sciences (Gillham 2001 ). group. After his fi rst application, he was given the status of a temporary intern before becoming one of six interns. Internship consisted of four Jean-Martin Charcot yearlong rotations at different hospitals under different chiefs (Goetz et al. 1995 ). Born in Paris in 1825, Jean-Martin Charcot was Two of these supervisors had a particularly raised in a lively section of town that is now con- great infl uence on Charcot, P. A. Piorry and sidered the Right Bank, north of the Le Seine Pierre Rayer. Piorry did not study the nervous River. In Charcot’s time this was between the system, but his emphasis on anatomical localiza- central city and the countryside. He was the tion and nosology infl uenced Charcot. Piorry eldest of four sons. He did not go far to attend also taught Charcot the practical use of micros- grammar school at the Pension Sabatier. Among copy in medicine. The most infl uential supervisor a very standardized curriculum for all schools in Charcot’s early career was Pierre Rayer. He that included a national examination at the end of was a mentor to Charcot. As a result of Charcot each school year, he studied Greek and Latin in not having any familial connections into the preparation to transfer to the Collège Royal de closed scientifi c and social circles of medicine, Bourbon after graduation. Although he took the Rayer befriended him and aided in networking prerequisites to enter medical school, such as him into those closed groups. This was a practice Latin, he was not sure medicine was his calling. not lost on Charcot, as later in his career he pro- He was torn between medicine and art as a career tected and nurtured the careers of his protégés path. He eventually chose medicine and began one of which was Sigmund Freud. Freud was his training at the Paris Medical School in 1843 interested in studying children’s brains and was (Goetz et al. 1995 ). welcomed by Charcot. Freud offered to translate So began Charcot’s career in medicine. He text into German and was quickly accepted into was described at that time as a thin, pale man Charcot’s good graces. 102 J. Rigby

Charcot found himself in a very advantageous closer to his goal was to teach both outside and situation. He was being trained by important and within the medical program. infl uential physicians in what was the medical It was Charcot’s desire to return to the capital of the world. He was intelligent and pos- Salpêtrière. This is where he had begun to study sessed a unique capacity to work. But what was the nervous system and where he would become perhaps a catalyst to this “perfect storm” that per- known all over the world. Most physicians at the petuated him into international success was a time would rotate from hospital to hospital. determination to publicize his ideas. He was Charcot broke that pattern and established him- established in publications very early and main- self at the Salpêtrière for the majority of his tained his scientifi c involvement throughout his career. Given his work upon starting at the career. Salpêtrière, he may have been motivated by the Charcot took 9 years to graduate. This was the opportunity to conduct original research as he typical length taken before completion of medical made this an emphasis at the very beginning. He school (Goetz et al. 1995). He was awarded his focused on pathology, diseases of the nervous diploma in 1853. He received an exceptional grade system, and chronic rheumatism above the rest on his fi nal examination, which was a thesis on (Goetz et al. 1995 ). rheumatoid arthritis. He collected his data where Over the course of many years, Charcot was he completed his internship, at the Salpêtrière. able to overcome that which held him back from Overall, his grades ranged from “rather good” to a failed bid for a medical professorship in inter- “very good.” There is no evidence to suggest that nal pathology. A vote took place, the results of Charcot had to retake or failed any courses, a com- which were overwhelmingly not in Charcot’s mon occurrence for many average students who favor. He received only a handful of votes for one did not pursue internships. of two positions. However, the period following Following graduation Charcot accepted a posi- his defeat has been described as advantageous tion as a head of a clinic. There he worked for two due to his efforts for research and teaching at the years under his old supervisor, Piorry. This posi- Salpêtrière. This additional experience led to a tion was not highly sought after or lucrative but convincing victory years later. His research, spe- was a natural step-up in the academic hierarchy . cifi cally focusing on pathology of the nervous Following his clinic position he became a hospital system, and his style of teaching made him a very doctor for the next 4 years. This may have been strong candidate. He became chair of pathologi- disappointing due to few opportunities to study cal anatomy at the Paris Medical School in 1873, patients. He primarily handled outpatient triage in but still continued his work at the Salpêtrière. the central hospital admitting offi ce. Despite his Obtaining professorship when he did, not earlier position, Charcot worked toward building a last- or later, clearly contributed to his career in neu- ing reputation (Goetz et al. 1995 ). rology as he was later appointed the professor- His ultimate goal was to secure a position at a ship for the study of disease of the nervous Paris hospital with the medical faculty. He set out system in 1882. This allowed him to focus on to do this by occupying his free time with build- neurological issues. ing a private practice. A private practice allowed Charcot’s interests progressed from general Charcot to develop a positive reputation and medicine among a geriatric population to chronic helped fi nancially. He also continued to publish illnesses affl icting all ages. At the Salpêtrière and conduct research within medical scientifi c Charcot was responsible for groups classifi ed as societies. He managed to make signifi cant contri- rheumatological and neurological. In an effort to butions involving the subjects of pathology and explain the medical fi ndings in arthritic and neu- critical reviews. Even these earliest publications rological populations, Charcot developed an demonstrated his abilities as a capable, wide- anatomo-clinical method of research. This would ranging, and discerning consumer of the avail- become the framework of which all of Charcot’s able literature. The last thing he did to move anatomical studies in neurology were based. 7 The Life and Evolution of Early Intelligence Theorists: Darwin, Galton, and Charcot 103

This linked clinical signs with anatomical lesions. sparked ideas that resulted in hysterical sympto- Charcot utilized longitudinal observation to mology (Hergenhahn 2009 ). At the time, these meticulously document clinical signs. This data symptoms were often disregarded as malingering. was then cross-examined with the brain and spi- The majority of physicians were suspicious. It nal cord following an autopsy. He hoped to estab- was Charcot’s thought that this directly related to lish anatomo-clinical correlations by combining hypnosis given that symptoms of hysteria could clinical and anatomical data. Through sound be reproduced so readily in a hypnotic state. research he hoped to locate specifi c lesions which During a hypnotic trance Charcot could induce would correspond to specifi c clinical signs (Goetz hysterical body contractures. This gave evidence et al. 1995 ). This was the foundation for a new to his theory that hypnosis could expose the true anatomy-based classifi cation of neurological mechanisms of hysteria (Fancher 1985 ). Being pathology. Perhaps the best-developed example components of the same neurological state, the of this method was Charcot’s work with motor capacity to be hypnotized and the diagnoses of system degenerative disorders, specifi cally amy- hysteria were linked (Goetz et al. 1995 ). This was otrophic lateral sclerosis. These studies led to the something Charcot advocated, even near the end international description of amyotrophic lateral of this life. sclerosis as Charcot’s disease, now commonly Charcot, with Binet’s help, wrote case studies known as Lou Gehrig’s disease. Also, numerous that would illustrate these theories and highlight stroke syndromes and the relation of specifi c the complexity of real people. Charcot’s theories signs of multiple sclerosis to specifi c lesions are complicated the established distinction between the product of the anatomo-clinical method. This biological and psychological investigation, discipline pushed neurologists toward the notion allowing for further research focusing on intelli- of cortical localization theory, suggesting the gence and its relationship with neurophysiologi- brain controlled specifi c motor, sensory, and lan- cal and genetic factors. His theories did not go guage functions (Goetz 2010 ). unchallenged. Many claimed that any suggestible Charcot became increasingly interested in person could be hypnotized, independent of hys- studying hysteria and hypnosis. He employed the teria. This thinking made hypnosis available as help of volunteer researchers, one of whom was an intervention for diagnosis outside of hysteria. Alfred Binet. Questions being raised concerning Many clinicians interested in exploring hypno- these two conditions were motivation and uncon- tism, including Freud, tended to agree. As time scious psychological effects. He displayed a con- passed, Charcot’s views on hysteria and its rela- siderable amount of interest in the psychology tionship to hypnotism became increasingly diffi - side of hysteria, more than most of his colleagues cult to defend as new insights on the subject in medicine. Even psychiatrists of the day were emerged (Goetz et al. 1995 ). committed to somatic and hereditary assumptions Another contribution by Charcot was with dis- about mental diseases and had very little interest eases of the nervous system. Charcot made an in the possibility of a psychological etiology. argument that as an organism evolved to become What sparked Charcot’s interest in hysteria was more complex, the nervous system took on an that the symptoms so closely imitated known neu- ever progressively dominant role over all other rological pathology but were absent from a neuro- functions along with a close connection anatomi- logical or organic cause. After an exhaustive cally (Goetz et al. 1995 ). neurological exam exposed no pathology, those Although Charcot may have held others’ suffering from hysteria still presented with anes- health and well-being in the highest regard, he thesias, amnesias, or paralyses. He also noted a did not do the same for his own. He sustained a traumatic event prior to the onset of symptoms in high degree of intensity, both personally and several of the hysteric patients. Although not professionally, that likely attributed to his death. physiologically or neurologically damaging, He also ate heavily and smoked endlessly. He had Charcot suggested that these traumatic events personal interests, but he seemed to always 104 J. Rigby

incorporate some type of medical element. Browne, J. (1995). Charles Darwin: Voyaging . Princeton: Despite serious cardiac problems, Charcot never Princeton University Press. Browne, J. (2002). Charles Darwin: The power of place . cut back his work. He died at age 67 of pulmo- Princeton: Princeton University Press. nary edema early in the morning on August 16, Darwin, C. (1859). The origin of species: By means of 1893, at an inn near the Morvan area of France. natural selection or the preservation of favoured races Just before his death he wrote a letter to his wife in the struggle for life . John Murray. London: England, doi: 10.1037/13681-000 stating that he had not been feeling well. His Darwin, C. (1871/1896). The decent of man and selection death was news that quickly traveled worldwide in relation to sex. New York: D. Appleton and via medical journals and newspapers. By the time Company. of his death, Charcot was a well-respected con- Darwin, C. (1892/2000). The autobiography of Charles Darwin . Amherst: Prometheus Books tributor to the fi eld of neurology. Although much Fancher, R. E. (1985). The intelligence men: Makers of of his work with hysteria was imperfect and most the IQ controversy. New York: W.W. Norton & was replaced by others such as Freud, his obser- Company. vations contributed to the concept of treatment Galton, F. (1869). Hereditary genius: An inquiry into its laws and consequences . New York: D. Appleton and for psychosomatic conditions and the association Company. of psychology into the manifestation of nervous Galton, F. (1894). Natural inheritance . New York: illness (Goetz et al. 1995 ). His anatomo-clinical Macmillan. model was unique and was being developed as a Galton, F. (1908). Memories of my life . London: Methuen & Co. discipline all over the world. Charcot’s unique Gillham, N. W. (2001). A life of Sir Francis Galton: From personality and history directly infl uenced its African exploration to the birth of eugenics . New York: development and is still used to distinguish neu- Oxford University Press. rology from other medical disciplines. The clas- Goetz, C. (2010). Jean-Martin Charcot and the anatomo- clinical method of neurology (Chapter 15). In sifi cation system that Charcot created for defi ning Handbook of clinical neurology (pp. 95203–95212). the brain and nervous system and muscle dis- doi: 10.1016/S0072-9752(08)02115-5 . eases to this day can be found in modern neuro- Goetz, C. G., Bonduelle, M., & Gelfand, T. (1995). logical classifi cations. For this reason, he will Charcot: Constructing neurology. New York, NY: Oxford University Press. forever be considered the “founder of modern Guilford, J. P. (1967). The nature of human intelligence . neurology.” New York: McGraw-Hill. Hergenhahn, B. R. (2009). An introduction to the history of psychology (6th ed.). Belmont: Wadsworth/ Thomson Learning. References Malthus, T. (1914). Essay on the principle of population . New York: Dutton. (original work published 1798). Alland, A., Jr. (1985). Human nature: Darwin’s view . Pearson, K. (2011). The life, letters and labours of Francis New York: Columbia University Press. Galton: Correlation, personal identifi cation and Berra, T. M. (2009). Charles Darwin: The concise story of eugenics . New York: Cambridge University Press. an extraordinary man . Baltimore: The Johns Hopkins Stefoff, R. (1996). Charles Darwin: And the evolution University Press. revolution . New York: Oxford University Press. Social Competition and the Evolution of Fluid 8 Intelligence

David C. Geary

The hallmark of human competence is the ability of novelty was climatic change, the nuances of to modify evolved brain and cognitive systems to hunting other species, or from the dynamics of create evolutionarily novel abilities that in turn human cooperation and competition. undergird much of culture as we experience it It is of course possible that all of these factors today. The modifi cation of these primary, evolved contributed, perhaps at different points during systems and the cross-generational accumulation hominid evolution. The key question concerns the of these advances have resulted in writing sys- selection pressure that was central to the evolution tems, mathematics, science, a myriad of techno- of these competencies, following the emergence logical advances, as well as literature, art, and of the genus Homo and especially the emergence many other features of the modern world. General of H. sapiens. I fi rst provide respective reviews of fl uid intelligence is critical to the creation and research on hominid brain evolution and psycho- learning of these non-evolved abilities and repre- metric and cognitive neuroscientifi c research on sents an evolved system itself (Geary 2005 ). general intelligence. These reviews provide the Proposals regarding how intelligence evolved background needed to appreciate the substantive emerged soon after Darwin’s ( 1859) On the changes that have occurred in the brain and the Origin of Species (Darwin 1871 ; Huxley 1863 ) mind during human evolution and to understand and in many respects anticipated the central argu- the component competencies that defi ne intelli- ment of modern-day theorists (Alexander 1989 ; gence. I close with an attempt to integrate these Ash and Gallup 2007 ; Dunbar 1998 ; Flinn et al. fi ndings with models of climatic, ecological, and 2005 ; Geary 2005 ; Kaplan et al. 2000 ; Mithen social selection pressures and within an integra- 1996). The central theme is that intelligence rep- tive model of human brain, cognitive, and psycho- resents the ability to anticipate and predict varia- logical evolution (see Geary 2005 ). tion and novelty that arise within lifetimes (as contrasted with cross-generational change) and to devise strategies to cope with this novelty. Evolution of the Hominid Brain Theorists differ largely on whether the key driver Brain Volume and Organization D. C. Geary , Ph.D. (*) Department of Psychological Sciences, The mean brain volume estimates for key spe- Interdisciplinary Neuroscience Program, cies of Homo and the predecessor genus University of Missouri , 210 McAlester Hall , Columbia , MO 65211-2500 , USA Australopithecus are shown in the top portion of e-mail: [email protected] Fig. 8.1 (Asfaw et al. 1999 ; Falk et al. 2000 ;

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 105 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_8, © Springer Science+Business Media New York 2015 106 D.C. Geary

Fig. 8.1 The top portion presents the estimated brain may be earlier and later forms of the same species. The volume for chimpanzees (Pan troglodytes ) and key spe- bottom portion presents estimated encephalization quo- cies of australopithecines ( A. afarensis , A. africanus , A. tients (EQ) for these species as a percentage of the EQ of garhi ) and Homo (H. habilis , H. ergaster , H. erectus , H. modern humans ( H. sapiens ) (Adapted from Geary 2010 , sapiens ). The middle portion presents estimated time of p. 136. Copyright 2010 by American Psychological existence of these species; H. ergaster and H. erectus are Association)

Holloway 1973 ; McHenry 1994 ; Tobias 1987 ; 1982 ). Expansion of these areas is potentially Wood and Collard 1999 ). The australopithecines important, because as they are implicated in stud- ( A. afarensis , A. africanus , and A. garhi ) had a ies of ﬂ uid intelligence, working memory, and modestly larger brain volume than modern-day self-awareness (Jung and Haier 2007 ; Levine chimpanzees (Pan troglodytes ). Further increases 1999 ; Tulving 2002 ). The area of the frontal lobe in brain volume are evident in H. habilis (Tobias involved in human speech and gesture (i.e., 1987 ) and with the emergence of H. erectus to Broca’s area) appears to have expanded with the modern humans (McHenry 1994 ; Wood and emergence of H. habilis and had architecture Collard 1999 ). Casts of the fossilized crania pro- similar to that of modern humans. vide impressions of the inner surface of the brain The threefold increase in brain volume from and suggest that the australopithecine brain, the australopithecines to modern humans is of including the shape of the posterior portions of course substantial but potentially misleading. the left hemisphere and anterior portions of the The absolute size of the brain increases with right hemisphere, differed from that of chimpan- increases in overall body size and thus confounds zees (Holloway and Kimbel 1986 ). The visual cross-species comparisons. An adult male A. afri- area of the left hemisphere is smaller than canus likely weighed 30 % less than a modern expected based on body and brain size, and the adult human male, suggesting some proportion left parietal area (e.g., area 39) and right frontal of the difference in brain volume is due to differ- pole (area 10, Fig. 8.2 ) area are larger than ences in overall body size (McHenry 1994 ). The expected (Holloway and de al Coste-Lareymondie encephalization quotient (EQ) is an often- used 8 Social Competition and the Evolution of Fluid Intelligence 107

Fig. 8.2 Cartoon maps of Brodmann’s ( 1909 ) areas of the complex auditory and speech perception; areas 20, 21, human neocortex. The top section is a lateral (outer ) view 26–28, 34–37, and 52 are also part of the temporal lobe of the cortex, whereas the bottom section is a medial (cen- but support a variety of complex visual competencies; ter ) view. Many of these areas can be subdivided into spe- areas 4, 6, and 8 are involved in complex motor move- cialized subregions that may process different forms of ments and are part of the frontal cortex; Area 44 and sub- information. Generally, areas 1, 2, 3, 5, 31, and 43 are part regions of area 45 are involved in speech generation and of the parietal cortex and support a variety of functions are part of the frontal cortex; areas 9, 10, 11, 25, 46, 47, including sense of body position, attention, and spatial and subregions of 45 are part of the prefrontal cortex and competencies; areas 17, 18, and 19 are part of the occipital support behavioral control, executive functions, and many cortex and support simple and complex visual perception; complex social competencies; areas 23, 24, 30, (parts of areas 22, 41, 42, and subregions of areas 40 and 38 are 31), 32, and 33 are part of the cingulate and support atten- part of the temporal cortex and support simple and tional and emotional functions measure to control for the relation between brain The EQ of chimpanzees is about 2.0 and about and body size; EQ indexes brain size relative to 6.0 for modern humans (McHenry 1994). The bot- that of a mammal of the same body weight tom portion of Fig. 8.1 shows EQ estimates that are (Jerison 1973 ), with an EQ of 1.0 indicating an expressed in terms of a percentage of that of modern average brain size. humans, and thus although the EQ of chimpanzees 108 D.C. Geary is twice that of the typical mammal, it is estimated Falk et al. 2000 ). Other theories have focused on to be only 34 % of that of modern humans. The EQ the benefi ts to being able to predict seasonal of australopithecines is greater than that of chim- changes and prepare for climatic variation (Potts panzees but less than 50 % that of modern humans, 1998 ). Some studies do fi nd a correlation between with large increases in EQ evident with the emer- estimates of variability in global temperature and gence of H. ergaster and H. erectus . Following the changes in brain volume and EQ from H. habilis emergence of H. ergaster, there was a period of lit- to early modern humans (Ash and Gallup 2007 ). tle change in EQ for about 1.2 million years ago Other studies, however, fi nd a smaller or no rela- (Ma), followed by a modest 12 % increase from tion once other factors are controlled (Bailey and 500,000 to 400,000 years ago, and then very rapid Geary 2009 ; Pearce et al. 2013 ). increases until about 20,000 years ago, and fol- Ecological models of hominid brain evolution lowed by a 3 % to 4 % decline (Holloway 1996 ). have focused on improvements in the ability to extract biological resources from the ecology and through this reduce premature mortality. The Climatic, Ecological, and Social basic idea is supported by comparative studies of Selection Pressures nonhuman species: Species with complex foraging or predatory demands have a larger brain vol- Climatic and geological changes are common and ume and higher EQ values than related species can result in variation in temperature, rainfall, and with less complex foraging or predatory demands other ecological conditions that in turn can infl u- (Barton 1996 ). The basic argument is that homi- ence the mix of vegetation, woodland, and other nids evolved into super predators (Wrangham factors that can have dramatic evolutionary effects et al. 1999 ). Humans in traditional societies are on species. Vrba (1995 ) argued that signifi cant indeed highly effi cient at extracting life- glaciation between 2.8 and 2.5 million years ago supporting resources from the ecology through resulted in decreased temperature and rainfall in hunting and foraging (Kaplan et al. 2000 ). If this Africa and substantial evolutionary changes in ability to extract and process (e.g., through cook- many species, including hominids. Evolutionary ing) biological resources was the driving force in responses to decreasing temperature often include hominid brain evolution, then there should be increases in physical size to enable retention of evidence of social, behavioral, and other adapta- body heat and a prolonged period of physical tions that allowed evolving hominids to forage development to enable growth of a larger body. and hunt in increasingly sophisticated ways. Vrba proposed that these physical adaptations There is some evidence that A. garhi con- resulted in an accompanying change in brain vol- structed stone tools, including tools used to cut ume and EQ: “The conclusion is inescapable that and process meat (Semaw et al. 2003 ). Moreover, hominine encephalization in the latest Pliocene features of tooth morphology and behavioral started a new trend, of higher evolutionary rates adaptation suggest that australopithecines were than before” (Vrba 1995 , p. 406). able to eat a wider range of foods than their pre- The implication is that the increase in EQ with decessors and thus were able to exploit a wider the emergence of Homo was not the direct result range of ecologies. Wrangham et al. (1999 ) of ecological or social pressures but rather an argued that H. erectus used fi re for cooking, incidental effect of physical adaptations to cli- which enables the use of an even wider range of mate change. Evidence for Vrba’s (1995 ) hypoth- plant and animal species as foods. Foley and oth- esis is mixed. The emergence of various species ers have demonstrated that tools related to hunt- of australopithecine 3.0 to 2.0 MYA may have ing and foraging (e.g., digging sticks) have been driven, at least in part, by climatic and eco- become increasingly sophisticated since the logical changes in Africa during this epoch, but appearance of A. afarensis (Foley 1987 ; Foley does not appear to be the primary evolutionary and Lahr 1997 ). H. habilis appears to have pressure since that time (Elton et al. 2001 ; used simple stone tools, with increases in the 8 Social Competition and the Evolution of Fluid Intelligence 109

complexity of these tools and their wide geo- Once ecological dominance was achieved, an graphic distribution coinciding with the emer- evolutionary Rubicon was crossed: gence and migration patterns of H. erectus . These the ecological dominance of evolving humans advances continued with the appearance of early diminished the effects of ‘extrinsic’ forces of natu- modern humans, with the most complex stone ral selection such that within-species competition tools found in archeological sites dating less than became the principle ‘hostile force of nature’ guiding the long-term evolution of behavioral capaci- 50,000 years ago (Foley and Lahr 1997 ). The pat- ties, traits, and tendencies (Alexander 1989 , tern of technological advances in tools and their p. 458). utility in hunting and foraging is consistent with ecological models of human brain evolution (e.g., Ecological dominance manifests as the ability Kaplan et al. 2000 ). to effi ciently extract biological resources from Patterns of human migration and subsequent the ecology, as described above, and the ability to mass extinctions of other species are consistent manipulate the ecology in ways – building of shel- with the evolution of a super predator. Wallace ters, use of fi re, clothing, and so forth – that reduce noted as much in 1911 (p. 264) – “the rapidity of mortality risks and supports subsequent popula- the extinction of so many large Mammalia is tion expansions (Hill et al. 2001; Kaplan et al. actually due to man’s agency.” Decades later, 2000). The lower mortality that would follow Martin ( 1973 ) presented systematic evidence from ecological dominance would have necessar- that indeed mass extinctions of many species of ily resulted in population increases, which would large prey were evident in Africa about result in pressures for migration to new regions, 50,000 years ago, and later mass extinctions as happened with H. erectus . Population increases occurred in Australia, Asia, America, and New also carry the risk of expansion beyond the carry- Zealand after the migration of humans into these ing capacity of the ecology, which results in regions (e.g., Alroy 2001), although some of fewer available resources per capita and an ensu- these extinctions appear to have been due to cli- ing increase in the competition for these dimin- matic or other ecological changes. In any case, ishing resources. The pattern was of course declines in many populations of large fi sh and described by Malthus (1798 ) and confi rmed in mammals have been directly linked to human subsequent studies of population dynamics in hunting and fi shing (e.g., Hsieh et al. 2006 ). developing Europe and developing nations today Consistent with these patterns is evidence that (e.g., Hed 1987 ; United Nations 1985 ). The end the evolutionary increase in brain volume was result is often a population crash that dispropor- associated with a corresponding decrease in the tionately affects individuals who are economi- mass of the metabolically expensive gastrointes- cally poor and of lower social status. The inverse tinal tract (Aiello and Wheeler 1995 ). Reduction relation between social status, resource control, in the size of the gastrointestinal tract requires and mortality risk creates a never-ending cycle change from a low-quality (e.g., plants) to a whereby Darwin’s and Wallace’s (1858 , p. 54) high-quality (e.g., fruits, meat) diet, as appears conceptualization of natural selection as a “strug- to have been the case during hominid evolution. gle for existence” becomes in addition a struggle It is not that reduction in the gastrointestinal with other human beings for control of the tract directly caused the evolutionary expansion resources that support life and allow one to repro- of brain volume and EQ but rather released a sig- duce (Geary 2005 ). nifi cant metabolic constraint, creating opportu- These dynamics set the stage for the within- nity for evolutionary change. species arms race, whereby cognitively and The gradual improvement of hunting and socially sophisticated individuals are able to out- foraging abilities, especially after the emergence maneuver and manipulate other individuals in of Homo, resulted in ecological dominance and order to gain control of local resources and to set the stage for a within-species evolutionary exert disproportionate infl uence over the lives of arms race (Alexander 1989 ; Flinn et al. 2005 ). other people. Much of the competition that might 110 D.C. Geary arise under these conditions is muted in modern Psychometric Intelligence democracies (Pinker 2011 ), but even so the relation between social status and health and well- In Spearman’s ( 1904 ) seminal study, children, being remains (Marmot 2004). Prior to the rise of adolescents, and adults were administered sen- these societies, resource control and social infl u- sory and perceptual tasks and were rated by ence had substantive effects on survival and teachers and peers on their in-school intelligence reproductive outcomes (Betzig 1986; Hed 1987 ; and out-of-school common sense. Exam scores Malthus 1798 ; United Nations 1985 ), and thus were also available for the adolescents. the supporting sociocognitive competencies and Correlations revealed that above-average perfor- brain systems would have necessarily evolved. mance on one task was associated with above- The gist is that heightened social competition, average performance on all other tasks, on exam following the evolution of ecological dominance, scores, and for ratings of intelligence and accommodates most of the core features of the common sense, leading Spearman (1904 , p. 285) climatic and ecological theories of human brain to conclude “that all branches of intellectual evolution (Alexander 1989 ; Flinn et al. 2005 ; activity have in common one fundamental func- Geary 2005 ), as well as other evolutionary mod- tion (or group of functions).” Spearman termed els that focus primarily on the importance of the fundamental function or group of functions social competition during human evolution (e.g., general intelligence or g . Dunbar 1998 ; Humphrey 1976 ). Decades later, Cattell and Horn (Cattell 1963 ; Horn 1968 ) proposed that general intelligence should be subdivided into crystallized intelli- Fluid Intelligence gence (Gc; learned knowledge and skills) and fl uid intelligence (Gf). The latter represents a bio- Across the various evolutionary theories, the logically based ability to acquire skills and basic defi nitions of intelligence include planning, knowledge. As Cattell (1963 , p. 3) stated: “Fluid foresight, scenario building, and so forth, but general ability … shows more in tests requiring seminal psychological research on intelligence is adaptation to new situations, where crystallized almost never mentioned. One goal of an earlier skills are of no particular advantage.” Cattell’s work was to attempt to meld the modular view of description of fl uid intelligence is consistent with the human mind associated with these various the gist of all of the various models of hominid evolutionary models with the more than 100 brain and cognitive evolution; specifi cally, this years of psychometric and more recent cognitive evolution was driven by selection pressures that neuroscience research on intelligence (Geary required humans and our ancestors to cope with 2005 ). The key hypothesis is that fl uid intelli- variation and novelty in their day-to-day lives. In gence evolved because of the advantages pro- other words, the ability to anticipate and cope vided by the ability to mentally represent potential with novelty and change that are central to changes in social conditions and to engage in models of human brain and cognitive evolution is explicit problem solving to devise behavioral reliably assessed by tests of Gf. strategies to cope with these changes. The evolution of fl uid intelligence probably accelerated after hominids achieved ecological dominance Cognitive Components of Intelligence (e.g., through tools, fi re, shelter) and the resulting uptick in the intensity of social competition (i.e., Hundreds of studies have focused on identifying within-species arms race). I summarize this the cognitive components of intelligence (Deary framework in a later section (see Geary 2005 for 2000 ; Jensen 1998 ). These efforts led to the full description). In the following, I provide a identifi cation of speed of information processing brief review of empirical studies of general and working memory as core components of intelligence. intelligence, especially Gf; many of these 8 Social Competition and the Evolution of Fluid Intelligence 111

studies include measures of Gc or those that tap systems or through access to overlearned infor- a combination of Gc and Gf and thus are termed mation, the result is an automatic shift in atten- “intelligence.” tion to this information (Botvinick et al. 2001 ). Situations that trigger attentional shifts are, by Speed of Processing this defi nition, novel or rapidly changing. The fi rst of three important fi ndings to emerge is Attentional focus results in an explicit repre- that faster speed of cognitive processing is related sentation of this information in working memory to higher scores on measures of intelligence and simultaneous inhibition of irrelevant informa- (Jensen 1982 ; Jensen and Munro 1979 ), although tion (Cowan 1995 ; Engle et al. 1995 ). Once active strength of the relation is moderate (r s ~ 0.3–0.4). in working memory and available to conscious The second is that individuals who are consis- awareness, the information is amendable to prob- tently fast in executing the same processes time lem solving. The attentional system that controls after time have the higher intelligence scores the explicit manipulation of information during (r s ~ 0.4; Deary 2000 ; Jensen 1998 ). Finally, the problem solving is called the central executive, speed with which individuals can identify very which operates on information in several repre- briefl y (e.g., 50 ms) presented information (e.g., sentational systems, including auditory, visual, whether “>” is pointed left or right) is moderately spatial, or episodic (Baddeley 1986, 2000). correlated with intelligence (Deary and Stough Episodic memory binds information from multi- 1996 ). ple systems and is important for recall of memo- The gist is that performance on psychometric ries of personal experiences (Tulving 2002 ). test of intelligence and especially Gf is related to Individual differences in performance on mea- the speed and accuracy with which information is sures of Gf are moderately (r s ~ 0.5; Ackerman identifi ed and then processed by sensory and per- et al. 2005 ) to strongly associated with individual ceptual systems. For individuals who score differences in working-memory capacity (r s > 0.8; highly on measures of Gf, the processing of this Conway et al. 2002 ). On the basis of these pat- information occurs more rapidly than for other terns, Horn (1988 ) and other scientists (Carpenter individuals. For all individuals, the information is et al. 1990 ; Stanovich 1999 ) have argued that fi rst implicitly (i.e., below conscious awareness) measures of strategic problem solving and abstract represented in short-term memory. For the infor- reasoning defi ne Gf, and the primary cognitive mation to become available to conscious aware- system underlying problem solving, reasoning, ness and thus amendable to explicit problem and thus Gf is attention-driven working memory; solving, it must become represented in working a proposal regarding the potential relation between memory. speed of processing and working memory is presented in Integration. Working Memory and Problem Solving The processing of the majority of information Summary Intelligent individuals identify, gathered by our sensory and perceptual systems process, and bind together bits of social and and represented in short-term memory occurs ecological information more easily and quickly automatically and implicitly. Typically, mental or than do other people. Their perceptual systems behavioral responses also occur automatically, process this information such that it is activated that is, without the need to engage attentional and quickly and accurately in short-term memory. If working-memory resources (Gigerenzer et al. evolved or learned heuristics are available for 1999). These automatic responses are the result responding to the situation, then intelligent of evolved heuristics – fast and effi cient behav- people will be able to execute these responses ioral or cognitive responses that require minimal, more quickly and consistently than other people. explicit cognitive resources – or heuristics learned If evolved or learned heuristics are not available, during the life span. However, when information there is an automatic shifting of attention to the cannot be automatically processed by evolved information represented in short-term memory. 112 D.C. Geary

Once attention is focused, intelligent people are that is, these frontal and parietal regions are typi- able represent more information in working cally active while people are engaged in diffi cult memory than are other people and have an problem-solving and reasoning tasks (e.g., enhanced ability to consciously manipulate this Duncan et al. 2000; Gray et al. 2003; Haier et al. information. The manipulation is guided and 1988), although other task-specifi c regions may constrained by reasoning and inference-making be engaged as well. mechanisms (Embretson 1995 ; Stanovich 1999 ). Many of the regions identifi ed as being critical to Gf overlap with those engaged in working- memory tasks, consistent with the previously Brain Systems and Intelligence described correlations, although there appear to be differences as well. In a unique and large-scale I have proposed that this attention-driven ability study, Barbey et al. (2014 ) studied defi cits in to explicitly represent and systematically and intelligence and the maintenance aspect of work- logically manipulate information in working ing memory (keeping something in mind while memory are core features of an evolved human engaged in other mental processes) in individuals ability to adapt to social and ecological variation with focal brain damage. They confi rmed the and novelty within the life span (Geary 2005 ). If importance of the parieto-frontal network for correct, then the evolutionary emergence of Gf intelligence, especially in the right hemisphere. should track the above-described evolutionary The working-memory task was dependent on changes in brain size and EQ. Unfortunately, a some of these same regions but also many regions direct test of this hypothesis is not possible due to of the left hemisphere (e.g., areas 41, 42, Fig. 8.2 ) the absence of Gf data in the paleontological that were not important for intelligence. These record. There are some interesting patterns, nev- results need to be interpreted with some caution, ertheless. I noted earlier that there have been dis- however, because the working-memory task used proportionate increases in the size and in this study does not assess all components of organization of some prefrontal and parietal working memory (e.g., inhibitory control), and regions, regions that have also been implicated in lesion studies may not capture the dynamic inte- brain imaging studies of Gf and self-awareness. gration of areas during complex reasoning and problem solving, as noted in another large-scale Brain Size and Regional Activation lesion study (Gläscher et al. 2009 ). There is a modest relation between gross brain In any case, the anterior cingulate (area 32, volume and intelligence ( r ~ 0.3–0.4; Deary 2000 ; Fig. 8.2) is included in the P-FIT model and is McDaniel 2005 ; Rushton and Ankney 1996 ). important because this brain region is engaged Recent reviews also indicate modest correlations when goal achievement requires dealing with between the size of specifi c regions in the pre- some degree of novelty, confl ict, or making a dif- frontal, parietal, and temporal cortices and poten- fi cult decision (Miller and Cohen 2001 ; tially the corpus callosum (Deary et al. 2010 ; Ranganath and Rainer 2003 ) – these are situa- Luders et al. 2009 ; Jung and Haier 2007 ). Of par- tions in which a goal cannot be readily achieved ticular importance may be the size and integra- by means of heuristics. Areas of the anterior cin- tion of regions in the prefrontal cortex (areas 6, 9, gulate cortex are thus the potential mechanism 10, 32, and 45–47, Fig. 8.2 ), the attentional con- that results in the automatic attentional shift to trol and imagery integration areas of the parietal novel, confl icted, or changing information repre- cortex (areas 7, 39, 40, Fig. 8.2), and the white sented in short-term memory and a correspond- matter tracts that connect them, which are coming activation of the dorsolateral and other ponents of Jung and Haier’s parieto-frontal inte- prefrontal areas (Botvinick et al. 2001 ). These gration theory (P-FIT) of intelligence. Results areas in turn enable the attentional focus and from functional MRI and other imaging tech- explicit, controlled problem solving needed to niques are also consistent with the P-FIT theory; cope with novel situations, resolve confl icts, and 8 Social Competition and the Evolution of Fluid Intelligence 113 make decisions that involve cost-benefi t trade- of processing and an attention-driven working- offs (Kerns et al. 2004 ; Miller and Cohen 2001 ). memory system are not competing explanations of Gf but rather may be coevolved and complementary mechanisms. Integration

Brain imaging studies suggest that an integrated Integrated Model: The Motivation system of brain regions supports the explicit con- to Control trolled problem solving that is the core of fl uid intelligence and that many, but not all, of these The theoretical model that places general fl uid same regions support working memory (Duncan intelligence in the context of the human brain and et al. 2000 ; Gray et al. 2003 ; Kane and Engle cognitive evolution can be organized around a 2002 ). High scores on measures of fl uid intelli- “motivation to control.” This does not mean indi- gence are associated with activation of the dorso- viduals necessarily have a conscious, Machiavellian lateral prefrontal cortex and several other brain motivation to control others . Rather, the “motiva- regions associated with attentional control, tion to control” is a heuristic for conceptualizing including the anterior cingulate cortex and the function of evolved traits. regions of the parietal cortex (Jung and Haier 2007 ). Many of these same regions contribute to the ability to inhibit irrelevant information from General Theory intruding into working memory and conscious awareness (Esposito et al. 1999) and inhibit the The basic proposal is that the brain and mind of execution of evolved or learned heuristics (Geary all species evolved to process the forms of infor- 2005 ). Awareness of information represented in mation (e.g., facial expressions, movement pat- working memory and the ability to mentally terns of predators) that covaried with survival and manipulate this information may result from a reproductive prospects during the species’ evolu- synchronization of the prefrontal brain regions tionary history. These systems operate implicitly that subserve the central executive and the brain and bias the organism to behave in ways that regions that process the specifi c forms of infor- result in attempts to gain control of these out- mation (e.g., voice, face, object; Damasio 1989 ; comes, such as capturing prey or avoiding being Posner 1994 ). captured by a predator (Gigerenzer et al. 1999 ; An attention-driven synchronization of the Simon 1956). The gist is consistent with the well- activity of the P-FIT network and activation of replicated fi nding that people’s subjective well- domain-specifi c brain regions would be facili- being and physical health are associated with tated by faster speed of processing and rich inter- having some level of control over relationships, connections among these brain regions. The events, and resources that are of signifi cance in latter are associated with larger brain size and their life (Heckhausen and Schulz 1995 ). As noted especially a greater volume and myelination of earlier, in conditions in which we evolved, achiev- axons (i.e., white matter; Deary et al. 2010 ). ing control of social relationships (e.g., as related Speed of processing may be important for the to social status) and biological (e.g., food) and synchronization process, because faster speed of physical (e.g., safe shelter, water) resources often processing would enable more accurate adjust- meant the difference between living and dying. ments in regional synchronization per feedback The control-related behavioral focus is repre- cycle. With repeated synchronized activity, the sented by the apex and adjoining section of result appears to be the formation of a neural net- Fig. 8.3 . The bottom of the fi gure represents work that automatically links the processing of the folk modules that result in implicit and these information patterns. In other words, speed automatic, bottom-up processing of social (e.g., 114 D.C. Geary

Fig. 8.3 The apex and following section represent the focus of behavior on achieving control of the social, biological, and physical resources that have tended to covary survival and reproductive outcomes during human evolution. The midsection shows the supporting affective, conscious- psychological (e.g., self-awareness), and cognitive (e.g., working memory) mechanisms that support the motivation to control and operate on the modular systems shown at the base (Adapted from Geary 2005 , p. 74. Copyright 2005 by American Psychological Association)

facial expressions), biological (e.g., features of are centered on the self and one’s relationships hunted species), and physical (e.g., objects poten- with other people or one’s access to biological tially useable as tools) information patterns that and physical resources. The representations often have tended to be the same across generations involve a form of mental time travel; that is, men- and within lifetimes and have covaried with sur- tal simulations of past, present, or potential future vival or reproductive prospects during human states and can be cast as visual images, in lan- evolution; a corresponding taxonomy of folk guage, or as episodic memories (Suddendorf and abilities is provided elsewhere (Geary 2005 ; Corballis 1997 ; Tulving 2002 ). The key is the Geary and Huffman 2002 ). If the evolutionary ability to create a mental representation of a expansion of the human brain and presumably desired or fantasized state and to compare this to enhancement of fl uid intelligence was driven by a mental representation of one’s current situation. the need to deal with rapid change in climatic, These are self-centered conscious-psychological ecological, and especially social conditions, then representations of present and potential future cognitive systems that function to mentally repre- states that are of personal signifi cance and are the sent and manipulate dynamic change in informa- content on which more conscious and effortful tion patterns should be identifi able. These reasoning and problem-solving processes are systems are represented by the center section of applied (Evans 2002 ; Stanovich and West 2000 ), Fig. 8.3 and are the key to understanding the evo- with the goal of reducing the difference between lution of fl uid intelligence. the current and desired states (Geary 2005 ). I have argued elsewhere that the evolution of Conscious-Psychological self-awareness and infusion of this awareness The core of the conscious-psychological mecha- into these mental simulations is best explained by nism is the human ability to form a conscious, social selection pressures, that is, awareness of explicit mental representation of situations that oneself is particularly important when attempting 8 Social Competition and the Evolution of Fluid Intelligence 115 to cope with other people with confl icting inter- require novel behaviors and in circumstances ests who know your personality, behavioral when one is attempting to outwit competitors or biases, etc. (Geary 2005 ). Mental simulation of avoid being outwitted by them (Geary 2005 ). the behavior of other people and especially of the self engages an identifi able brain network that Affective Mechanisms includes the prefrontal (area 32, 10, Fig. 8.2 ) and Affective mechanisms include emotions, that is, parietal (area 7) cortices (Gallagher and Frith observable behaviors (e.g., facial expressions), 2003 ; Lou et al. 2004 ). These systems appear to and feelings – conscious representations of an reconstitute the activity of the brain regions that emotional state or other conditions that can were engaged during personal experiences or potentially infl uence the individual’s well-being activate more abstracted representations of com- (Damasio 2003 ). Emotions result in observable mon features of these experiences (Damasio feedback to others, and feelings provide unob- 1989) and are highly integrated with the prefron- servable feedback to the individual (Campos tal regions described in the next section (Miller et al. 1989). The latter is a useful indicator of the and Cohen 2001 ). effectiveness of control-related behavioral strategies and an indicator of the potential benefi ts of a Cognitive Mechanisms simulated behavior. Positive feelings provide The cognitive mechanisms include working reinforcement when strategies are resulting in a memory, attentional control, and the ability to reduction in the difference between the current inhibit automatic processing of evolved or learned and desired state, and negative feelings promote behavioral responses or cognitive biases (e.g., disengagement when behaviors are not resulting attributional biases; Baddeley 2000 ; Cowan in this end (Gray 1987 ). The supporting brain 1995). The mechanisms also include the ability to systems, such as the amygdala (not shown in systematically problem-solve and to reason about Fig. 8.2 ), are predicted to function in part to patterns represented in working memory (Newell amplify attention to evolutionarily signifi cant and Simon 1972 ). In effect, working memory and forms of information and produce emotions, feel- attentional and inhibitory control are the content- ings, and corresponding behavioral biases that free mechanisms that enable the integration of a are likely to automatically reproduce outcomes current conscious-psychological state with mem- that have covaried with survival or reproduction ory representations of related past experiences during hominid evolution. and the generation of mental models or simulations of potential future states (Alexander 1989 ). Attentional and executive control is dependent Conclusion on several regions of the prefrontal cortex, such as the dorsolateral regions, as well as areas of the Since the emergence of australopithecines more parietal cortex (Kane and Engle 2002 ). These than 4 MYA, the hominid brain has tripled in size, areas are highly integrated with other regions of with much of the change occurring after the emer- the prefrontal cortex, such as the ventromedial gence of Homo and especially during the past sev- areas (e.g., area 11) of both hemispheres, the eral hundred thousand years. The proposed right frontal pole (area 10), and areas of the pari- selection pressures driving this change include etal cortex (e.g., area 7) that support social cogni- climatic variation (Potts 1998 ; Vrba 1995 ), the tion, including a sense of self (Tulving 2002 ). complexities of hunting and foraging (Kaplan The integration of these areas is consistent with a et al. 2000 ), and social dynamics (Alexander coevolutionary process, whereby working mem- 1989 ; Dunbar 1998 ; Humphrey 1976 ). It may be ory and fl uid intelligence are readily activated that each class of pressure contributed to the evo- during the processing of social information and lution of the human brain over the past 4 million self-relevant situations. My proposal is that these years, but the most interesting issue concerns the systems are coactivated in social contexts that particularly rapid and substantial increases in 116 D.C. Geary brain volume and EQ following the emergence of that is, a conscious-psychological (explicit) rep- H. erectus and continuing through to modern resentation of past, present, or potential future humans (Bailey and Geary 2009 ; Pearce et al. situations that are of personal relevance. 2013 ). Alexander’s ecological dominance model Although climatic and ecological conditions can provides a way of integrating these selection pres- create variation and novelty, the most dynamic sures and for making inferences about when the and variable conditions faced by humans are various pressures may have been most critical in those that arise from the competing interests of driving human brain and cognitive evolution. On other people. My proposal is that the evolved the basis of this model, my colleagues and I have function of these mental models is to generate a argued that the complexity and dynamics of social self-centered simulation of the “perfect” world, competition and cooperation within and between one in which other people behave in ways con- groups is likely to have been the most potent sistent with one’s best interest, and biological selection pressure for human brain and cognitive and physical resources are under one’s control. evolution since the emergence of H. erectus (Flinn The function of mental simulations is to create et al. 2005 ; Geary 2005 ). and rehearse strategies that can be used to reduce The combination of selection pressures and the difference between this perfect world and especially those related to social competition can current conditions. The cognitive systems that be integrated with psychological research using evolved to support the use of these self-centered the motivation to control framework. The folk mental models are known as working memory physical systems at the base of Fig. 8.3 and and general fl uid intelligence. described elsewhere support the evolution of modularized brain and cognitive systems that enable humans to conceptualize, construct, and References use very complicated tools (Geary 2005 ), consistent with Kaplan et al.’s (2000 ) hunting hypothe- Ackerman, P. L., Beier, M. E., & Boyle, M. O. (2005). sis and Alexander’s (1989 ) ecological dominance Working memory and intelligence: The same or different constructs? Psychological Bulletin, 131 , 30–60. model. As noted, the model also predicts the evo- Aiello, L. C., & Wheeler, P. (1995). The expensive-tissue lution of modularized folk psychological or hypothesis: The brain and digestive system in human sociocognitive competencies that are uniquely and primate evolution. Current Anthropology, 36 , human, which would include language, theory of 199–221. Alexander, R. D. (1989). Evolution of the human psyche. mind, face processing, among others (Geary In P. Mellars & C. Stringer (Eds.), The human revolu- 2005 ). Much of human behavior and cognition tion: Behavioural and biological perspectives on the and perhaps all of that of most other species can origins of modern humans (pp. 455–513). Princeton, be accommodated by the fast, implicit function- NJ: Princeton University Press. Alroy, J. (2001). A multispecies overkill simulation of the ing of such modularized systems, for example, end-Pleistocene megafaunal mass extinction. Science, resulting in rapid responses to predators or threat- 292 , 1893–1896. ening facial expressions. Asfaw, B., White, T., Lovejoy, O., Latimer, B., Simpson, The motivation-to-control model includes S., & Suwa, G. (1999). Australopithecus garhi: A new species of early hominid from Ethiopia. Science, 284 , modules in these domains as well as mechanisms 629–635. that enable organisms to anticipate and cope Ash, J., & Gallup, G. G., Jr. (2007). Paleoclimatic varia- with variation, novelty, and change within their tion and brain expansion during human evolution. life span. The critical differences comparing Human Nature, 18 , 109–124. Baddeley, A. D. (1986). Working memory . Oxford: Oxford humans to other species appear to be in the abili- University Press. ties to generate representations of the self in Baddeley, A. (2000). The episodic buffer: A new compo- working memory – self-awareness – and to men- nent of working memory? Trends in Cognitive tally time-travel (Suddendorf and Corballis Sciences, 4 , 417–423. Bailey, D. H., & Geary, D. C. (2009). Hominid brain evo- 1997 ; Tulving 2002). The combination enables lution: Testing climatic, ecological, and social compe- the generation of self-centered mental models, tition models. Human Nature, 20 , 67–79. 8 Social Competition and the Evolution of Fluid Intelligence 117

Barbey, A. K., Colom, R., Paul, E. J., & Grafman, J. Dunbar, R. I. M. (1998). The social brain hypothesis. (2014). Architecture of fl uid intelligence and working Evolutionary Anthropology, 6 , 178–190. memory revealed by lesion mapping. Brain Structure Duncan, J., Rüdiger, J. S., Kolodny, J., Bor, D., Herzog, and Function , 219, 485–494. H., Ahmed, A., Newell, F. H., & Emslie, H. (2000). A Barton, R. A. (1996). Neocortex size and behavioural neural basis for general intelligence. Science, 289 , ecology in primates. Proceedings of the Royal Society 457–460. of London B, 263 , 173–177. Elton, S., Bishop, L. C., & Wood, B. (2001). Comparative Betzig, L. L. (1986). Despotism and differential reproduc- context of Plio-Pleistocene hominin brain evolution. tion: A Darwinian view of history . New York: Aldine. Journal of Human Evolution, 41 , 1–27. Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, Embretson, S. E. (1995). The role of working memory C. S., & Cohen, J. D. (2001). Confl ict monitoring and capacity and general control processes in intelligence. cognitive control. Psychological Review, 108 , Intelligence, 20 , 169–189. 624–652. Engle, R. W., Conway, A. R. A., Tuholski, S. W., & Brodmann, K. (1909). Vergleichende Lokalisationslehre Shisler, R. J. (1995). A resource account of inhibition. der Grosshirnrinde in ihren Prinzipien dargestellt auf Psychological Science, 6 , 122–125. Grund des Zellenbaues . [Comparative localization of Esposito, G., Kirkby, B. S., van Horn, J. D., Ellmore, the cerebral cortex based on cell composition.] T. M., & Berman, K. F. (1999). Context-dependent, Leipzig: Barth. neural system-specifi c neurophysiological concomi- Campos, J. J., Campos, R. G., & Barrett, K. C. (1989). tants of ageing: Mapping PET correlates during cogni- Emergent themes in the study of emotional develop- tive activation. Brain, 122 , 963–979. ment and emotion regulation. Developmental Evans, J. S. B. T. (2002). Logic and human reasoning: An Psychology, 25 , 394–402. assessment of the deduction paradigm. Psychological Carpenter, P. A., Just, M. A., & Shell, P. (1990). What one Bulletin, 128 , 978–996. intelligence test measures: A theoretical account of Falk, D., Redmond, J. C., Jr., Guyer, J., Conroy, G. C., processing in the Raven Progressive Matrices Test. Recheis, W., Weber, G. W., & Seidler, H. (2000). Early Psychological Review, 97 , 404–431. hominid brain evolution: A new look at old endocasts. Cattell, R. B. (1963). Theory of fl uid and crystallized Journal of Human Evolution, 38 , 695–717. intelligence: A critical experiment. Journal of Flinn, M. V., Geary, D. C., & Ward, C. V. (2005). Educational Psychology, 54 , 1–22. Ecological dominance, social competition, and coali- Conway, A. R. A., Cowan, N., Bunting, M. F., Therriault, tionary arms races: Why humans evolved extraordi- D. J., & Minkoff, S. R. B. (2002). A latent variable nary intelligence. Evolution and Human Behavior, 26 , analysis of working memory capacity, short-term 10–46. memory capacity, processing speed, and general fl uid Foley, R. (1987). Hominid species and stone-tool intelligence. Intelligence, 30 , 163–183. assemblages: How are they related. Antiquity, 61, Cowan, N. (1995). Attention and memory: An integrated 380–392. framework . New York: Oxford University Press. Foley, R., & Lahr, M. M. (1997). Mode 3 technologies and Damasio, A. R. (1989). Time-locked multiregional retro- the evolution of modern humans. Cambridge activation: A systems-level proposal for the neural Archaeology Journal, 7 , 3–36. substrates of recall and recognition. Cognition, 33 , Gallagher, H. L., & Frith, C. D. (2003). Functional imag- 25–62. ing of ‘theory of mind’. Trends in Cognitive Sciences, Damasio, A. (2003). Looking for Spinoza: Joy, sorrow, 7 , 77–83. and the feeling brain . Orlando: Harcourt. Geary, D. C. (2005). The origin of mind: Evolution of Darwin, C. (1859). The origin of species by means of nat- brain, cognition, and general intelligence . Washington, ural selection . London: John Murray. DC: American Psychological Association. Darwin, C. (1871). The descent of man, and selection in Geary, D. C. (2010). Male, female: The evolution of relation to sex . London: John Murray. human sex differences (2nd ed.). Washington, DC: Darwin, C., & Wallace, A. (1858). On the tendency of spe- American Psychological Association. cies to form varieties, and on the perpetuation of vari- Geary, D. C., & Huffman, K. J. (2002). Brain and cogni- eties and species by natural means of selection. tive evolution: Forms of modularity and functions of Journal of the Linnean Society of London, Zoology, 3 , mind. Psychological Bulletin, 128 , 667–698. 45–62. Gigerenzer, G., Todd, P. M., & ABC Research Group Deary, I. J. (2000). Looking down on human intelligence: (Eds.). (1999). Simple heuristics that make us smart . From psychophysics to the brain . Oxford: Oxford New York: Oxford University Press. University Press. Gläscher, J., Tranel, D., Paul, L. K., Rudrauf, D., Rorden, Deary, I. J., & Stough, C. (1996). Intelligence and inspec- C., Hornaday, A., & Adolphs, R. (2009). Lesion tion time: Achievements, prospects, and problems. mapping of cognitive abilities linked to intelligence. American Psychologist, 51 , 599–608. Neuron, 61 , 681–691. Deary, I. J., Penke, L., & Johnson, W. (2010). The neuro- Gray, J. A. (1987). Perspectives on anxiety and impulsiv- science of human intelligence differences. Nature ity: A commentary. Journal of Research in Personality, Reviews Neuroscience, 11 , 201–211. 21 , 493–509. 118 D.C. Geary

Gray, J. R., Chabris, C. F., & Braver, T. S. (2003). Neural Jung, R. E., & Haier, R. J. (2007). The parieto-frontal inte- mechanisms of general fl uid intelligence. Nature gration theory (P-FIT) of intelligence: Converging Neuroscience, 6 , 316–322. neuroimaging data. Behavioral and Brain Sciences, Haier, R. J., Siegel, B. V., Jr., Nuechterlein, K. H., 30 , 135–187. Hazlett, E., Wu, J. C., Paek, J., Browning, H. L., & Kane, M. J., & Engle, R. W. (2002). The role of prefrontal Buchsbaum, M. S. (1988). Cortical glucose metabolic cortex in working-memory capacity, executive atten- rate correlates of abstract reasoning and attention stud- tion, and general fl uid intelligence: An individual- ied using positron emission tomography. Intelligence, differences perspective. Psychonomic Bulletin & 12 , 199–217. Review, 9 , 637–671. Heckhausen, J., & Schulz, R. (1995). A life-span theory of Kaplan, H., Hill, K., Lancaster, J., & Hurtado, A. M. control. Psychological Review, 102 , 284–304. (2000). A theory of human life history evolution: Diet, Hed, H. M. E. (1987). Trends in opportunity for natural intelligence, and longevity. Evolutionary selection in the Swedish population during the period Anthropology, 9 , 156–185. 1650–1980. Human Biology, 59 , 785–797. Kerns, J. G., Cohen, J. D., MacDonald, A. W., III, Cho, Hill, K., Boesch, C., Goodall, J., Pusey, A., Williams, J., & R. Y., Stenger, V. A., & Carter, C. S. (2004). Anterior Wrangham, R. (2001). Mortality rates among wild chim- cingulate confl ict monitoring and adjustments in con- panzees. Journal of Human Evolution, 40, 437–450. trol. Science, 303 , 1023–1026. Holloway, R. L. (1973). Endocranial volumes of early Levine, B. (1999). Self-regulation and autonoetic con- African hominids, and the role of the brain in human sciousness. In E. Tulving (Ed.), Memory, conscious- mosaic evolution. Journal of Human Evolution, 2 , ness, and the brain: The Tallinn conference 449–459. (pp. 200–214). Philadelphia: Psychology Press. Holloway, R. (1996). Evolution of the human brain. In Lou, H. C., Luber, B., Crupain, M., Keenan, J. P., Nowak, A. Lock & C. R. Peters (Eds.), Handbook of human M., Kjaer, T. W., Sackeim, H. A., & Lisanby, S. H. symbolic evolution (pp. 74–116). New York: Oxford (2004). Parietal cortex and representation of the University Press. mental self. Proceedings of the National Academy of Holloway, R. L., & de al Coste-Lareymondie, M. C. Sciences of the United States of America, 101 , (1982). Brain endocast asymmetry in pongids and 6827–6832. hominids: Some preliminary fi ndings on the paleon- Luders, E., Narr, K. L., Thompson, P. M., & Toga, A. W. tology of cerebral dominance. American Journal of (2009). Neuroanatomical correlates of intelligence. Physical Anthropology, 58 , 101–110. Intelligence, 37 , 156–163. Holloway, R. L., & Kimbel, W. H. (1986). Endocast mor- Malthus, T. R. (1798). An essay on the principle of popu- phology of Hadar hominid AL 162–28. Nature, 321 , lation as it affects the future improvement of society 536–537. with remarks on the speculations of Mr. Godwin, Horn, J. L. (1968). Organization of abilities and the devel- M. Condorcet, and other writers. London: Printed for opment of intelligence. Psychological Review, 75 , J. Johnson, in St. Paul’s church-yard. 242–259. Marmot, M. (2004). The status syndrome: How social Horn, J. L. (1988). Thinking about human abilities. In standing affects our health and longevity . New York: J. R. Nesselroade & R. B. Cattell (Eds.), Handbook of Henry Holt and Company, LLC. multivariate experimental psychology (2nd ed., Martin, P. S. (1973). The discovery of America: The fi rst pp. 645–685). New York: Plenum. Americans may have swept the Western Hemisphere Hsieh, C.-h., Reiss, C. S., Hunter, J. R., Beddington, J. R., and decimated its fauna in 1000 years. Science, 179 , May, R. M., & Sugihara, G. (2006). Fishing elevates 969–974. variability in the abundance of exploited species. McDaniel, M. A. (2005). Big-brained people are smarter: Nature, 443 , 859–862. A meta-analysis of the relationship between in vivo Humphrey, N. K. (1976). The social function of intellect. brain volume and intelligence. Intelligence, 33 , In P. P. G. Bateson & R. A. Hinde (Eds.), Growing 337–346. points in ethology (pp. 303–317). New York: McHenry, H. M. (1994). Tempo and mode in human evo- Cambridge University Press. lution. Proceedings of the National Academy of Huxley, T. H. (1863). Evidence as to man’s place in Sciences USA, 91 , 6780–6786. nature . New York: Appleton and company. Miller, E. K., & Cohen, J. D. (2001). An integration of Jensen, A. R. (1982). Reaction time and psychometric g. theory of prefrontal cortex function. Annual Review of In H. J. Eysenck (Ed.), A model for intelligence Neuroscience, 24 , 167–202. (pp. 93–132). New York: Springer. Mithen, S. (1996). The prehistory of the mind: The cogni- Jensen, A. R. (1998). The g factor: The science of mental tive origins of art and science. New York: Thames and ability . Westport: Praeger. Hudson. Jensen, A. R., & Munro, E. (1979). Reaction time, move- Newell, A., & Simon, H. A. (1972). Human problem ment time, and intelligence. Intelligence, 3 , 121–126. solving . Englewood Cliffs, NJ: Prentice-Hall. Jerison, H. J. (1973). Evolution of the brain and Pearce, E., Stringer, C., & Dunbar, R. I. M. (2013). New intelligence . New York: Academic. insights into differences in brain organization between 8 Social Competition and the Evolution of Fluid Intelligence 119

Neanderthals and anatomically modern humans. Stanovich, K. E., & West, R. F. (2000). Individual differ- Proceedings of the Royal Society B, 280 , 2130–2140. ences in reasoning: Implications for the rationality Pinker, S. (2011). The better angels of our nature: Why debate? Behavioral and Brain Sciences, 23 , 645–726. violence has declined . New York: Viking. Suddendorf, T., & Corballis, M. C. (1997). Mental time Posner, M. I. (1994). Attention: The mechanisms of travel and the evolution of the human mind. Genetic, consciousness. Proceedings of the National Academy Social, and General Psychology Monographs, 123 , of Sciences USA, 91 , 7398–7403. 133–167. Potts, R. (1998). Variability selection in hominid evolu- Tobias, P. V. (1987). The brain of Homo habilis : A new tion. Evolutionary Anthropology, 7 , 81–96. level of organization in cerebral evolution. Journal of Ranganath, C., & Rainer, G. (2003). Neural mechanisms Human Evolution, 16 , 741–761. for detecting and remembering novel events. Nature Tulving, E. (2002). Episodic memory: From mind to Reviews Neuroscience, 4 , 193–202. brain. Annual Review of Psychology, 53 , 1–25. Rushton, J. P., & Ankney, C. D. (1996). Brain size and United Nations. (1985). Socio-economic differentials in cognitive ability: Correlations with age, sex, social child mortality in developing countries . New York: class, and race. Psychonomic Bulletin & Review, 3 , Author. 21–36. Vrba, E. S. (1995). The fossil record of African antelopes Semaw, S., Rogers, M. J., Quade, J., Renne, P. R., Butler, (Mammalia, Bovidae) in relation to human evolution R. F., Dominquez-Rodrigo, M., Stout, D., Hart, W. S., and paleoclimate. In E. S. Vrba, G. H. Denton, T. C. Pickering, T., & Simpson, S. W. (2003). 2.6-million- Partridge, & L. H. Burckle (Eds.), Paleoclimate and year-old stone tools and associated bones from OGS-6 evolution, with emphasis on human origins (pp. 385– and OGS-7, Gona, Afar, Ethiopia. Journal of Human 424). New Haven, CT: Yale University Press. Evolution, 45 , 169–177. Wallace, A. R. (1911). The world of life: A manifestation Simon, H. A. (1956). Rational choice and the structure of of creative power, direction mind and ultimate pur- the environment. Psychological Review, 63 , 129–138. pose . New York: Moffat, Yard & Company. Spearman, C. (1904). General intelligence, objectively Wood, B., & Collard, M. (1999). The human genus. determined and measured. American Journal of Science, 284 , 65–71. Psychology, 15 , 201–293. Wrangham, R. W., Holland Jones, J., Laden, G., Pilbeam, Stanovich, K. E. (1999). Who is rational? Studies of D., & Conklin-Brittain, N. (1999). The raw and stolen: individual differences in reasoning. Mahwah, NJ: Cooking and the ecology of human origins. Current Erlbaum. Anthropology, 40 , 567–594. Part III Theories of Intelligence Intelligence Deﬁ ned: Wundt, James, Cattell, Thorndike, 9 Goddard, and Yerkes

John D. Greenwood

Most texts in the history of psychology credit Like Charles Darwin (1809–1882), Spencer Francis Galton (1822–1911) and Alfred Binet believed in strong psychological continuity (1857–1911) as the fi rst to develop theories of between humans and other animals: that the psy- intelligence as well as instruments for its mea- chological capacities of humans and other ani- surement. However, credit should probably go to mals differ in degree but not in fundament kind.1 the Victorian polymath Herbert Spencer (1820– While for Darwin this just meant that complex 1903) as the fi rst to develop a substantive theory and distinctively human capacities such as rea- of intelligence (Guilford 1967 ) and one which soning and language could be attributed to higher implied––at least for Spencer––individual, racial, animals in at least insipient form, Spencer held and species differences in intelligence. Spencer that humans and other animals differ only in the treated the empiricist principle of association by complexity of their associative processes. For contiguity as the foundation of intelligence in Spencer, the complex capacities of humans such both animals and humans: as reasoning and language are merely elabora- Hence the growth of intelligence at large depends tions of the basic associative processes common upon the law, that when any two psychical states to all animals. occur in immediate succession, an effect is pro- Spencer’s theory was enormously infl uen- duced such that if the fi rst subsequently recurs tial in its day, prompting theorists such as John there is a certain tendency for the second to follow it. (1855, p. 530) Hughlings Jackson (1835–1911) and Ivan Sechenov (1829–1910) to develop refl exive Spencer held that intelligence is determined sensorimotor theories of the nervous system, by the quantity and quality of adaptive associa- which treated cognitive capacities such as rea- tions made by organisms to their environment, soning and language as merely the pinnacles of by the “the continuous adjustment of internal a hierarchy of increasingly more complex and relations to external relations” ( 1855 , p. 374), sophisticated refl exive machinery (Jackson which he believed was in turn determined by neurophysiological complexity. Consequently, he maintained that intelligence is a function of 1 “The difference in mind between man and the higher ani- brain size. mals, great as it is, is certainly one of degree and not of kind. We have seen that the senses and intuitions, the various emotions and faculties, such as love, memory, atten- J. D. Greenwood , Ph.D. (*) tion, curiosity, imitation, reason, &c., of which man CUNY Graduate Center , 365 Fifth Avenue , boasts, may be found in an incipient, or even sometimes in New York , NY 10016 , USA a well-developed condition, in the lower animals.” e-mail: [email protected] (Darwin 1871 , p. 105)

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 123 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_9, © Springer Science+Business Media New York 2015 124 J.D. Greenwood

1931 ; Sechenov 1863 ). They developed theo- apperception, which shape human perception and ries of strong continuity between so-called cognition, are distinct from the forms of associa- “higher” cognitive processes such as reason tion that humans share with animals. Like Conwy and language and “lower” associative or refl ex- Lloyd Morgan (1852–1936), Wundt held that all ive processes, mirroring Spencer and Darwin’s animal psychology and behavior could “be theory of strong continuity between human and accounted for by the simple laws of association” animal psychology.2 (Wundt 1863 , p. 350) and treated apperception as Thus in a very real sense, scientifi c psycho- the highest stage of distinctively human psychological theories of intelligence in the nineteenth logical evolution. Consequently, Wundt had little century began with theories of animal intelli- to say about intelligence per se, except to distin- gence, and the question of the relation between guish the forms of apperceptive intelligence— human and animal intelligence remained a live involving the creative synthesis of relational issue for many early scientifi c psychologists. psychological elements––distinctive of humans from the forms of associative intelligence characteristic of both humans and animals. Wundt and James: Apperception William James managed to persuade the and Similarity President and Board of Trustees at Harvard that the “new psychology” provided an intellectual Wilhelm Wundt (1832–1920) and William James bastion against the materialist threats of evolu- (1842–1910) are generally held to be the found- tionary psychology and experimental physiology, ing fathers of scientifi c psychology in Germany which according to his future colleague at Yale, and America (respectively). However, neither George Trumbull Ladd (1842–1921), threatened followed Spencer in treating intelligence in terms to reduce conscious experience to “a stream of of the complexity of association by contiguity. mechanically associated ‘epiphenomena,’ thrown This is because both Wundt and James rejected off by the molecular machinery of the cerebral the principles of strong psychological continu- hemispheres” (1895 , p. x). Like Wundt and Lloyd ity––between humans and animals and between Morgan, James held that all animal psychology cognitive and associative processes––champi- and behavior could be explained in terms of asso- oned by evolutionary theorists like Spencer and ciation by contiguity and claimed that distinctive experimental physiologists like Hughlings human cognition went beyond this to the recogni- Jackson and Sechenov. Both Wundt and James tion of similarities and analogies: championed the autonomy of a scientifi c psy- We may then, we think, consider it proven that chology devoted to the study of conscious experi- the most elementary single difference between ence by rejecting what they perceived to be the human mind and that of brutes lies in this reductive challenges to the reality and effi cacy of defi ciency on the brute’s part to associate ideas by similarity —characters, the abstraction of human conscious experience. which depends on this association, must in the Thus Wundt, for example, maintained that the brute always remained drowned, swamped in the creative and selective attentional processes of total phenomenon which they help constitute. (James 1890, p. 360) While James did not develop a theory of intel- 2 Although it is doubtful if Darwin himself thought that rationality and language are just more complex elabora- ligence per se, it is clear that he thought that the tions of associative or refl exive processes. That is, he seems most intelligent men and women were those who to have held that cognition and association are psychologi- were able to most fully develop their ability to cally discontinuous, even though he held that the psychol- form associations by similarity: ogy of humans and animals is strongly continuous: he seems to have supposed that the discrete higher cognitive Genius then, as has been already said, is identical capacities of humans—such as rationality and language–– with the possession of similar association to an could be found in animals, in at least incipient form. extreme degree. (1890, p. 360) 9 Intelligence Defi ned: Wundt, James, Cattell, Thorndike, Goddard, and Yerkes 125

Although neither Wundt nor James developed in 1886. Galton, who was Darwin’s half cousin, substantive theories or measures of intelligence, maintained that individual differences in a number of their students did and made major sensorimotor reaction times were a natural con- contributions to debates about the nature of intel- sequence of the chance distribution of inherited ligence in the early twentieth century. To these characteristics described by Darwin in On the we now turn. Origin of Species (Darwin 1857 ) and set about measuring these differences in the general population. Galton gathered data on nearly 20,000 Cattell and Mental Testing people at his anthropometric laboratories at the International Health Exhibition in London (in James McKeen Cattell (1860–1944) was one of 1884) and the Science Galleries of the South Wundt’s earliest Leipzig students and his fi rst full- Kensington Museum (in 1888), employing physi- time American student. He also became Wundt’s cal and sensory acuity measures, such as head fi rst research assistant, based upon Cattell’s own size, physical strength, visual and auditory acu- recommendation (Boring 1957). While at Leipzig, ity, and reaction time. Cattell did experimental studies of the time taken Galton claimed that sensory acuity was corre- to complete “cerebral operations,” a topic suited to lated with intelligence and thus could serve as a Wundt’s program of research on mental chronom- convenient indirect measure of intelligence. 3 etry, based upon measures of reaction time. Galton also later claimed that he had demon- Cattell’s paper “On the Time It Takes to See and strated the correlation, based upon the statistical Name Objects” was published in Wundt’s journal measures of correlation that he developed toward Philosophical Studies in 1885 (Cattell 1885 ); a the end of the 1880s (Galton 1888 ), although he shorter version appeared in the British journal seems to have come to this conclusion sometime Mind a year later (Cattell 1886). earlier. For example, in Inquiries Into Human Most unusually, Wundt also allowed Cattell to Faculty (1883 ), Galton announced: pursue research on individual differences in reaction time, a topic that was anathema to Wundt, The trials I have as yet made on the sensitivity of different persons confi rm the reasonable expecta- who refused to allow students like Edward tion that it would on the whole be highest among B. Titchener (1867–1927) and Lightner Witmer the intellectually ablest. (1883, p. 20) (1867–1956) to participate in reaction time In this peculiar fashion, the measures of sen- experiments because they were not properly “cal- sory acuity and reaction time developed by ibrated” (trained) to reproduce the supposedly Wundt and his students in the Leipzig laboratory universal measures of sensory and motor reaction were appropriated by Galton, and later by Cattell, times (O’Donnell 1985 ). This may have been as measures of intelligence. because Wundt respected Cattell’s intellectual Galton also employed his newly developed independence and self-confi dence, qualities statistical measures––and his pioneering use of Wundt characterized as “typically American.” twin studies––to supposedly demonstrate that There was, however, one major theorist who human intelligence is largely determined by recognized individual differences in sensory and heredity. While the statistical calculations of motor reaction times, and that was Francis Galton Natural Inheritance (Galton, 1889 ) are impres- (1822–1911), with whom Cattell went to study in sive, the data on which they were founded are England after he completed his degree in Leipzig doubtful, based upon family records anony- mously submitted by correspondents hoping to 3 This program was based upon the complication experi- win the cash prizes for best entries offered by ments of the Dutch physiologist Franciscus Cornelis Galton (Boakes 1984 ). Donders (1818–1889), in which the time taken for components of a complex mental task was calculated by sub- Given his commitment to the hereditarian tracting the time taken for other components of the task. determination of intelligence, Galton was 126 J.D. Greenwood

consequently dismissive of optimistic utilitarian individual differences. He also embraced Galton’s theories––such as those advanced by John Stuart theory of the hereditarian determination of intel- Mill (1806–1873)––that held that all human ligence and contributed his small share to the pro- beings are capable of attaining the same intellec- gram of positive eugenics by offering each of his tual and moral levels, given similar nurturing seven children one thousand dollars if they mar- experiences. As Galton put it (once again in ried college professors. When he returned to advance of his statistical calculations): America for a position at the University of I have no patience with the hypothesis …that Pennsylvania, Cattell began to use the techniques babies are born pretty much alike, and that the sole he had developed at Leipzig for his own anthropo- agencies in creating differences between boy and metric studies and initiated a program of “mental boy, and man and man, are steady application and testing” based upon psychophysical measures of moral effort. It is in the most unqualifi ed manner that I object to pretensions of natural equality. grip strength, speed of movement, skin sensitivity, (Galton 1869, p. 12) and sensory and motor reaction times. In a paper published in Mind in 1890 (Cattell 1890), Cattell Although Galton did not believe that the intel- extolled the utility of such “mental tests,” and ligence levels of men and women could be sig- when he moved to Columbia in 1891, he sub- nifi cantly raised through education, he did think jected hundreds of students to them. they could be raised through selective breeding. Like Galton, Cattell assumed that his tests of Like others who refl ected on Darwin’s theory of sensory acuity were indicators of intelligence and evolution, Galton recognized that natural selec- consequently assumed that there would be a sig- tion operating on chance variations would not nifi cant correlation between test scores and aca- ensure the evolution of socially desirable charac- demic performance. However, at the turn of the teristics like high intelligence and moral virtue century, one of Cattell’s own students, Clark and indeed might very well lead to their attenua- Wissler (1870–1947), tried to demonstrate this tion, if “idiots and imbeciles” were allowed to by measuring the degree of correlation between overbreed. Galton coined the term eugenics performance on 21 of Cattell’s mental tests and (Greek for “wellborn”) to describe his recom- course grades, employing the Galtonian mea- mended form of artifi cial selection, designed to sures of correlation (including Pearson’s newly produce more intelligent and productive human developed correlation coeffi cient). To Cattell’s stock (as farmers used selective breeding to pro- consternation, Wissler (1901 ) found no correla- duce desired qualities in their animal stock, such tion between the test scores and the grades, nor as high quality of fl eece in sheep and body mass between the tests scores themselves (although he in cattle). did fi nd correlations between the student’s Originally, Galton recommended a form of grades). This effectively brought to an end to the positive eugenics. He thought that those persons program of the Galtonian anthropometric intelli- identifi ed as the most intelligent (via his sensory gence testing in the United States.4 However, acuity measures) should be encouraged to breed, intelligence testing would shortly resurface in an and to breed regularly, via fi nancial inducements entirely different guise. provided by the government. However, in reaction to the moral panic created by the failure of the mighty British Empire to crush a nation of Spearman and the “g” Factor farmers in the Second Boer War (1899–1902), Galton and his protégé Karl Pearson (1857–1936) Wissler’s studies were not quite the end of the recommended a form of negative eugenics Galtonian psychometric story, however, although through institutionalization and sterilization of they are commonly supposed to have been. the “idiots and imbeciles.” Cattell was greatly impressed by Galton’s 4 Although it was continued for a few years by Joseph work, which reinforced his own interest in Jastrow (1863–1944) at the University of Michigan. 9 Intelligence Defi ned: Wundt, James, Cattell, Thorndike, Goddard, and Yerkes 127

Charles Spearman (1863–1945), after serving as (Thomson 1916 ). Spearman did not always insist a regular offi cer in the British Army for 15 years, on the unitary nature of the g factor, at least in his resigned his commission in 1897 to study for a later work (Spearman 1925 ). He believed g to be doctoral degree in Wundt’s Laboratory. While at grounded in two “ubiquitously” cooperating abil- Leipzig, Spearman was surprised to learn about ities: an educative ability to make “meaning from Wissler’s results, but eventually came to the con- confusion” and a reproductive ability to recall clusion that Wissler’s data were unreliable and that meaning. And although Spearman remained that he had underestimated the degree of relation- a champion of intelligence testing and of the ship between them. Using a corrective statistical view that the innate g factor accounted for most formula, Spearman was able to demonstrate posi- individual differences of intelligence, he thought tive correlations between the test scores and that intelligence testing had no place in schools, between the test scores and academic grades. He which ought to be engaged in maximizing the parleyed this success into his theory that the cor- varied native abilities of individual students.6 relation between various mental tests could be explained by reference to a single, unitary capacity, or general factor “g,” which he identifi ed via Thorndike and Connectionism a new statistical technique that came to be known as factor analysis.5 Edward L. Thorndike (1874–1949) was a student Spearman called his theory, which he also of William James and received his master’s developed and published while a student at degree at Harvard in 1897. Like James (and Leipzig, as the “law of the universal unity of the Wundt), he was not impressed by anecdotal intellectual function”: reports of the apparently intelligent behavior of Whenever branches of intellectual activity are at animals, such as the apparent rationality of ani- all dissimilar, then their correlations with one mals reported by Darwin in The Descent of Man, another appear wholly due to their being all vari- and Selection in Relation to Sex (Darwin 1971 )7 ously saturated with some common fundamental or their apparent knowledge of mathematical and Function (or group of Functions). (Spearman, 1904, p. 124) mechanical principles reported by George Romanes (1848–1894) in Animal Intelligence Like Galton and Cattell, Spearman also (Romanes 1882).8 Thorndike disparaged anec- believed that this general factor ‘’g’’ is innate: dotal accounts of animal behavior because he G is in the normal course of events determined maintained they were generally unrepresentative innately; a person can no more be trained to have it of animals’ cognitive abilities: in higher degree than he can be trained to be taller. Dogs get lost hundreds of times and no one ever (Spearman 1931, in Deary et al 2008, p 157) notices it or sends an account of it to a scientifi c magazine. But let one fi nd his way from Brooklyn After Leipzig, Spearman returned to England to Yonkers and the fact immediately becomes a cir- culating anecdote. Thousands of cats on thousands and took up a position at University College, London, where he continued to develop his theory of intelligence. His theory was later criticized 6 Unlike his later colleague at University College, Sir Cyril by Sir Geoffrey Thompson (1881–1955), who Burt (1883–1971), who promoted the 1940s British government program of intelligence testing in schools––the argued that although g was a statistical reality, it “11 plus” exam––on the basis of which students were did not designate a unitary intelligence factor, but streamed into academic or trade classes. a variety of highly correlated intellectual skills 7 Darwin claimed that most people would agree that “animals possess some power of reasoning,” on the grounds that they “may constantly be seen to pause, deliberate, and 5 Strictly speaking, Spearman developed a two-factor theory resolve” (1871, p. 46). of intelligence, postulating a general factor “g” common to 8 Darwin bequeathed his notebooks on animal behavior to all tasks requiring intelligence and a factor “s” specifi c to Romanes after the two men became close friends toward different kinds of intellectual tasks (Sternberg 2003 ). the end of Darwin’s life. 128 J.D. Greenwood

of occasions sit helplessly yowling, and no one linking learning with reinforcement, and the “law takes thought of it or writes to his friend, the proof exercise,” linking learning with repetition. fessor; but let one cat claw at the knob of a door supposedly as a signal to be let out, and straight- Thorndike was at pains to insist that learning way this cat becomes the representative of the cat- was not based upon imitation or insight and that mind in all the books…In short, the anecdotes give connections between behavior and response were really the abnormal or supernormal psychology of “stamped in” by reinforcement and repetition. animals. (Thorndike 1911, pp. 23–25) Consequently, he called his theory “connection- Thorndike insisted that the scientifi c study of ism” and on the basis of his limited experiments animal “intelligence” should be based upon care- maintained that all animal and human behavior fully controlled experimental studies, which he could be explained in terms of the laws of effect began with chickens in William James’s base- and exercise and instinct: ment and continued at Columbia University when The higher animals, including man, manifest no Cattell offered him a fellowship there in 1897. At behavior beyond expectations from the laws of Columbia, Thorndike began a series of experi- instinct, exercise and effect. (Thorndike 1911, p. 274) ments in which food-deprived cats learned to escape from specially constructed slatted cages While he was later forced to modify the law of or “puzzle boxes” to gain a food reward, described effect and drop the law of exercise, Thorndike in his doctoral dissertation of 1898, and pub- continued to insist that learning was based upon lished later that year as Animal Intelligence: An connections or associations. So although he Experimental Study of the Associative Processes rejected Darwin and Romanes’s attribution of in Animals (Thorndike 1898) as a monograph higher cognitive processes such as rationality and supplement in Psychological Review9 . mechanical understanding to animals, he main- Thorndike’s studies were based upon Lloyd tained––with Spencer––that higher cognitive Morgan’s (1894 ) explanation of how his dog processes are nothing more than complex forms Toby had learned to lift the latch on the gate of of connection or association, based ultimately the back courtyard of his house to escape into the upon the same psychological and physiological street. Based upon his repeated observation of the principles: dog’s behavior, Lloyd Morgan dismissed …the higher forms of intellectual operations are Romanes’s (1882 ) explanation of such behavior identical with mere association or connection in terms of the dog’s understanding of mechani- forming, depending upon the same sort of physio- cal principles and noted how after accidentally logical connections but requiring many more of them . (Thorndike et al. 1926 , p. 415) stumbling on a means of lifting the latch by a movement of its head the dog had managed to Consequently, he supposed that an individu- learn after repeated trial and error how to open al’s level of intelligence was determined by the the gate. Similarly, Thorndike’s cats initially number of connections that individual was capa- clawed at the bars and pushed their paws between ble of making: them, until they accidentally hit on the movement By the same argument the person whose intellect is required to release the latch. Like Morgan, greater or higher or better than that of another per- Thorndike found that the animals took progres- son differs from him in the last analysis in having, sively less time to hit on the required behavior not a new sort of physiological process, but simply a larger number of connections of the ordinary over a series of trials, until eventually they pro- sort. (Thorndike et al. 1926 , p. 415) duced the learned behavior the moment they were placed in the box. While he recognized the diffi culty of deter- On the basis of these experiments, Thorndike mining the nature and number of connections that articulated what he called the “law of effect,” needed to be made to in order to execute a complex human cognitive process such as language 9 It was republished as an independent monograph in 1911 comprehension, Thorndike developed his own (Thorndike 1911 ). intelligence tests when he took up a full- time 9 Intelligence Defi ned: Wundt, James, Cattell, Thorndike, Goddard, and Yerkes 129 position at Teachers College, where he remained Binet-Simon scale. This became the standard for the rest of his career, devoting much of his measure of intelligence in the United States until time to educational psychology and the develop- Lewis Terman (1877–1956), a Clark University ment of tests for education (mainly on reading graduate of 1905 who secured a position as and writing). Although he came to his theoretical Professor of Child Study at Stanford University, conception of human intelligence by a different brought out what came to be known as the route than Galton, Cattell, and Spearman, Stanford-Binet scale in 1916. Thorndike was equally committed to the view that In 1905 Alfred Binet (1857–1911) and his intelligence was largely determined by hereditary, research assistant Theodore Simon (1873–1961) publishing Heredity, Correlation and Sex had developed a test for measuring children’s Differences in School Abilities in 1903 (Thorndike intelligence, as a means of assessing the child’s 1903). “mental level” (the tests were designed to ensure that most children at any age would test at the appropriate mental level). Unlike Galton and Goddard and the Binet-Simon Cattell’s tests, Binet and Simon’s tests were Intelligence Scale based on direct measures of intellectual abilities such as comprehension, problem solving, and A good case can be made for the claim that logical and analogical reasoning—all forms of Granville Stanley Hall (1844–1924), rather than intelligent judgment. Their scales consisted of William James, deserves to be credited as the thirty items, ranked in order of diffi culty, so that founding father of scientifi c psychology in everyone could do the easier ones, and children America. Although Hall was himself a student of scored progressively better with every increase in James, he was the fi rst to complete a PhD at “mental level.” Harvard on a psychological topic (with a disser- Binet and Simon’s test items represented what tation on spatial perception). He was also the fi rst Spearmen called a “hodgepodge” of factors, but to set up a fully developed psychological labora- they did the job for which they were designed, tory and PhD program in psychology at Johns namely, to provide an objective means of identi- Hopkins University in 1884,10 which he trans- fying children in need of remedial education. ferred to Clark University in 1888. Hall is mainly Unlike Spearman, Binet and Simon did not sup- remembered for his contributions to educational pose that they were measuring a unitary capacity, psychology and (like Wundt and James) had little far less one that is innately determined. On the to say about intelligence and its measurement, contrary, they designed programs of remedial but he did have two graduate students who made education for those children who tested below the signifi cant contributions to the fi eld. mental level of their age group, designed to Henry Goddard (1866–1957) graduated from increase their mental level through special train- Clark University in 1899 and became Director of ing methods called “mental orthopedics” (which the Research Laboratory for the Study of involved exercises in attention, will, and disci- Feeblemindedness at the Vineland Training pline). The Binet-Simon scale was administered School for Feeble-Minded Boys and Girls in to French schoolchildren between 1905 and 1908 New Jersey in 1906. He translated the 1908 and was judged to be a great success in Europe. It Binet-Simon scale for measuring children’s intel- was seen as an objective measure of intelligence ligence into English and also the revised 1911 that was easy to administer and was translated into many languages. Binet and Simon were careful to stress the 10 Although William James set up a laboratory of sorts at limitations of their scales, given their belief in Harvard in 1875, it was little more than a small collection the malleability of intelligence and the inherent of instruments housed in a stairwell closet (Hall 1923 , p. 218), and Harvard did not attain an independent (of phi- margin of error. However their cautious approach losophy) psychology department until 1937. was discarded when Goddard and Terman 130 J.D. Greenwood brought the Binet-Simon scales to the United immigrants of “poor stock,” that is, among the States––both followed Galton, Cattell, and increasing number of immigrants from Eastern Spearman in supposing that intelligence was a and Southern Europe, as opposed to the earlier unitary ability, which was largely determined by and supposedly “superior stock” from Northern heredity. In 1914, William Stern (1871–1938) and Western Europe. had introduced the theoretical notion of a mental Goddard had also read the works of Galton quotient, defi ned as a child’s mental age––as and Mendel and had noted that many brothers determined by their performance on the Binet- and sisters of the children at Vineland were also Simon scale––divided by their chronological age feebleminded themselves, which he operation- (Stern 1914), and Terman defi ned the intelli- ally defi ned as having an IQ of less than 70. In gence quotient (IQ) as Stern’s mental age multi- 1911, he set out to demonstrate the inheritance of plied by 100, so that the average intelligence feeblemindedness, which he believed to be quotient for any mental age would be 100. caused by a recessive gene, by exploring the fam- Goddard and Terman used these IQ scales to ily tree of a 22-year-old Vineland girl whom defi ne levels of intelligence and “feebleminded- Goddard had named Deborah Kallikak. She per- ness,” or as Terman put it bluntly in his PhD dis- formed on the Binet-Simon scale at the mental sertation at Clark University, “genius and level of 9 years, on the basis of which Goddard stupidity” (Terman 1906 ). classifi ed her as a moron, a technical term (mean- Goddard was convinced of the utility of the ing dull) introduced by Goddard to describe those tests as measures of intelligence and trained with an IQ of between 50 and 70. teachers at local schools to administer them. He Goddard set about exploring her family tree, was not surprised to fi nd that the scores for which he found could be traced back to one Martin inmates at Vineland were much lower than those Kallikak, a Revolutionary War soldier. Martin had of children in the regular public schools but was married an upstanding Quaker girl, who had borne concerned to discover that many of the public him seven children. The descendants of this schoolchildren scored lower than their age “good” side of the family tree had gone on to norms, which Goddard took as a disturbing indi- become upright citizens, such as lawyers, doctors, cator of the extent of feeblemindedness in the judges, teachers, and landed gentry. However, general population . Martin Kallikak had also dallied with a serving Goddard also impressed the immigration wench of loose morals, who had borne him an ille- offi cers at Ellis Island with his apparent ability gitimate son, Martin Kallikak junior, who fathered to identify feebleminded immigrants by sight ten children. The descendants of this “bad” side of and then have his identifi cation “objectively” the family turned out to be horse thieves, brothel confi rmed by their low scores on the Binet- owners, prostitutes, and alcoholics. Goddard Simon tests. Goddard trained assistants to iden- claimed to have discovered a high incidence of tify the feebleminded and administer the tests, feeblemindedness among the offspring of the serv- and eventually they came to be employed by the ing wench and a low incidence among the off- inspectors themselves, although some objected spring of the upstanding Quaker girl. that they included questions (e.g., about the This study, published in 1912 as The Kallikak New York Giants) that they would not have been Family: A Study in Feeblemindedness (Goddard able to answer when they fi rst came to the coun- 1912), confi rmed Goddard in his belief that fee- try. Nevertheless, the use of psychological test- blemindedness and intelligence are inherited, and ing methods for screening immigrants was that feeblemindedness was the root cause of judged to be a great success, and the number of licentiousness and criminality. Although Goddard persons deported on grounds of feebleminded- promoted the study as a “natural experiment,” ness increased dramatically. Again, Goddard there was no attempt to control for environmental drew a pessimistic conclusion about the high or social differences, and Goddard and his percentage of feeblemindedness among recent coworkers, like many asylum and prison superin- 9 Intelligence Defi ned: Wundt, James, Cattell, Thorndike, Goddard, and Yerkes 131 tendents of his day, simply equated immoral and criminal behavior with feeblemindedness. Yerkes and the Army Testing Nevertheless, Goddard’s conclusions were Project widely accepted and regularly cited, and his study spawned a spate of similar studies suppos- Robert M. Yerkes originally came to Harvard as a edly demonstrating the link between feeblemind- graduate student in zoology, but the philosopher edness and social degeneration (Zenderland Josiah Royce (1855–1916) persuaded him to 1998). Many also accepted the implied threats to combine his interests in zoology and psychology the “national stock” posed by the overbreeding of to study comparative psychology. Yerkes trans- what Galton had called “the idle and the infi rm” ferred to the philosophy department, then headed and the infl ux of recent immigrants of poor stock. by James’s successor, Hugo Münsterburg (1863– For Goddard, the moral and social implica- 1916), and received his PhD degree in 1902. tions were as clear as they were for Galton, Münsterburg managed to secure Yerkes an assis- namely the need for eugenic programs designed tantship and later an assistant professorship in the to prevent the breeding of the feebleminded. department, where he developed “criteria of the Goddard served as psychology representative on psychic” (Yerkes 1905 ), which supposedly pro- the 1911 Committee to Study and to Report on the vided researchers with objective grounds for the Best Practical Means of Cutting Off the Defective attribution of higher cognitive states to animals, Germ-Plasm in the American Population, estab- as a means of assessing their intelligence.11 lished by the Eugenics Section of the American However, Münsterburg objected to Yerkes’s fail- Breeders Association, which recommended the ure to follow his advice to devote more of his segregation and sterilization of mental defectives, time to educational psychology and threatened to mercifully drawing the line against euthanasia. shut down the comparative psychology program. Goddard also served on the 1913 Committee for Fortunately, Yerkes secured a position as state the Heredity of Feeblemindedness, which psychologist at Boston Psychopathic Hospital included fellow psychologists Edward Thorndike, just before the First World War, which also Lewis Terman, and Robert M. Yerkes (1876– enabled him to keep a half-time teaching position 1956), as well as the inventor Alexander Graham at Harvard at double his former pay. Bell (1847–1922) and the Harvard physiologist In later years, Yerkes became known for his Walter B. Canon (1871–1945), which also recom- seminal contributions to comparative psychology, mended that “defective classes” be eliminated but during his time at Boston Psychopathic Hospital, through sterilization. he developed his own intelligence test, the Yerkes- Such eugenic ideas were not themselves a Bridges Point Scale of Intelligence (Yerkes et al. product of intelligence testing, but had been in 1915 ). Yerkes was also a born organizer and admin- circulation since the beginning of the century. istrator, and his great chance came in April 1917, Charles Davenport (1866–1944), the author of when Titchener’s Experimentalists were meeting at Eugenics: The Science of Human Improvement Harvard. Yerkes was serving that year as President by Better Breeding ( 1911), had founded the of the American Psychological Association and, on Eugenics Records Offi ce at Cold Spring receiving the news that America had entered the Harbor, New York, in 1910, and similar societies and organizations had been founded in 11 Canada and Europe. However, the major impe- John B. Watson (1878–1958), who was a friend of Yerkes, was so skeptical of his criteria for the objective tus of the development of eugenics legislation determination of animal cognition and intelligence that in the United States was the result of the pro- he began to promote the claim that animal psychology gram of intelligence testing in the United should be restricted to the description of observable stim- States Army conducted during the First World ulus–response sequences (Watson 1909 ), a position he extended to human psychology 4 years later in his War, engineered by the psychologist Robert “Psychology as a Behaviorist Views It” lecture at M. Yerkes. Columbia (Watson 1913 ). 132 J.D. Greenwood

First World War, chaired a special session on the “feeblemindedness …is of much greater fre- proposed contribution of psychologists to the war quency than has been previously supposed” effort. Shortly afterward, Yerkes traveled to Canada (Yerkes 1921 , p. 789). This led to a moral panic to study the contribution of Canadian psychologists analogous to that generated in Britain by the poor to the war effort, and later that month, at the meeting performance of the British army during the of the APA Council in Philadelphia, he formed a Second Boer War and promoted similar fears committee to explore ways in which American about the decline in “national effi ciency.” These psychologists could contribute. This committee, fears were stoked by books like Madison Grant’s whose members included Cattell, Hall, The Passing of the Great Race (Grant 1916 ), Thorndike, and Watson, suggested the develop- Goddard’s Human Effi ciency and Levels of ment of psychological tests to facilitate the selec- Intelligence (Goddard 1920 ), and Carl Brigham’s tion of offi cers and the discharge of feebleminded A Study of American Intelligence (Brigham 1923, recruits. Yerkes promptly formed the Committee with a foreword by Yerkes), which claimed that on Methods of Psychological Examining for the average intelligence of Americans had Recruits, whose members included Goddard and declined since the turn of the century due to the Terman, which spent 2 weeks at Vineland creat- overbreeding of recent immigrants from Eastern ing group intelligence tests for the army and run- and Southern Europe, which was itself seen as a ning trials at local institutions and army bases. sign of feeblemindedness. When the army fi nally approved what became While there were those who urged caution in known as the Army Testing Project, some 400 drawing conclusions from the army data (Boring commissioned psychologists administered group 1923a ) and those who disputed Goddard’s intelligence tests to some two million soldiers (Goddard 1919 ) claim that they demonstrated between 1917 and 1919: the Alpha 12 test to liter- that the average intelligence of adult Americans ate soldiers and the Beta (pictorial) test to illiter- was below the level of moron (Freeman 1922 ; ates. The mass testing was of doubtful military Lippmann 1922 ), they were drowned out by more utility, and the army discontinued the project at strident calls for immigration quotas and pro- the end of the war (Samelson 1977 ), although it grams of sterilization of the feebleminded. did serve to promote the public perception of the In 1924, the Congress passed the National utility of intelligence testing and the professional Origins Act, which imposed quotas on the national- status of psychologists––as Cattell put it, the ity of immigrants based upon the 1890 census (i.e., Army Testing Project “put psychology on the before the wave of Eastern and Southern European map” (Cattell 1922 , p. 5). Yerkes was appointed immigration at the turn of the century). Harry to the National Research Council, where he Laughlin (1880–1943) of the Eugenics Records worked on the development of the National Offi ce had testifi ed before the Congressional Intelligence Test (National Research Council Immigration and Naturalization Committee earlier 1920 ), administered to over seven million school- that year, claiming that the American gene pool children in the 1920s. was being “polluted” by the growing numbers of However, the most important outcome of the intellectually inferior immigrants. Army Testing Project was the alarming fi nding By the end of the 1920s, close to 30 states 13 that around half the army recruits tested at or had passed laws legitimizing the compulsory below the level of moron. In his fi nal report on sterilization of the feebleminded, a procedure the project, Yerkes concluded that 13 The fi rst state law licensing the compulsory sterilization of confi rmed criminals, idiots, imbeciles, and rapists was 12 Carl Brigham (1890–1943), who worked on the Army passed by Indiana in 1907, on the recommendation of Alpha test, later adapted it as an admissions test when he Harry Clay Sharp (1869–1940), a pioneer of vasectomy, joined the faculty at Princeton University, where it became who originally used the procedure to treat masturbation, known as the Scholastic Admissions Test (SAT) and later which he believed to be a major cause of intellectual as the Scholastic Assessment Test. degeneracy. The 1907 Indiana Law was struck down by 9 Intelligence Defi ned: Wundt, James, Cattell, Thorndike, Goddard, and Yerkes 133 infl icted on at least 12,000 inmates by 1930. In hot to handle and shifted his attention back to 1926, the US Supreme Court put its seal of comparative psychology. Nonetheless, the quota approval on the eugenic use of the procedure by system established by the National Origins Act sanctioning the sterilization of Carrie Buck remained in place until 1965, and the sterilization (1906–1983), a 17-year-old girl who was an laws remained on the books for many years later, inmate of the Virginia Colony for the Epileptic with the State of Virginia repealing the last law as and Feebleminded. Carrie and her mother, who late as 1981. was also an inmate, were judged to be feebleminded, as was Carrie’s illegitimate daughter Vivian, and Carrie was sterilized by court order Conclusion: So What Is Intelligence? under Virginia’s Eugenical Sterilization Act.14 The case was appealed to the US Supreme Court, Beyond the Sturm und Drang over the inheri- which upheld the Virginia Law, with Oliver tance of intelligence and feeblemindedness, Wendell Holmes Jr. (1841–1935), who wrote the what was learned about the nature of intelligence majority opinion in the case of Buck vs. Bell in itself? Not much it seemed. In a 1921 study pub- 1927, famously declaring: lished in the Journal of Educational Psychology It is better for all the world if instead of waiting to that canvassed the defi nitions of intelligence by execute degenerate offspring for crime, or let them 14 “experts” (Thorndike et al 1921 , there was starve for their imbecility, society can prevent remarkably little agreement, and many practitio- those who are manifestly unfi t from continuing ners of intelligence testing rested content with their kind. Three generations of imbeciles are enough. (Buck vs. Bell 1927, p. 207) Boring’s (1923b) operational defi nition of intelligence as what intelligence tests measure, which Eventually, the eugenic excesses of the seemed suffi cient to guarantee its reference Nazis––who did endorse euthanasia on a grand while saying nothing about its nature. scale––demonstrated the dangers of this over- Nevertheless, one can discern at least the outline reaching science and dampened the enthusiasm of a common theme running through the works of many former supporters. Many original psy- of Wundt, James, Spearman, Thorndike, and the chological advocates, such as Goddard and developers of the Binet-Simon tests of intelli- Brigham, recanted their original positions. 15 And gence. Wundt’s apperceptive synthesis of rela- Yerkes, although he continued to believe in tional elements, James’ discernment of genetically determined racial differences in intel- similarities, Spearman’s “education” of relations ligence, recognized that the topic had become too and correlates, Thorndike’s formation of connections, and Binet and Simon’s trio of “compre- the state supreme court in 1921, but was quickly replaced hension, invention, and direction” all point to by another that survived legislative challenge. some form of cognitive achievement involving 14 There is little evidence that Carrie was feebleminded, the discernment or determination of connections and she seems to have been institutionalized on grounds of sexual promiscuity––commonly treated as an indicator and confi gurations. However, it would take the of feeblemindedness––on the basis of her pregnancy, development of cognitive psychology in the later despite the fact that this was the result of a rape by a part of the century to begin to specify the rele- nephew of the foster parents who committed her. Nor is vant cognitive processes, while still leaving open there much evidence that Carrie’s daughter Vivien was feebleminded. Although she died at the age of eight of an and contentious the original question of whether intestinal disease, her fi rst grade report card indicated that “higher” cognitive processes such as logical rea- she was a solid B student, with an A in deportment, who soning and abstract thought are continuous with once made the honor role. “lower” forms of association and whether human 15 Brigham also repudiated the use of the Scholastic intelligence is continuous with animal intelli- Achievement Test as a measure of intelligence (Angier et al. 1926) and opposed its use as the basis of a National gence (Mitchell et al. 2009 ; Penn et al. 2008 ; Educational Testing Service (Brigham 1938 ). Shanks 2007 ) . 134 J.D. Greenwood

Grant, M. (1916). The passing of the great race. New York: References Scribner’s. Guilford, J. P. (1967). The nature of human intelligence . Angier, R. P., MacPhail, A. H., Rogers, D. C., Stone, C. J., New York: McGraw-Hill. & Brigham, C. C. (1926). Scholastic aptitude tests: A Hall, G. S. (1923). Life and confessions of a psychologist . manual for the use of schools . New York: College New York: Appleton. Entrance Examination Board. Intelligence and Its Measurement: A Symposium. (1921). Boakes, R. (1984). From Darwin to behaviorism: Psychology Journal of Educational Psychology, 12, 123–147, and the minds of animals . Cambridge, MA: Cambridge 195–216, 271–275. University Press. Jackson, J. H. (1931). Selected writings of Hughlings- Boring, E. G. (1923a). Facts and fantasies of immigration. Jackson (Ed., J. Taylor, Vols. 1–2). London: Hodder & New Republic, 36 245–246. Stroughton. Boring, E. C. (1923b). Intelligence as the tests test it. New James, W. (1890). The principles of psychology (Vol. 1–2). Republic, 36 , 35–37. New York: Holt. Boring, E. G. (1957). A history of experimental psychol- Ladd, G. T. (1895). Philosophy of mind: An essay on the ogy (2nd ed.). New York: Appleton Century Crofts. metaphysics of psychology . New York: Scribner’s. Brigham, C. C. (1923). A study of American intelligence . Lippmann, W. (1922). The mental age of Americans. New Princeton: Princeton University Press. Republic, 32 , 213–215. Brigham, C. C. (1938). Letter to J. B. Conant, 3 Mitchell, C. J., Houwer, J. D., & Lovibond, P. F. (2009). January. Princeton: ETS Archive:Brigham file. The propositional nature of human associative learn- Buck v. Bell – 274 U.S. 200 (1927). ing. Behavioral and Brain Sciences, 32 , 183–246. Cattell, J. M. (1885). Über die Zeit der Erkennung und Morgan, C. L. (1894). Introduction to comparative Benennung von Scriftzeichen, Bilden und Farben. psychology . London: Walter Scott. Philosophische Studien, 2 , 635–650. National Research Council. (1920). Fourth annual report . Cattell, J. M. (1886). On the time it takes to see and name Washington, DC: Government Printing Offi ce. objects. Mind, 11 , 63–65. O’Donnell, J. M. (1985). The origins of behaviorism: Cattell, J. M. (1890). Mental tests and measurements. American psychology, 1870–1920. New York: Mind, 15 , 373–381. New York University Press. Cattell, J. M. (1922). The fi rst year of the psychological Penn, D. C., Holyoak, K. J., & Povinelli, D. J. (2008). corporation. Report at annual meeting of the psycho- Darwin’s mistake: Explaining the discontinuity logical corporation, December 1. Lewis M. Terman between human and non-human minds. Behavioral Papers, Stanford University Archives. and Brain Sciences, 31 , 109–178. Darwin, C. (1859). On the origin of species by means of Romanes, G. J. (1882). Animal intelligence . New York: natural selection . London: John Murray. Appleton. Darwin, C. (1871). The descent of man and selection in Samelson, F. (1977). World War I intelligence testing and relation to sex . London: John Murray. the development of psychology. Journal of the History Davenport, C. (1911). Eugenics: The science of human of the Behavioral Sciences, 13 , 274–282. improvement by better breeding. New York: Henry Sechenov, I. (1863). Refl exes of the brain (ed.: G. Gibbons, Holt & Co. Deary Trans: S. Belsky). Cambridge, MA: MIT Press, 1965. Deary I. J., Lawn, M., & Bartholomew, D. J. (2008). Shanks, D. R. (2007). Associationism and cognition: A conversation between Charles Spearman, Godfrey Human contingency learning at 25. Quarterly Journal Thomson, and Edward L. Thorndike: The International of Experimental Psychology, 37B , 1–21. Examinations Inquiry Meetings 1931–1938. History Spearman, C. (1904). General intelligence, objectively of Psychology, 11 , 122–142. determined and measured. The American Journal of Freeman, F. N. (1922). The mental age of adults. Journal Psychology, 15 , 201–292. of Educational Research, 6 , 441–444. Spearman, C. (1925). The abilities of man: Their nature Galton, F. (1869). Hereditary genius. London: Macmillan. and measurement . London: McMillan. Galton, F. (1883). Inquiries into human faculty and its Spencer, H. (1855). Principles of psychology . London: development . London: Macmillan. Longmans. Galton, F. (1888). Co-relations and their measurement: Stern, W. (1914). The psychological methods of testing Chiefl y from anthropometric data. Proceedings of the intelligence (trans: G. M. Whipple). Baltimore: Royal Society, 45 , 135–145. Warwick and York. Galton, F. (1889). Natural inheritance . London: Sternberg, R. J. (2003). Intelligence. In D. K. Freedheim Macmillan. & I. B. Weiner (Eds.), Handbook of psychology Goddard, H. H. (1912). The Kallikak family: a study in (History of psychology, Vol. 1). Hoboken: Wiley. the heredity of feeblemindedness . New York: Terman, L. M. (1906). Genius and stupidity. Pedagogical Macmillan. Seminary, 13 , 307–373. Goddard, H. H. (1919). Psychology of the normal and Thomson, G. H. (1916). A hierarchy without a general subnormal . New York: Dodd, Mead. factor. British Journal of Psychology, 8 , 271–281. Goddard, H. H. (1920). Human effi ciency and levels of Thorndike, E. L. (1898). Animal intelligence: An intelligence . Princeton: Princeton University Press. experimental study of the associative processes in 9 Intelligence Defi ned: Wundt, James, Cattell, Thorndike, Goddard, and Yerkes 135

animals. Psychological Review, Monograph Wundt, W. (1863) Lectures on human and animal Supplement, 2, (8), 1–109. psychology (trans: J. E. Crighton & E. B. Titchener). Thorndike, E. L. (1903). Heredity, correlation and sex dif- New York: Macmillan, 1894. ferences in school abilities . New York: Macmillan. Yerkes, R. M. (1905). Animal psychology and the crite- Thorndike, E. L. (1911). Animal intelligence . New York: rion of the psychic. Journal of Philosophy, Psychology Hafner. and Scientiﬁ c Methods, 2 , 527–528. Thorndike, E. L., Henmon, V. A. C., & Buckingham, B. Yerkes, R. M. (Ed.) (1921). Psychological examining in R. (eds.) (1921). Intelligence and its measurement: A the United States Army. Memoirs of the National Symposium. Baltimore. Academy of Sciences, 15 , 1–890. Thorndike, E. L., Bregman, E. D., Cobb, M. V., & Yerkes, R. M. (1923). Introduction to C. C. Brigham’s A Woodyard, E. I. (1926). The measurement of intelli- study of American intelligence. Princeton: Princeton gence . New York: Teachers College Press. University Press. Watson, J. B. (1909). A point of view in comparative Yerkes, R. M., Bridges, J. W., & Hardwick, R. S. (1915). psychology. Psychological Bulletin, 6 , 57–58. A point scale for measuring mental ability . Baltimore: Watson, J. B. (1913). Psychology as the behaviorist views Warwick & York. it. Psychological Review, 20 , 158–177. Zenderland, L. (1998). Measuring minds: Henry Herbert Wissler, C. (1901). The correlation of mental and physical tests. Goddard and the origins of American intelligence Psychological Review Monograph Supplements, 3(6). testing . Cambridge, MA: Cambridge University Press. Piaget’s Theory of Intelligence 10 Ulrich Müller , Kayla Ten Eycke , and Lesley Baker

[I]ntelligence, the most plastic and at the same time most durable structural equilibrium of behavior, is essentially a system of living and acting operations (Piaget 1976a , p. 7)

Readers may be puzzled by the quote above because sion. In this book, there is no description of any the way in which Piaget uses the term intelligence is method by means of which intelligence might be rather broad and does not correspond with the more assessed in order to assign individuals a particular tightly defi ned way in which it is used today. number (their IQ), nor does it refer to individual dif- Puzzlement likely will make way to outright confu- ferences in intelligence, nor does it identify specifi c sion when the reader attempts to make sense of cognitive processes such as working memory or pro- Piaget’s defi nition of intelligence: “Intelligence con- cessing speed as being involved in intelligence, nor stitutes the state of equilibrium towards which tend is there a discussion of a specifi c neural basis of all the successive adaptions of a sensori-motor and intelligence, nor is any attempt being made to under- cognitive nature, as well as all assimilatory and stand intelligence in the context of the selection pres- accommodatory interactions between the organism sures exerted by the environment of evolutionary and the environment” (Piaget 1976a , p. 11). This adaptedness. Instead, in the fi rst part of the book the quote is taken from the book The Psychology of reader encounters a classifi cation of theories of intel- Intelligence, but despite its title, the following pages ligence, a discussion of old theories of intelligence of the book might do little to lift the reader’s confu- (e.g., Russell’s theory of logical atomism, the rzburg school of thought psychology, GestaltصW psychology), and remarks on the relation between logic and psychology that culminate in a mathematical treatment of higher states of equilibrium. The goal of this chapter is to clarify Piaget’s U. Müller (*) • L. Baker theory of intelligence. We attempt to do this by Department of Psychology, P. O. Box 1700 STN fi rst showing how Piaget’s conception of intelli- CSC , University of Victoria , Victoria , BC V8W 2Y2 , Canada gence builds on that of his contemporaries. Next, e-mail: [email protected]; [email protected] we describe how his conception of intelligence K. Ten Eycke is contextualized within his larger theoretical Behavioural Research Unit , Alberta Children’s framework, his genetic epistemology. This leads Hospital, University of Calgary , 2888 Shaganapi Trail us to a discussion of core features of Piaget’s NW , T3B 6A8 Calgary , AB , Canada theoretical framework such as self-organization, e-mail: [email protected]

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 137 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_10, © Springer Science+Business Media New York 2015 138 U. Müller et al. assimilation and accommodation, and operative 1963 , pp. 395–407, for an extended discussion of and fi gurative aspects of intelligence. We next Claparède’s theory of intelligence). summarize the four major stages of the develop- Given these contradictory ways of demarcat- ment of intelligence. Finally, we describe how ing intelligent and unintelligent behaviors, Piaget Piaget’s theory of intelligence relates to semiotic (1976a , p. 10) saw himself confronted with the function, affectivity, and social interaction. following alternatives: [E]ither we must be satisfi ed with a functional defi - nition at the risk of encompassing almost the entire Piaget’s Defi nition of Intelligence range of cognitive structures, or else we must choose a particular structure as our criterion, but the choice remains arbitrary and runs the risk of Similar to his contemporaries (e.g., see Binet overlooking the continuity which exists in reality. 1894 , 1975 ), Piaget uses the term intelligence in a broad sense, a common practice commented on Piaget (1976a ) resolved this issue by combin- already by the French historian Taine (1872 , ing functional continuity and structural change p. vii): “If I am not mistaken, we mean nowadays and defi ned intelligence in terms of the direction by intelligence, what was formerly called in which development moves, “without insisting Understanding or Intellect—that is to say, the on the question of boundaries, which become a faculty of knowing.” Refl ecting such a broad matter of stages of successive forms of equilib- sense of intelligence, Piaget (1976a , pp. 8–9) rium” (p. 10). From the functional perspective, identifi es a central feature of intelligence as a behavior becomes more intelligent as the spatio- movement toward increasing spatiotemporal dis- temporal distance between person and world tances in the functional interaction between sub- increases. At the same time, functional continuity ject (i.e., person, animal) and world. implies that there is neither an absolute starting Piaget (1976a , p. 9) notes that a problem point for intelligence nor a predetermined end arises when we want to draw a line and precisely point: “it is an ultimate goal, and its origins are demarcate behavior that is intelligent from indistinguishable from those of sensori-motor behavior that is not. He illustrates this by com- adaptation in general or even from those of bio- paring a variety of different defi nitions of intelli- logical adaptation itself” (Piaget 1976a , p. 7). gence. For example, Karl Bühler ( 1933 ) From the point of view of structure, earlier stages distinguishes between three stages of purposeful of development are characterized by a lack of behavior: instinct, training, and intellect. reversibility, with intelligence moving toward Instinctive behaviors are innate, rigid, and exe- increasingly reversible mobility. Let us now cuted in the same manner in all members of a examine this defi nition in the context of Piaget’s species. Training involves trial-and-error learn- theoretical framework. ing such that successful behaviors are reinforced and unsuccessful ones are eliminated, which allows individuals to adapt to new situations. Genetic Epistemology True intelligence emerges at the stage of the intellect at which the individual “makes discover- Piaget (1950a , b , c ) called his theoretical frame- ies by means of insight and refl ection . work genetic epistemology. Here, genetic speci- INVENTION, in the true sense of the term, is the fi es developmental. Epistemology refers to the biological achievement of the intellect” (Bühler study of the nature, sources, scope, and validity 1933 , p. 10; emphases in original). Piaget ( 1976a , of knowledge. Usually, epistemology is consid- pp. 9–10) contrasts Bühler’s defi nition with that ered a branch of philosophy, and empirical of Claparède (1917 ), who argues that the defi ning research is argued to have no bearing on episte- feature of intelligence is the adaptation to new mological questions (Hamlyn 1971 ). Piaget, situations. As a result, Claparède considers trial- however, did not believe that epistemological and-error behavior as intelligent behavior and issues fall under the sole jurisdiction of philoso- distinguishes it from instinct and habit (see Piaget phy. One important reason for this is that knowl- 10 Piaget’s Theory of Intelligence 139 edge itself is in constant fl ux and always remains used the necessary truth of arithmetic judgments as incomplete. In this context, Piaget (1972a , p. 2) an example to illustrate this irreducibility: approvingly quotes the neo-Kantian philosopher [the] truth of 2 + 2 = 4 is not the ‘cause’ of the truth Natorp (1910 , pp. 14–15): of 4–2 = 2 in the same way that a cannon causes the movement of two billiard balls, or a stimulus is one Like Kant, we start with the actual existence of of the causes of a reaction: the truth … of 2 + 2 = 4 knowledge and seek the basis from there. But what ‘implies’ that of 4–2 = 2, which is quite a different is this existence since, as we know, knowledge is matter (Piaget 1968 , p. 187). constantly evolving? Progression, method is everything… in consequence, the existence of knowledge cannot be comprehended except as a fi eri (i.e., to be The normative dimension of knowledge, then, made, to become; authors’ note). This fi eri alone is puts a further constraint on the genetic explana- the fact. Any entity (or object) which knowledge tion of knowledge: the explanation must provide attempts to crystallize must dissolve again in the for the possibility of the emergence of normative current of development. It is in the last phase of this development, and in this alone, that we have the knowledge. Causal, reductionist explanations right to say: “this is (a fact).” What we can and must (e.g., in terms of neurophysiological causality) seek, then is the law underlying this process. will not suffi ce. However, Piaget did not deny that states of knowledge and intelligence have a Hence, if constant evolution is constitutive of biological dimension: intelligence has a “dual scientifi c knowledge, as witnessed in the natural nature,” it is “both biological and logical” (Piaget and human sciences, and even in logic and math- 1976a , p. 3). Indeed, the major challenge is to ematics (Piaget 1950a , b , 1972a , b ), then the explain how intelligence as a system of living and study of the conditions of the possibility of acting operations can be rooted in biology yet knowledge must include the development of incessantly generate novelties and result in rigor- knowledge. The study of the development of ous (i.e., logically necessary) and normative knowledge, in turn, falls under the purview of the knowledge. For Piaget, this challenge can be met empirical sciences. only when life is conceived as self-organization. Thus, although Piaget is mostly known as a child psychologist, the study of cognitive development in children was for Piaget only a means Self-Organization to address epistemological issues (Vonéche and Vidal 1985 ). Piaget’s focus on epistemological It is generally accepted that psychological devel- issues explains why he was not at all interested in opment has a biological basis. The question is determining the cognitive level of an individual how the relation between biology and psycho- child or in examining interindividual differences logical development should be conceptualized. in cognition (see Bringuier 1980, p. 86). Instead, Piaget used the concept of self-organization to he was interested in what is common to all per- characterize the relation between biology and sons at a specifi c level of thinking—what Piaget psychological development (see Piaget 1971 ; see ( 1987) referred to as the epistemic subject. The Boom 2009 ; Chapman 1992). At the biological epistemic subject is able to attain states of knowl- level, self-organization is the process by which a edge, and the goal of genetic epistemology is to system perpetually reconstitutes its processes explain how the attainment of this knowledge is (e.g., metabolic cycles) and elements (e.g., cells) possible. Knowledge, however, “is not to be in order to preserve its continuous functioning. naively equated with mere belief (or the brute This self-organizing activity is not something factual existence of a cognitive structure): additional or external to the living system; rather, knowledge has an inescapable normative dimen- it pervades matter as active form and it arises sion, one concerning concepts like evidence, when the reciprocal interactions between the ele- objectivity, rationality, validity, truth, etc.” ments or subsystem of the system lead to the for- (Kitchener 1993 , p. 141). mation of a higher-order system that regulates the Piaget argued that the normative dimension of elements or subsystems (Piaget 1971 , p. 327 fn.). knowledge cannot be reduced to causality. He often Living systems are self-organizing systems; in 140 U. Müller et al. exchange processes with their environment, they time, constitute the most complex instruments for spontaneously reproduce their organization. regulating the exchange with the environment The idea of self-organization can be traced (Piaget 1971 ). Moreover, cognitive functioning back to Aristotle, and a brief excursion into how refl ects reason in a double sense: it is the product he explicated self-organization will help clarify of reason that is intrinsic to nature (i.e., in the its conceptual status. For Aristotle, form and mat- logic of self-organization) and through cognitive ter are inseparable; matter is always already orga- functioning reason in nature becomes conscious nized (Aristotle 1956 , 412b, 4–25). Each thing of itself. has its substantial form or entelechy; psyche is Thus, similar to Binet (Binet and Simon 1916 , the entelechy of living beings. Realized structure pp. 141, 153; see Bennour and Vonéche 2009 ), (entelecheia) of natural kinds must be viewed as Piaget considered intelligence a biological adap- the result of immanent processes (their energeia) tation. Self-organizing activity is the biological (Aristotle 1971 , 1046a:30, 1050a:22). foundation and origin of intelligence, and cogni- Psyche is best interpreted as self-organizing tive processes are the outcome of and extend the activity that preserves its own organization processes of organic self-organization. Cognitive (Aristotle 1971 , 1048b27; see Hübner 1999). processes extend the processes of organic self- Organic self-preservation has its goal inside itself, organization by using and adapting to new cir- and this goal is achieved with the occurrence of cumstances the different systems of organic organic self-preservation. Self- organizing activity self-regulation that can be found on the genetic, preserves its organic body and the disposition that morphogenetic, physiological, and nervous lev- it exercises (Aristotle 1956 , 417b2-7; 1971 , els. In support of this claim, Piaget (1971 ) 1048b18-34). Thus, psyche fulfi lls the body and describes many functional and structural analo- preserves the body and the body’s disposition to gies between cognitive and organic functioning. be psyche, and, because psyche is organic self- Central among these analogies is the triad of preservation, body and psyche possess the same assimilation, accommodation, and scheme. being. As a consequence, Aristotle characterizes a living being by the fact that fi nal cause and formal cause are fused together: the goal or telos is the Assimilation, Accommodation, form of the thing; the form of a living being is the and Scheme continuous self-organizing activity; therefore, the telos of the living being is to preserve its form. A The complementary functions of assimilation self- organizing system is like a physician healing and accommodation describe the general charac- himself (Aristotle 1970 , 199b, 27–32). Machines, teristics of the exchange between organism by contrast, are not self-organizing systems; their and environment (Piaget 1963 , 1970 , 1971 ). functioning does not result in their reconstitution, Assimilation is the aspect of an organism’s and formal cause and fi nal cause are not fused activity wherein elements of the environment are together. integrated into the organism’s preexisting organi- Piaget expanded on Aristotle by putting an zational structures (i.e., the relations between ele- evolutionary spin on the idea of self- organization. ments). Accommodation, on the other hand, As he put it, “the very nature of life is constantly provides the material for the structuring activity to overtake itself” (Piaget 1971 , p. 362). of assimilation. Accommodation is the aspect of Phylogenetically, the interaction between organ- the activity wherein an organism’s existing ism and environment leads to the emergence of schemes are differentiated and modifi ed in higher-order self- regulatory processes, which, on response to the environment. For example, a the level of cognitive functioning, refl ect the basic preexisting metabolic cycle assimilates particular mechanisms of self-organization and, at the same nutrients by breaking them down into the elements 10 Piaget’s Theory of Intelligence 141 that contribute to the continued functioning of the “Assimilation is hence the very functioning of the living system. The assimilatory cycle needs to be system of which organization is the structural modifi ed when the organism encounters a new aspect” (Piaget 1963 , p. 410). At the same time, nutrient (accommodation) (Piaget 1963 ). Assimi- the incorporation of new elements leads to the lation and accommodation maintain the equilib- modifi cation of the scheme and thus to accom- rium between an organism and its environment. modation. Accommodation brings about adapta- Assimilation and accommodation at the psy- tion to the environment, but this adaptation is chological level extend the physiological interac- always a function of the structuring activity of tions between the organism and the environment assimilation. The act of assimilating objects to because their functioning no longer depends on schemes is rather complex because it involves the incorporation of material elements but now affect, sensation, (internalized) motor elements, incorporates informational content (Piaget 1963 , and perception (Piaget 1963 , 1981 ). 1971). At the psychological level, schemes are characterized by what is repeatable in actions and thought processes (i.e., internalized actions). Operative and Figurative Aspects Assimilation refers to the incorporation of new of Intelligence information into already existing schemes, a process giving meaning to the content (Piaget 1963 , Closely related to the concepts of assimilation 1985). For example, when a baby grasps a rattle, and accommodation are Piaget’s (1969 ) notions of the rattle is assimilated to the grasping scheme fi gurative and operative aspects of intelligence. The and thereby attains the functional meaning of fi gurative aspect of intelligence includes the func- being “graspable.” Assimilation always uses the tions of perception, imitation, imagery, and (in existing psychological schemes; its functioning part) language that are supplied by the accommo- carries the history of the subject’s interaction datory aspect of activity (Piaget 1969 ; Piaget and with the world into each particular act. For exam- Inhelder 1971 ). The fi gurative aspect provides sig- ple, an infant who has differentiated various ways nifi ers, which, in turn, bear on the “states” of reality of interacting with the rattle will have different and provide data on which the structuring activity action potentialities available compared to an of assimilation acts. For example, an infant may infant who has not. perceive a rattle and, assimilating it to an action Accommodation refers to the modifi cation of scheme, she recognizes rattle as something that can existing schemes to account for particular fea- be shaken (i.e., the sight of the rattle serves as a tures of the object or situation. Because schemes signifi er of what can be done with it). are structures with varying degrees of generality, In contrast, the operative aspect of intelligence applying them to particular situations always refers to the transforming and form-giving, or requires an adjustment or accommodation. structuring, aspect of knowledge (Piaget and Accommodation thus particularizes the general Inhelder 1971 ). It includes sensorimotor actions, schemes, supplies them with specifi c content, internalized actions that are carried out mentally, and modifi es them in doing so (e.g., the preexist- and operations. Operations are internalized ing grasping scheme needs to be modifi ed, actions that have become reversible because they becoming more specifi c to take into account the are organized in a structure such that each opera- particular spatial position of the rattle). tion is coordinated and can be carried out simul- Assimilation, accommodation, and scheme taneously with an another operation that cancels are inseparable. Assimilation is always a struc- it out (e.g., uniting and dissociating elements, turing activity because it involves integrating adding and subtracting; Piaget 1976a ). Piaget content into existing schemes; thus, structures do (1976a ; Piaget et al. 1992 ) used different mathe- not exist independently of structuring activity: matical and logical tools to analyze the properties 142 U. Müller et al. of the operational structures. Piaget (1973 , 1974) which he focused in more detail on the specifi c was very clear that the subject is only aware of processes involved in equilibration. Although the outcome (i.e., his or her performance) of Piaget identifi ed several processes as playing an these structures, but not of the structures important role in equilibration, such as dialec- themselves. tics, contradiction, affi rmation and negation, the The operative aspect of intelligence trans- generation of possibilities, and the process of forms subject-object relations by inserting the becoming aware, his theory of equilibration data provided by the fi gurative functions into remains unfi nished (Campbell 2009 ; Piaget increasingly complex structures. In other words, 1976b , 1980 , 1987). Here we focus on the role the operative activity of the human mind results of refl ecting abstraction in equilibration, as in the construction of more and more complex refl ecting abstraction is central to the construc- relations (spatial, causal, logical, etc.) between tion of more powerful knowledge structures person and world. The operative aspect of intel- (Piaget 1971 , 1985 , 2001 ). ligence, then, is central to understanding the Refl ecting abstraction is an elaborative pro- kinds of qualitative changes that occur in cess by which children discover the structural Piaget’s account of cognitive development. The aspects of their cognitive activity (Piaget 2001 ). fi gurative aspect is an auxiliary of the operative For instance, putting marbles, one after the other, aspect of intelligence in that it provides knowl- in a receptacle is an action with several structural edge about states that are coordinated and trans- aspects, one of which is based on the creation of formed by operations. For example, the meaning a serial order and another on the creation of a set of what the subject perceives is relative to his or with a growing number of elements. By becom- her action tendencies, and perceptual activities ing aware of the relations between and coordina- (i.e., what the subject pays attention to) are tion of their actions, children abstract structure dependent on the subject’s operative develop- (the coordinatory or operative aspect of actions) ment (Piaget 1969 ). from content and, in turn, project this structure to a higher cognitive level. Piaget (1950a , 1972a , 2001 ) suggests that the Equilibration general coordinations of actions (e.g., putting things together, establishing correspondences, At each point in development, children are in a ordering) are the source of logico-mathematical state of equilibrium with the environment, char- knowledge. For example, to understand the com- acterized by a particular balance of assimilation mutativity of addition (3 + 2 = 5 = 2 + 3), the child and accommodation. Development is a process needs to put down objects in a different orders that leads to increasingly more stable (complete (e.g., fi rst 3, then adding 2; after that 2, adding 3) and consistent) forms of equilibrium. Piaget and then realize that the total remains the same (1985 ) termed this process equilibration. (i.e., the product of the actions is independent of Equilibration must ensure two things: (1) It must the particular order in which the actions are exe- always open up new possibilities (as life is cre- cuted; Piaget and Inhelder 1976 ). With the emer- ative); (2) it must conserve previous structures as gence of the semiotic function (see below), the substructures in new and elaborated structures. knowledge abstracted from the coordination of The second requirement is necessary to account actions becomes internalized, and the commuta- for the fact that logico-mathematical knowledge tivity of addition can be deduced by mental does not become invalid with the construction of operations. new knowledge (Piaget 1972a ). Equilibration The mechanism of refl ecting abstraction then thus must reconcile the two contradictory tenden- ensures that development has an intrinsic logic cies of openness and closure of structures. and proceeds by way of successively conceptual- The theory of equilibration takes central izing the structures or forms of knowledge under- place in Piaget’s later work (Boom 2009 ), in lying previous knowing levels (Piaget 1971 , 10 Piaget’s Theory of Intelligence 143

2001). Thus, the form of stage n becomes the nor preformed, but an active construction that “at content of stage n + 1. With each new and higher its origin, neither arises from objects nor from the stage, the forms become increasingly abstract. subject, but from interactions—at fi rst inextrica- Through the mechanism of refl ecting abstraction, ble—between the subject and the object” (Piaget then, development proceeds by way of succes- 1970, p. 704). It is in the course of these interac- sively conceptualizing and reconstructing the tions that the subject becomes aware of herself as knowledge structures underlying previous know- the (social and physical) world provides resis- ing levels, thus ensuring the generativity and tance to her projects, and she constructs an rigor of knowledge. increasingly complex knowledge of the world as she coordinates her actions and operations (Piaget 1954 ). Constructivism At the psychological level, constructivism amounts to a pragmatist approach to knowledge Piaget’s way of conceptualizing intelligence because Piaget (1970 , p. 704) emphasized action and knowledge clashes with the way it is con- as the source of knowledge: “in order to know ceptualized by empiricism. Piaget himself was objects, the subject must act upon them, and there- ardent in his opposition to empiricism (e.g., fore transform them: he must displace, connect, Piaget and Inhelder 1976 ). According to Piaget, combine, take apart and reassemble them.” Action empiricism conceives of human beings as pas- is goal-directed and, at least early in life, aims at sive, emphasizing sense perception, which pro- success and not truth; it is a lived and not a con- vides copies of reality, and association as major templative intelligence (Müller 2009 ). Action also sources of knowledge. Piaget argued that empir- transforms reality itself: “The construction of an icists misconstrue the fundamentally active rela- electronic machine or a sputnik not only enriches tion between subject and world as a passive, our knowledge of reality, it also enriches reality causal relation. Furthermore, Piaget (1970 , itself, which until then did not include such objects. 1972a , b ) contended that the idea that knowl- This creative nature of action is central” (Piaget edge is a copy of reality is fl awed because there and Inhelder 1976 , p. 33). would be no way to evaluate the accuracy of Constructivism combines genesis (empiri- such copies because they cannot be directly cism) and structuralism (nativism): every genesis compared to reality itself. Rather, he was infl u- originates from one knowledge structure and enced by Kant’s (1929 ) idea that objectivity is results in another structure, and, conversely, constituted by the subject. Kant argued that our every structure has a genesis (Piaget 1967 ). We intuition (i.e., sensibility) and understanding now turn to the stages defi ned by the succession use a priori (i.e., independent of all experience) of structures. forms and categories, which are the condition of the possibility for experiencing objectivity. Piaget (1963 , pp. 376–395) subscribes to the Stages in the Development ordering and organizing function of the mind— of Intelligence indeed, this is implied in the notion of assimilation. However, Piaget believed that the forms In standard psychology textbooks (e.g., Berk and categories are not a priori or innate but 2012 ), Piaget is typically portrayed as a stage rather undergo development as a result of the theorist who claimed that stages are general subject’s interaction with the world—which is structures that defi ne a child’s behavior in each implied in the notion of accommodation. area of cognitive functioning and that age is a cri- As an alternative to empiricist and nativist terion for stage. Consequently, it is argued that interpretations of knowledge, Piaget proposed a Piaget’s theory is fl awed because empirical evi- constructivist interpretation, according to which dence shows that at any point in development, knowledge is neither a simple recording of reality children’s behavior is heterogeneous and not 144 U. Müller et al. homogeneous (e.g., they may reason at a Sensorimotor Stage Piaget (1963 ; see Müller preoperational level in one conservation task, and 2009 ) termed the developmental period during at a concrete-operational level in another) and approximately the fi rst 18 months of life that particular stages emerge earlier than Piaget sensorimotor intelligence. It plays a key role in would predict. bridging the gulf between the biological level of This portrayal of Piaget’s stage theory is functioning and rational thought. Sensorimotor utterly incorrect (Chapman 1988 ; Smith 1993 ). intelligence is a practical, embodied intelligence Piaget did not claim that stages are characterized on the basis of which infants interact with the by homogeneity, and, in fact, Piaget often made world through perception–action cycles. At the the opposite point that variability should be sensorimotor stage, meaning is originally expected (Chapman 1988 ). Furthermore, vari- embedded in unrefl ective activities; objects ability in children’s performance on structurally have a functional, practical meaning, they are similar tasks is entirely consistent with the basis things at hand, utensils for practical use or of Piaget’s grounding assumption that thought manipulation. Infants employ action schemes originates in action. Based on this assumption, like sucking, pushing, hitting, and grasping to cognitive structures should, at fi rst, be context explore and manipulate the world. At the outset, and content specifi c. That is, cognitive structures the newborn has no self- consciousness and no cannot be separated from their content, and clear awareness of what effects she herself although structures involving different content produces through actions on the world and what (e.g., number and volume), may be of the same effects occur independently of her actions. By logical form, they develop independently in a coordinating her actions and applying them in functional sense through the child’s activity with the social domain (imitation), the infant these different areas of content. gradually learns to distinguish between self, Stages are also not defi ned in terms of age. other persons, and world. Rather, they are defi ned in terms of performance Piaget traced the process of differentiation on particular tasks that Piaget analyzed in terms and coordination of action schemes through of the operations and structure they require. He several sensorimotor substages. Drawing on acknowledged that the age of acquisition of Claparède’s ( 1917 ) suggestion that intelligence operations is highly variable and infl uenced by is an adaptation to new circumstances, Piaget the amount of cognitive stimulation. Furthermore, highlighted the transition between sensorimotor central to Piaget was not the age at which the substages III and IV. During sensorimotor substages emerge but the mechanisms involved in stage III (approximately 4–8 months), children stage transitions. Each stage is a temporary equi- use secondary circular reactions (i.e., actions librium in the process of equilibration (Piaget that focus on the effects they produce in their 1985 , p. 139). Higher stages are in a better equi- world) to interact with the world. Essentially, librium; they are characterized by a more com- secondary circular reactions aim at reproducing plex understanding of the world (greater the effect by repeating the action that generated spatiotemporal distance) and more mobile oper- this effect in the fi rst place. For example, Piaget’s ative structures. Because the stages build on each daughter Lucienne moved her legs vigorously, other, they constitute an invariant sequence. But thereby shaking her bassinet. The movement there is no fi xed, predetermined end point to the made the dolls swing that were hanging from the development of intelligence because a new stage hood. Lucienne looked at the dolls, smiled, and always opens up new possibilities of interacting repeated the movement (Piaget 1963 , Obs. 94). with the world, which may lead to further At sensorimotor substage VI (approximately development. We next briefl y describe the main 8–12 months), by contrast, infants construct characteristics of each stage (see Chapman hierarchical relations between secondary circu- 1988 ). lar reactions by subordinating one action as a 10 Piaget’s Theory of Intelligence 145 means (e.g., removing an obstacle) to another coordinated by a subject who begins to be aware action as an end (e.g., grasping the rattle). The of himself as the source of actions” (Piaget coordination of secondary schemes differs from 1972a , pp. 21–22). the behaviors displayed at the previous substage in two ways (Piaget 1963 , p. 229). First, whereas Preoperational Stage The emergence of the secondary circular reactions simply tried to symbolic or, as Piaget also termed it, semiotic reproduce an interesting event, the coordination function marks the onset of the preoperational of secondary circular reactions becomes neces- stage, which extends from about 2 to about 7 sary when infants in pursuit of their goals years (Piaget and Inhelder 1969 ). The semiotic encounter an obstacle that requires them to function underlies children’s abilities to engage accommodate existing schemes to a new situa- in a number of different activities, such as tion. Second, whereas secondary circular reac- deferred imitation (i.e., imitation in the absence tions lead to the differentiation between means of the model), pretend play, drawing, and ends only after the fact, means and ends are psychological functions based on mental images differentiated in substage IV from the outset. At (e.g., recall memory), and language. These substage IV, then, children coordinate two inde- activities are practiced and refi ned during the pendent schemes, the scheme assigning an end fi rst substage of this stage, the level of to the action (e.g., grasping the rattle) and the preconceptual thought (approximately 2–4 scheme used as a means (e.g., removing the years of age). At the same time, preoperational obstacle). Because the behavior at substage IV thought is characterized by profound cognitive requires an adaptation to a new situation, Piaget limitations. For example, although preconceptual ( 1963 , p. 228) terms means–end coordination a thought is no longer tied to particular objects or “true act of intelligence” and “the beginning of events (the here and now), it fails to distinguish intelligent action.” between individual members of a concept and The coordination and differentiation between the generality of a concept. To illustrate, when actions result in the construction of increasingly Piaget’s daughter Jacqueline was 31 months complex relations between objects—“the objec- old, she cried upon seeing a slug, “There it is!” tivization of reality” (Piaget and Inhelder 1976 , When she saw another slug a few yards further p. 32)—as refl ected in the development of such she said, “There’s the slug again” (Piaget 1962 ). basic categories as space, time, causality, and At this substage, concepts thus remain midway object (Piaget 1954 ). For example, in order to between the generality of the concept and the remove a cushion that is placed in front of an individuality of elements composing it. On the object, the child must realize for herself that the one hand, there is no concept of a general class; cushion, in fact, is placed in front of the object on the other hand, particular objects have less (space), that she must remove it before grasping individuality and easily lose their identity. the object (temporal series), that the object At the second substage of preoperational behind the cushion still exists (object perma- thought—termed intuitive thought—symbolic nence), and that in order to remove the cushion representational schemes become increasingly she must grasp it (spatialized and objectifi ed coordinated, and children become capable of causality). relating two such schemes to each other by means The sensorimotor period ends with the emer- of an unidirectional logical relation (Piaget gence of symbol representations, which allow 1970). For example, in comparing the liquid in infants to transcend the immediate here and now. two differently shaped containers, children may At the completion of the sensorimotor stage, for use height in order to infer the amount of liquid, the infant, his own action is no longer the whole but ignore the width of the container. Intuitive of reality and instead now becomes “one object thought thus remains centered on one dimension among others in a space containing them all; (e.g., height) and fails to establish bidirectional and actions are related together through being relations between dimensions (Piaget et al. 1977 ). 146 U. Müller et al.

Concrete Operational Stage During the (roses) = B (fl owers). The second step involves concrete-operational stage, which emerges performing the inverse (negative) operation asso- around 6–7 years, operations (i.e., internalized ciated with this combination of subclasses. This actions such as putting like objects together, entails subtracting each subclass from the super- putting objects in one-to-one correspondence) ordinate class such that A = B − A’ and become coordinated and integrated into logical A’ = B − A. The inverse operation, thus, implies systems (see Bibok et al. 2009). As a result, that children construct each subclass through children no longer center on one aspect of a negation under the including class. Piaget termed situation, and they can mentally reverse this type of negation partial because it is applied transformations that have occurred in reality. to a part of a larger whole. Through partial nega- The coordination of operations into systems tion children realize that the subclass A is an also leads to the emergence of logical necessity autonomous whole, which enables them to recog- (Piaget 1976a ). nize that there are some B’s that are not A’s (e.g., Piaget devised a variety of conservation tasks there are some fl owers that are not daisies) and to assess concrete-operational thought. that, therefore, there are more B’s than A’s. The Conservation refers to the understanding that a different operations required by class inclusion whole exists as a quantitative invariant and there- provide an example of an operational structure fore remains intact despite the quantitative rear- (Piaget 1976a ). rangement of its parts (Piaget and Inhelder 1974 ). For example, the number of objects in a set does Formal Operational Stage The last stage of not change by rearranging them (e.g., spreading cognitive development described by Piaget them out). To understand that the quantity has not emerges during adolescence (Inhelder and Piaget changed, children need to coordinate transforma- 1958; see Moshman 2009 ). Piaget and his tions in two dimensions (density of objects, collaborator Bärbel Inhelder studied formal length of row of objects). An operative under- operations by presenting children and adolescents standing of conservation is logical in nature, and with concrete material (e.g., different weights, it is not given by empirical observation of strings of different length) to be manipulated in transformations. order to discover scientifi c laws or the cause of a Another concept that children understand at result from several possible factors (e.g., which the concrete-operational level is class inclusion factor—weight, length of string, height of (Piaget and Inhelder 1969 ; Piaget 1980 ). A typi- dropping point, force of push—determines the cal class inclusion task requires children to com- frequency of the pendulum’s oscillation). These pare the number of objects in the including or studies revealed that children approached superordinate class with the number of objects in scientifi c problems in a qualitatively different the most numerous of two of its subclasses. For way than adolescents. Although children were example, given 12 daisies and 4 roses, children capable of classifying and cross-classifying the are asked, “Are there more daisies or more fl ow- independent variables, of properly ordering ers?” A correct answer requires that children con- magnitudes of the independent variable along serve the including class (B) while making the one dimension, and of putting these seriations quantitative comparison between it and the into correspondence with their effects on the included class (A). Although this may sound sim- dependent variable, they failed to separate the ple enough, such a comparison actually involves involved variables by varying only one variable a multistep process in which children must not and holding all others constant. As a result, these only be able to construct the including class but children did not supply adequate proof for their also be able to reverse this affi rmative operation statements. By contrast, from the outset, by properly decomposing it. The fi rst step adolescents formulated hypotheses and derived involves being able to combine two subclasses to conclusions from these hypotheses. They then form a superordinate class, or A (daisies) + A’ proceeded to test these hypotheses by 10 Piaget’s Theory of Intelligence 147

systematically controlling all variables except It also increases the speed of processing because the one under investigation. Thanks to their it makes it possible to imagine at the same time systematic experimental approach, adolescents the successive phases of an action. Finally, it excluded hypotheses that were contradicted by opens up the possibility of refl ecting on and observations and converged on the hypothesis understanding the reasons why some actions are that was actually true. successful and others not (Piaget 1954 ). For Piaget, the difference between children’s During the preoperational period, children use and adolescents’ approaches to these problems symbols in symbolic play (e.g., a toy cup stands suggested the reversal of the direction between for a real cup), deferred imitation (e.g., imitating reality and possibility: whereas in the concrete- an action of an absent model), and drawing (Piaget operational stage, possibility remains an exten- 1962 ; Piaget and Inhelder 1971 ). Piaget believed sion of reality, in the formal-operational stage, that particularly young children need to rely on the reality is subordinated to possibility. Adolescents use of individualized and personal systems of are capable of thinking hypothetico-deductively symbols (Piaget and Inhelder 1971 ). This is by drawing necessary conclusions from truths because personal symbols make fewer processing that are considered merely possible. demands than language which is based on collective and arbitrary signs. Piaget recognized that language is essential to socialization, which, in turn, Semiotic Function and Intelligence modifi es action and behavior. Verbal exchange between individuals allows children to share ideas, Piaget (1963 ; Piaget and Inhelder 1971 ) held that and the resulting “collective concepts” reinforce consciousness is always based on signs or, better, individual thinking (Piaget 1995 ). Being more signifi ers. Signifi ers are items that convey mean- mobile than symbols, language also makes a ing. At the sensorimotor level, signifi ers are not unique contribution to the mobility of thought. yet differentiated from their referent (signifi eds). At the same time, neither language nor sym- Signifi ers at this level are termed indications. An bols are the source of the forms of thought found indication is an “objective aspect of external real- at the concrete- and formal-operational stages. ity” (Piaget 1963 , p. 193), “a perceptible fact According to Piaget (1970 ), these forms of thought which announces the presence of an object or the are grounded in the practical coordination of imminence of an event (the door which opens and actions (e.g., grouping objects, seriating objects) at announces a person)” (Piaget 1963 , pp. 191– the sensorimotor stage. The semiotic function, and 192). Signifi eds at this level are sensorimotor particularly language, is necessary for the internal- schemes that confer meaning on the elements ization of actions (i.e., without the semiotic func- interacted with. tion, operations would have to be executed as At the end of the sensorimotor stage, the coor- successive actions and could not be condensed dination and differentiation of schemes culmi- into a simultaneous whole), but it is not suffi cient nate in the emergence of signifi ers that are to explain logical thought. In sum, Piaget consid- differentiated from their signifi eds. Piaget (Piaget ered the semiotic function only a tool used by and and Inhelder 1969 , 1971 ) termed a system of dependent on the operative aspect of intelligence such signifi ers the semiotic function. The semi- (Piaget and Inhelder 1969 , 1971 ). otic function subsumes both symbols and signs. Piaget defi ned symbols such as mental images as signifi ers (i.e., they resemble the things signifi ed) Affectivity and Intelligence and signs, such as words, as arbitrary and conventional signifi ers. The semiotic function makes There is a long tradition of treating intelligence it possible for children to form mental and emotion as distinct. Even in present-day representations and to think about absent objects psychology, IQ and EQ are thought of as separate as well as past, future, and even fi ctitious events. (or even opposing) constructs (Goleman 1990 ). 148 U. Müller et al.

In contrast to dualistic conception, Piaget ( 1981 ; development from the fi rst to the last day of life” see Sokol and Hammond 2009 ) believed that all (Piaget 1995 , p. 278). In his work, Piaget strug- behaviors involve an affective aspect and a cogni- gled with the fundamental epistemological ques- tive aspect. The affective aspect is responsible for tion: “Is it the individual as such or is it the social motivating the organism’s interaction with the group that constitutes the motor or, if you prefer, environment by assigning a value or goal to the the ‘context’ of intellectual evolution?” (Piaget behavior. However, achieving a particular end 1995 , p. 215). Piaget contrasted his own solution can involve a number of different paths. It is the to this question with theoretical positions which cognitive aspect of behavior that structures such suggest that rationality is derived either from the paths and thus the relation between the individual individual or the collective. By reducing the and the environment. In other words, affect pro- social to the aggregation of ready-made individ- vides the values and ends for actions, whereas ual consciousnesses, individualism provides an cognitive functions are the means for achieving atomistic explanation of the social and rational- the ends (see Binet and Simon 1916 p. 142). ity. Collectivism, on the other hand, considers the To illustrate that any intelligent act contains social as a whole that cannot be derived from an both affective and cognitive contributions, take additive composition of individuals. Rather, the the following sensorimotor action: a child reaches collective whole is characterized by emergent, for a toy by pulling on the blanket under the toy. novel properties and structures, and it modifi es This act has an affective component. In fact, two its members (i.e., individual persons; see Piaget types of affectivity are involved in this act: syn- 1995 ). chronic affectivity (in the moment) and dia- Piaget criticized both individualism and col- chronic affectivity (over time). First, the child lectivism and proposed an interactive relational evaluates his current actions with feelings of suc- position as an alternative explanation of the role cess and failure (synchronic affectivity), and sec- of social interaction in intellectual development. ond, the child’s evaluations of the situation According to the interactive relational position, involve a system of values that he has developed “there are neither individuals as such nor society over time, engaging his interest in obtaining the as such. There are just interindividual relations” toy (diachronic affectivity). These affective com- (Piaget 1995 , p. 210). These relations between ponents regulate the cognitive component of the individuals are primary and “constantly modify act that facilitated obtaining the object by pulling individual consciousnesses themselves” (Piaget on the blanket. Thus, affect provides direction for 1995 , p. 136). The interactive relational point of intelligence, fi rst by regulating interest and effort view leads to a more fi ne-grained analysis of spe- and, second, by assigning value to solutions cifi c social relationships and their implications sought. As such, for Piaget, affect and intelli- for development. Piaget describes two extreme gence are inseparable, and Piaget (1981 ) often types of social interaction: constraint and coop- underscored the role of affectivity in intellectual eration. Whereas constraint involves the imposi- growth. tion of authority and group traditions on the individual, cooperative interactions are based on reciprocity and equality. In his early work, Piaget Social Interaction and Intelligence (1932 , 1995) argued that cooperative relations among equals are necessary for the development Piaget is often accused of failing to address the of rationality and autonomous morality; by con- role of social interaction in development. This, trast, relations of unilateral respect in which one however, is not the case (see Kitchener 2009 ). individual has to submit to another individual’s Piaget was by no means oblivious to the role of authority impede the development of morality the social in development, as attested by, for and rationality. example, his statement that “human intelligence The upshot of the interactive relational view is is subject to the action of social life at all levels of that because individuals must coordinate their 10 Piaget’s Theory of Intelligence 149 actions vis-à-vis the world, “[i]ndividual address the biological dimension of intelligence. operations and cooperations form one insepara- Elegantly, Piaget’s insight into self-organization ble whole in such a way that the laws of the gen- grounded his conception of intelligence in the eral coordination of actions are, in their functional very feature of life itself. Second, intelligence is nucleus, common to inter- and intraindividual not passive, not a process triggered by an input actions and operations” (Piaget 1971 , p. 98). and, in turn, triggering some sort of output. Individual operations and cooperations are sub- Theories of intelligence must capture the fact that ject to the same kind of combinations and trans- human beings are active and transform through formations as actions and operations, thus the their actions the world. Intelligence is not just question of whether rationality is essentially comprised of a set of theoretical abilities; it social or individual becomes moot: includes, and is grounded in, practical skills. To wonder whether it is intrapersonal operations Third, intelligence has a normative dimension. that engender interpersonal co-operations or vice This normative dimension includes values, moral versa is analogous to wondering what came fi rst, norms, and logical necessity (Smith 1993 , 2009 ). the chicken or the egg … The internal operations The normative dimension, which may be unique of the individual and the interpersonal coordination of points of view constitute a single and the to humans, has been recalcitrant to reductionist same reality, at once intellectual and social. (Piaget explanations. We submit that a successful theory 1995 , pp. 294, 307) of intelligence will be based on the valuable insights provided by Piaget’s theory, that is, mod- For that reason, social interactions are subject ern intelligence theories must be grounded in to and regulated by the same equilibration pro- biology, acknowledge the active nature of intel- cesses as intraindividual actions and operations. ligence and the role of practical intelligence, and capture the normative dimension of knowledge.

Conclusion References In this chapter, we described, in broad strokes, Piaget’s theory of intelligence. We showed that Aristotle. (1956). De anima . Oxford: Clarendon. Piaget defi ned intelligence as the cognitive orga- Aristotle. (1970). Physics . Oxford: Clarendon. Aristotle. (1971). Metaphysics . Oxford: Clarendon. nization of an organism. More specifi cally, Piaget Bennour, M., & Vonéche, J. (2009). The historical context argued that intelligence can be defi ned function- of Piaget’s ideas. In U. Müller, J. I. M. Carpendale, & ally as resulting in increasing spatiotemporal dis- L. Smith (Eds.), The Cambridge companion to Piaget tances between subject and world and structurally (pp. 45–63). New York: Cambridge University Press. Berk, L. (2012). Infants and children (7th ed.). Toronto: in terms of a sequence of stages that move toward Pearson. increasing reversibility. We showed how Piaget’s Bibok, M. B., Müller, U., & Carpendale, J. I. M. (2009). conception of intelligence is rooted in his larger Childhood. In U. Müller, J. I. M. Carpendale, & epistemological framework, and we described L. Smith (Eds.), The Cambridge companion to Piaget (pp. 229–254). New York: Cambridge University Press. major features of this framework. Binet, A. (1894). Introduction à la psychologie expéri- Even though many aspects of Piaget’s theory mentale [Introduction to experimental psychology]. have been heavily criticized (Lourenço and Paris: Alcan. Machado 1996 ), we believe that the way in which Binet, A. (1975). Modern ideas about children . Menlo Park: San Francisco State University. (Original work he conceptualized the relations between affectiv- published 1909). ity, symbols, social interaction, and intelligence Binet, A., & Simon, T. (1916). The intelligence of the was original and provides a fruitful direction for feeble-minded . Baltimore: Williams & Wilkins overcoming sterile dichotomies. Furthermore, we Company. (Original work published in 1909). Boom, J. (2009). Piaget on equilibration. In U. Müller, think that three features of his theory are essential J. I. M. Carpendale, & L. Smith (Eds.), The Cambridge to any comprehensive and coherent theory of companion to Piaget (pp. 132–149). New York: intelligence. First, theories of intelligence must Cambridge University Press. 150 U. Müller et al.

Bringuier, J.-C. (1980). Conversations with Jean Piaget . Piaget, J. (1932). The moral judgment of the child . Chicago: The University of Chicago Press. (Original London: Kegan Paul Trench Trubner. work published in 1977). Piaget, J. (1950a). Introduction à l'èpistemologie gènè- Bühler, K. (1933). The mental development of the child . tique, Vol. 1: La pensèe mathèmatique [Introduction in New York: Harcourt, Brace & Company. (Original genetic epistemology: Mathematical thought]. Paris: work published 1918). Press Universitaires de France. Campbell, R. L. (2009). Constructive processes: Piaget, J. (1950b). Introduction à l'èpistemologie gènè- Abstraction, generalization, and dialectics. In tique, Vol. 2: La pensèe physique [Introduction in U. Müller, J. I. M. Carpendale, & L. Smith (Eds.), The genetic epistemology: Physical thought]. Paris: Press Cambridge companion to Piaget (pp. 150–170). Universitaires de France. New York: Cambridge University Press. Piaget, J. (1950c). Introduction à l'èpistemologie gènè- Chapman, M. (1988). Constructive evolution: Origins tique, Vol. 3: La pensèe biologique, la pensèe psy- and development of Piaget’s thought . New York: chologique et la pensèe sociologique [Introduction in Cambridge University Press. genetic epistemology: Biological, psychological, and Chapman, M. (1992). Equilibration and the dialects of sociological thought]. Paris: Press Universitaires de organization. In H. Beilin & P. Pufall (Eds.), Piaget’s France. theory: Prospects and possibilities (pp. 39–60). Piaget, J. (1954). The construction of reality in the child . Hillsdale: Lawrence Erlbaum. New York: Basic Books. (Original work published in Claparéde, E. (1917). La psychologie de l’intelligence. 1937). Scientia, 22 , 253–268. Piaget, J. (1962). Play, dreams and imitation in childhood . Goleman, D. (1990). Emotional intelligence . New York: New York: W. W. Norton & Company. (Original work Bantam Books. published in 1945). Hamlyn, W. (1971). Epistemology and conceptual devel- Piaget, J. (1963). The origins of intelligence . New York: opment. In T. Mischel (Ed.), Cognitive development Norton. (Original work published in 1936). and epistemology (pp. 3–24). New York: Academic. Piaget, J. (1967). Genesis and structure in the psychology Hübner, J. (1999). Die Aristotelische Konzeption der of intelligence. In J. Piaget (Ed.), Six psychological Seele als Aktivität in de Anima II 1 [Aristotle’s con- studies (pp. 143–158). New York: Random House. cept of soul as activity in de Anima II 1]. Archiv für (Original work published in 1964). Geschichte der Philosophie, 81 , 1–32. Piaget, J. (1968). Explanation in psychology and psycho- Inhelder, B., & Piaget, J. (1958). The growth of logical physiological parallelism. In P. Fraisse & J. Piaget thinking from childhood to adolescence . New York: (Eds.), Experimental psychology: Its scope and Basic Books. (Original work published in 1955). method (History and method, Vol. 1, pp. 153–191). Inhelder, B., & Piaget, J. (1969). The early growth of logic London: Routledge & Kegan Paul. (Original work in the child. New York: Norton. (Original work pub- published in 1963). lished 1959). Piaget, J. (1969). The mechanisms of perception . Kant, I. (1929). Critique of pure reason . London: New York: Basic Books. (Original work published in Macmillan. (Original work published in 1787). 1961). Kitchener, R. F. (1993). Piaget’s epistemic subject and Piaget, J. (1970). Piaget’s theory. In P. Mussen (Ed.), science education: Epistemological vs. psychological Carmichael’s manual of child psychology (3rd ed., issues. Science and Education, 2 , 137–148. pp. 703–732). New York: Wiley. Kitchener, R. F. (2009). On the concept(s) of the social in Piaget, J. (1971). Biology and knowledge . Chicago: The Piaget. In U. Müller, J. I. M. Carpendale, & L. Smith University of Chicago Press. (Original work published (Eds.), The Cambridge companion to Piaget (pp. 110– in 1967). 131). New York: Cambridge University Press. Piaget, J. (1972a). Psychology and epistemology: Towards Lourenço, O., & Machado, A. (1996). In defense of a theory of knowledge . Harmondsworth: Penguin Piaget’s theory: A reply to 10 common criticisms. Press. (Original work published in 1970). Psychological Review, 103 , 143–164. Piaget, J. (1972b). The principles of genetic epistemology . Moshman, D. (2009). Adolescence. In U. Müller, J. I. New York: Basic Books. (Original work published in M. Carpendale, & L. Smith (Eds.), The Cambridge 1970). companion to Piaget (pp. 255–269). New York: Piaget, J. (1973). Main trends in psychology . London: Cambridge University Press. George Allen & Unwin. (Original work published in Müller, U. (2009). Infancy. In U. Müller, J. I. 1970). M. Carpendale, & L. Smith (Eds.), The Cambridge Piaget, J. (1976a). The psychology of intelligence . companion to Piaget (pp. 200–228). New York: London: Routledge & Kegan Paul. (Original work Cambridge University Press. published in 1947). Natorp, P. (1910). Die logischen Grundlagen der exakten Piaget, J. (1976b). The grasp of consciousness . Cambridge, Wissenschaften [The logical foundation of the exact MA: Harvard University Press. (Original work pub- sciences]. Leipzig and Berlin: B. G. Teubner. lished in 1974). 10 Piaget’s Theory of Intelligence 151

Piaget, J. (1980). Experiments in contradiction . Chicago, Piaget, J., & Inhelder, B. (1976). The gaps in empiricism. IL: University of Chicago Press. (Original work pub- In B. Inhelder & H. Chipman (Eds.), Piaget and his lished in 1974). school (pp. 24–35). New York: Springer. (Original Piaget, J. (1981). Intelligence and affectivity . Palo Alto, CA: work published in 1969). Annual Reviews. (Original work published in 1954). Piaget, J., Grize, J.-B., Szeminska, A., & Bang, V. (1977). Piaget, J. (1985). Equilibration of cognitive structures . Epistemology and psychology of functions . Dordrecht: Chicago: University of Chicago Press. (Original work Reidel. (Original work published in 1968). published in 1975). Piaget, J., Henriques, G., & Ascher, E. (1992). Morphisms Piaget, J. (1987). Possibility and necessity, Vol. 1: The role and categories: Comparing and transforming . of possibility in cognitive development . Minneapolis: Hillsdale: Erlbaum. (Original work published in University of Minnesota Press. (Original work pub- 1990). lished in 1981). Smith, L. (1993). Necessary knowledge. Hove: Lawrence Piaget, J. (1995). Sociological studies. London: Routledge. Erlbaum. (Original work published in 1977). Smith, L. (2009). Piaget’s developmental epistemology. Piaget, J. (2001). Studies in reﬂ ecting abstraction . Hove: In U. Müller, J. I. M. Carpendale, & L. Smith (Eds.), Psychology Press. (Original work published in 1977). The Cambridge companion to Piaget (pp. 64–93). Piaget, J., & Inhelder, B. (1969). The psychology of the New York: Cambridge University Press. child . New York: Basic Books. (Original work pub- Sokol, B. W., & Hammond, S. I. (2009). Piaget and affec- lished in 1966). tivity. In U. Müller, J. I. M. Carpendale, & L. Smith Piaget, J., & Inhelder, B. (1971). Mental imagery in the (Eds.), The Cambridge companion to Piaget (pp. 309– child . New York: Basic Books. (Original work pub- 323). New York: Cambridge University Press. lished 1966). Taine, H. (1872). On intelligence . New York: Holt & Piaget, J., & Inhelder, B. (1974). The child’s construction Williams. of quantities . London: Routledge & Kegan Paul. Vonéche, J., & Vidal, F. (1985). Jean Piaget and the child (Original work published in 1941). psychologist. Synthese, 65 , 121–138. Alfred Binet and the Children of Paris 11

Amber Esping and Jonathan A. Plucker

Alfred Binet was born in 1857 to a wealthy but his father forced him to touch a cadaver.1 He troubled French family. His father, a physician, suffered an emotional collapse and dropped out and his mother, an amateur artist, divorced when of medical school at age 22 (Fancher 1985 ). he was a child, and he grew up in his mother’s Following his psychological breakdown, Binet household. Family resources afforded him an spent considerable time resting and reading among excellent private school education in Nice and the peaceful stacks of the National Library of France later Paris. He distinguished himself in French in Paris. While browsing books on psychology, he composition, but his academic record was other- discovered some ideas in which he could at last wise unremarkable (Siegler 1992 ; Wolf 1964 ). become genuinely and passionately invested. He fi rst Binet’s career path appeared desultory and became intrigued by psychophysical experiments unpromising at fi rst. In 1878, he earned a law involving tests of two-point sensation thresholds, and degree from the University of Paris, but never he replicated some of the published experiments expressed any real interest in the fi eld. He would using himself and some friends as subjects. He con- later call law “a career for those without any [yet] cluded from his own results that extant theories about chosen vocation” (Binet 1904a , p. 14). Next he sensation thresholds should be modifi ed, and he pub- attempted a medical degree at the Sorbonne in lished a paper outlining his suggested corrections Paris, but he was profoundly distressed by the (Binet 1880 ). The article was well written and trauma and gore he witnessed in the operating cogently argued, but unfortunately it was also fueled room, to which he may have been especially sen- by naïve enthusiasm. The ideas Binet put forth as his sitive owing to a childhood experience in which own had in fact already been published—in much more sophisticated form—by a respected Belgian physiologist by the name of Joseph Delboeuf (1831– 1896). Delboeuf responded by publishing a humiliat- ing critique of Binet’s article (Delboeuf 1880 ; see also Fancher 1985 ; Wolf 1964 ). A. Esping Undeterred, Binet continued to read about College of Education , Texas Christian University , TCU Box 297900 , Fort Worth , TX 76021 , USA psychology and to publish articles independently, J. A. Plucker (*) Department of Educational Psychology , 1 Binet told the story of touching the cadaver in 1911, indi- University of Connecticut, Neag School of Education , cating that this distressing experience had happened to 2131 Hillside Road, Unit 3007 , Storrs , one of his “friends.” Compelling circumstantial evidence CT 06269-3007 , USA suggests that he was really talking about himself (see Wolf e-mail: [email protected] 1964 , pp. 762–763).

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 153 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_11, © Springer Science+Business Media New York 2015 154 A. Esping and J.A. Plucker leading one biographer to dub him a “library psy- as “The Queen of Hysterics.” Ms. Witman could chologist” (Siegler 1992 , p.180). He was most be relied upon to follow the three-stage hypnotic interested in English associationists like John pattern of lethargy, catalepsy, and somnambulism Stuart Mill (1806–1873), Herbert Spencer that Charcot believed was defi nitive of major (1820–1903), and Alexander Bain (1818–1903), hypnotism. Charcot and Binet saw this easy sus- but he was also infl uenced by Hippolyte Taine’s ceptibility to hypnotic states and consistent pat- (1828–1893) philosophical treatise On tern of responses as an indication of her Intelligence (1870/1872) and Théodule-Armand underlying hysterical disorder. Indeed, Charcot Ribot’s (1839–1916) treatments of English and believed generally that susceptibility to hypnosis German psychology (Foschi and Cicciola 2006 ). was an indication of an underlying hysteria. This Mill was Binet’s particular favorite, and he would conclusion would later turn out to be incorrect later refer to him as “my only master in psychol- (Fancher 1985 ). ogy” (Binet 1903 , p. 68). Binet’s second publica- In a series of related experiments with Ms. tion forwarded Mills’ associationism as an Witman, Binet and another of Charcot’s assis- all-encompassing explanation for the operations tants, Dr. Charles Féré (1852–1907), discovered of the intellect (Binet 1883 ), but here again he that they could reverse or transfer Ms. Witman’s was overconfi dent. Associationism was already behaviors under hypnosis simply by reversing the beginning to lose its cachet, and prominent psy- polarity of a large horseshoe magnet. For exam- chologists were routinely acknowledging its ple, if Ms. Witman seemed to be paralyzed on her inability to account for motivational or uncon- right side, they could transfer the paralysis to the scious infl uences (Fancher 1985 ). Binet eventu- left by reversing the magnet. They found that ally realized these defi ciencies of associationism, they could transform the expression of an emo- but he never abandoned it completely. Indeed, tion, such as sobbing, to its inverse (laughing) his associationist roots would later be evident in through the same mechanism. Binet and Féré his greatest achievement, the Binet-Simon were fully convinced by Charcot that deeply hyp- Intelligence Scale (1905 ) . notized people were not aware of their surround- In 1883 Binet’s independent means made it ings, so they did not consider the possibility that possible for him to volunteer his time assisting the their patient might be attending to the magnet or eminent neurologist Jean-Martin Charcot (1825– related experimenter cues. In a series of articles 1893) with his research at the Salpêtrière Hospital (Binet and Féré 1885a , b , c ), they attributed the in Paris. Charcot was interested in hysteria, a baf- magnet fi ndings to the existence of complemen- fl ing syndrome in which female patients experi- tary human emotions, akin to complementary enced paralysis, loss of sensation, seizures, and colors which produce white or black when mixed memory loss, with no apparent neurological (Fancher 1985 ). cause. Whereas some physicians attributed these Unfortunately for Binet, his fi rst critic, Joseph cases to malingering, Charcot believed that the Delboeuf, also had a side interest in hypnosis. patients experienced these symptoms as real Delboeuf respected Charcot and was ready to (Fancher 1985 ). Charcot’s associated interest in accept his theory of major hypnosis, but he was hypnosis grew from the fi nding that some of the skeptical of the magnet fi ndings—especially so same “symptoms”—paralysis, amnesia, dramatic when he saw Binet was one of the paper authors fi ts, and so on—could be induced through hyp- (Fancher 1985 ). Curious, Delboeuf visited the notic suggestion. Therefore, the study of hypnosis Salpêtrière hospital and saw immediately that Ms. offered promise in understanding the underlying Witman was aware of the magnet and was undoubt- causes of hysteria (Fancher 1985 ). edly yielding—consciously or unconsciously—to Binet assisted Charcot in hypnosis studies of the desires of Féré with whom she seemed to have hysterical patients. One particular responsibility a particular rapport. When Delboeuf undertook the was Blanche Witman, an intense and melodra- same experiments under more carefully controlled matic young woman who was known in the wards conditions, he discovered that both the magnet 11 Alfred Binet and the Children of Paris 155

fi ndings and Charcot’s theory of major hypnosis His time at the Salpêtrière also allowed Binet’s were invalid (see Delboeuf 1886). Binet was at passive associationist psychology to mature into fi rst reluctant to accept Delboeuf’s evidence, and a a sophisticated theory that recognized the active heated public exchange ensued (Binet and role of human attention, as well as the impor- Delboeuf 1886 ). However, Binet eventually tance of innate and hereditary factors in deter- acknowledged the “loopholes for error” which had mining one’s makeup (Fancher 1985 ). “pervert[ed]” interpretation of his study results (Binet 1896 /1977, p. 76). This proved to be an essential lesson for Binet, and he was ever after- Binet and Experimental Child wards aware that psychological tests will always Psychology contain some degree of error. His willingness to acknowledge and work within the constraints of Binet’s curiosity was unbounded, and he pro- this fact has been called by one scholar “[perhaps] duced many other signifi cant acts of scholarship Binet’s greatest contribution” to intelligence test- while working with Charcot at the Salpêtrière. ing (Kaufman 2009 , p. 22). His publications during this time included three Following this major career setback, Binet left books and more than 20 articles exploring a wide the Salpêtrière, and it took more than a year to variety of subjects. Among these were the psy- fi nd another position. Fancher (1985 , p. 57) notes chic life of microorganisms (Binet 1887a ), sexual that under the circumstances, it is not surprising fetishes (Binet 1887b ), and the nature of human that “prospective employers did not come fl ock- consciousness (Binet 1890a ). He also developed ing to his door.” However, he eventually con- an interest in the natural sciences, eventually nected with the director of the new Laboratory of earning a Ph.D. for a dissertation on the anatomy Physiological Psychology at the Sorbonne, and and physiology of the subintestinal nervous sys- he willingly accepted Binet’s offer to work there, tems of insects (Binet 1894 ). The birth of his as he had done at the Salpêtrière, without com- daughters, Madeleine (b. 1885) and Alice (b. pensation. He served there as a researcher and 1887), provided an avenue to study child psy- assistant director and ascended to become the chology, and in 1890 he published three articles laboratory’s director in 1894. He held this unpaid describing experiments he conducted using his position until he died in 1911 (Fancher 1985 ; girls and their friends as subjects (Binet 1890b , c , Siegler 1992 .) d). This emerging interest in the psychology of Binet’s early career may be characterized as a children evolved into Binet’s new career as an series of productive false starts. He made mis- experimental child psychologist (Fancher 1985 ). takes, to be sure, but in the process he gained Binet derived his fi rst experiments with skills and dispositions that would prove enor- Madeleine and Alice from Galton and Cattell’s mously benefi cial in his future intelligence work. psychophysical tests of reaction times and sen- From the hypnosis debacle, he learned the impor- sory acuity, which up to this point in history rep- tance of careful attention to experimental con- resented the state of the art in intellectual testing. trols. But he also learned to appreciate the He noted that on average, his young subjects advantages of the detailed case study approach to reacted to stimuli much more slowly than did research. This distinguished him from contempo- adults but also that they were far less consistent raries like Francis Galton (1822–1911) and James in their performances; a child’s reaction time McKeen Cattell (1860–1944), who favored gen- might be on par with the typical adult in one trial eralizations based on large sample sizes. This and substantially slower in the next. Binet appreciation of the uniqueness of individuals deduced that the salient difference between adults convinced Binet that measuring psychological and children, then, was not really the reaction variables was a far more complex and nuanced times but rather the children’s limited ability to process than other intelligence researchers had so sustain attention during the trials. This insight far been willing to acknowledge (Fancher 1985 ). about the importance of attention proved to be 156 A. Esping and J.A. Plucker fundamental to the eventual development of his children’s memories and their ability to resist the intelligence scale (Fancher 1985 ). infl uence of experimenter suggestion improved Binet’s psychophysical tests of color percep- with age (Binet 1900 ). He also initiated several tion also yielded interesting results. Child sub- in-depth case studies of people with extraordi- jects were much slower than adults in naming nary abilities and accomplishments, such as colors, and this outcome might have been used to chess prodigies and mathematical wizards (Binet support the hypothesis that children had less and Henneguy 1894 ) and eminent French authors developed sensory acuity than adults. However, (Binet and Passy 1895 ). From these, he deter- when Binet asked the child subjects to match col- mined some unanticipated facts about the human ors, they were nearly as fast as the typical adult. intellect. First, there are many ways of becoming From this Binet concluded that seeming adult- extraordinary; the great writers and math and child differences in color perception were in chess prodigies approached their cognitive tasks actuality methodological artifacts resulting from in a variety of ways. Second, for the most part, differences in language development—the kids these extraordinary individuals were quite ordi- could see the differences; they just could not say nary in domains other than one particular narrow them fast enough. Binet eventually lost faith in area of excellence. Binet recognized these fi nd- psychophysical testing as a reliable and valid ings as important evidence of the complexity and measure of intellectual ability and determined heterogeneity of intellectual operations (Fancher that more complex, language-based tasks were 1985 ). The psychophysical testing that had domi- needed to discriminate child from adult intellec- nated the fi eld to this point would never be able to tual capacity (Fancher 1985 ). tease out these kinds of nuances. New methods Binet advanced his understanding of the for testing individual differences in intellectual importance of language development by asking functioning had to be developed. his children and their friends to defi ne common words. He discovered that children typically responded by providing concrete, functional Binet and Individual Psychology examples of how the items were used rather than the abstract dictionary-type defi nitions most However valuable the in-depth case analyses adults provided. For example, a knife was defi ned Binet cut his teeth on, he also recognized that as “to cut meat.” The defi nition of snail was these long investigations were not always practi- “squash it.” From this adult-child difference, cal. Psychologists needed to be able to compare Binet concluded that the ability to think in intellectual functioning quickly along some stan- abstract terms must somehow be important to the dard dimensions, preferably in one sitting. His development of human intelligence (Binet 1890b ; prior research had illuminated the vulnerabilities see also Fancher 1985). He continued to study his of psychophysical testing, so the relatively fast children, retaining some of the Galtonian psycho- methods he sought would have to test higher- physical tasks and also including tests of mem- order cognitive processes. Binet and his research ory, judgment, imagination, and inkblot assistant, Victor Henri (1872–1940), identifi ed 10 interpretation, as well as qualitative impressions candidate variables for measurement: (1) mem- about their temperaments and personalities. He ory, (2) imagery, (3) imagination, (4) attention, published these results in book form in 1903 (5) comprehension, (6) suggestibility, (7) a (Binet 1903 ). esthetic sentiment, (8) moral sentiment, (9) mus- Other aspects of human intellectual develop- cular strength and willpower, and (10) motor abil- ment also caught Binet’s attention during these ity and hand-eye coordination. The last two early years at the Sorbonne. He expanded his variables resonated with earlier psychophysical subject pool to include children in the local testing approaches, but as conceived they were schools, undertaking studies of memory and sug- more complex than standard tasks of that kind. gestibility. He discovered that both accuracy of The other eight variables were refreshingly origi- 11 Alfred Binet and the Children of Paris 157 nal in fl avor (Fancher 1985 ). Binet named this government asked Binet to join a distinguished new approach “Individual Psychology” (Binet & commission of experts who could provide insight Henri, 1986 ). and leadership regarding the education of these In 1899, a young medical student named special cases. Binet’s Individual Psychology Théodore Simon (1873–1961) contacted Binet research, his publication record, and his particu- and requested an opportunity to work with him. lar experience with Simon’s institutionalized Binet did not need another assistant, and he was children uniquely qualifi ed him for this undertak- inclined to refuse the offer. However, Simon had ing. He immediately recognized the need for a recently obtained a medical internship working diagnostic system that could identify those chil- with approximately 300 abnormal children2 at the dren who could benefi t from special education Perray-Vaucluse asylum, near Paris, and Binet classes and, just as important, prevent intellectu- found the opportunity to apply Individual ally normal children from being misdiagnosed Psychology with this special population very (Binet and Simon 1905a ). One year later, he had attractive. He accepted the offer of help and one: The Binet-Simon Scale, the world’s fi rst trained Simon to use his testing techniques. modern intelligence test (Binet and Simon Simon returned to Perray-Vaucluse and spent 1905b ). In a 1909 book, Binet described the the next several months engaged in psychological enthusiasm with which he approached this work: testing. These data would later become his There is nothing like necessity to generate new doctoral thesis in medicine (Wolf 1961 ). ideas. We undoubtedly would have retained the Unfortunately, the results of Binet’s Individual status quo…if a matter of true social interest three Psychology research program were largely disap- years ago, had not made it mandatory for us to measure intelligence by the psychological method. pointing. In a 1904 paper, Binet reported that It had been decided to try to organize some special they had failed to produce a valid and discrimi- classes for abnormal children. Before these chil- nating psychological test that could be adminis- dren could be educated, they had to be selected. tered in a short period of time. The in-depth case How could this be done?…It was under these circumstances that our devoted collaborator, Dr. study, it seemed to him, was still the most prom- Simon, and I formulated a plan for measuring ising approach to individual psychology (Binet intelligence. (Binet 1909 /1973, pp. 104–105) 1904b; see also Fancher 1985 ). However, in short order Binet would be offered a challenge that The defi nition of “intelligence” is a diffi cult would change his mind. thing to pin down even in the twenty-fi rst century (Plucker and Esping 2014 ), and Binet and Simon were working from scratch. They began by look- Binet Invents the Intelligence Test ing for evidence of what might now be termed “face validity”—that is, by recruiting groups of By the early part of the twentieth century, French children who had previously been identifi ed by national laws had begun mandating public school experts as being obviously intellectually normal education for all children, including children or clearly subnormal in their intellectual func- with mental disabilities, who had previously been tioning. Drawing on their earlier work in excluded entirely or permitted to drop out early Individual Psychology, Binet and Simon admin- from schooling. In 1904, offi cials of the French istered a variety of tests to both groups, with the expectation that some of these tests might plainly differentiate normal from subnormal children. In 2 The language used to describe intellectual and developmental disability in the late nineteenth and early twentieth choosing their tasks, the researchers were partic- centuries included the (now offensive) terms abnormal ularly careful to avoid items that might rely heav- and feebleminded and the clinical labels moron, imbecile, ily on formal education, as they wanted their tests and idiot. Binet preferred the term débiles (“weak ones”). to show evidence of psychological functioning, The person-fi rst language considered respectful in the twenty-fi rst century (e.g., “persons with intellectual dis- not educational attainment. This remains an abilities”) was unheard of. essential goal of intelligence testing to the 158 A. Esping and J.A. Plucker

present day (Kaufman 2009 ). As one means of and items calibrated such that they could be accessing higher-order processes, they chose to “located” at ages where typical children started include some questions about typical life within to complete them successfully (Binet 1911 ; Binet the French cultural context. They believed that it and Simon 1908 ). For example, a 10-year-old was safe to assume that even poor children of child who completed all the tasks usually passed normal intelligence would have reasonable famil- by 10-year-olds—but nothing beyond—would iarity with this kind of information (Fancher have a mental level that exactly matched his or 1985 ). her chronological age, 10.0. Children who Binet and Simon’s fi rst attempts at differenti- attained a mental level 2 or more years behind ating intellectually normal from subnormal chil- their chronological age—e.g., a 10-year-old child dren were unsuccessful. There were able to fi nd with a mental level of 8—were generally diag- important differences in average performance on nosed as being mentally subnormal, providing the tasks, but they failed to fi nd any set of items that they were otherwise healthy and motivated that only the normal children could solve. There when they took the test.3 This diagnosis was was always overlap, with some normal children applied to approximately 7 % of the students who failing tests that some subnormal children passed. were tested (Fancher 1985 ). The “aha!” moment came when the researchers The creation of the Binet-Simon Scale marked recognized one essential difference between the the development of a completely revolutionary two groups: the normal children were able to approach to the measurement of human intellec- respond to the tasks correctly at an earlier age tual functioning. Rather than relying on simple than the other group. It was critical to take age measures of reaction time and sensory acuity, into consideration when scoring (Fancher 1985 ). Binet and Simon’s test purported to measure Armed with this insight, the researchers cre- higher-order processes such as memory, lan- ated a series of tasks of increasing complexity. guage, and attention. In particular, however, the Some of the simplest test items assessed whether researchers believed that their scale measured or not a child could follow a lighted match with the subjects’ capacity to exercise judgment. his eyes, take a candy out of a wrapper, or shake Although conventional academic wisdom hands with the examiner. Slightly harder tasks purports that Binet and Simon did not have a required children to point to various named body clear defi nition of intelligence guiding their parts, repeat back a series of 3 digits, repeat from work, they were rather clear about their concep- memory a 15-word sentence, and defi ne words tualization of the construct: like house, fork , and mama . More diffi cult test [I]n intelligence there is a fundamental faculty, the items required children to state the difference alteration or the lack of which, is of the utmost between pairs of things, reproduce drawings importance for practical life. This faculty is judg- from memory, and construct sentences from three ment, otherwise called good sense, practical sense, initiative, the faculty of adapting one’s self to cir- given words such as Paris , gutter , and fortune . cumstances. A person may be a moron or an imbe- Some of the hardest items asked children to cile if he is lacking in judgment; but with good repeat back seven random digits, fi nd three judgment he can never be either. Indeed the rest of rhymes for the French word obéissance , state the the intellectual faculties seem of little importance in comparison with judgment. (Binet and Simon difference between abstract concepts like sad and 1916 /1973, pp. 42–43) bored, and answer questions such as, “My neigh- bor has been receiving strange visitors. He has received in turn a doctor, a lawyer, and then a priest. What is taking place?” (Fancher 1985 ; Kaufman 2009 ). 3 Binet and Simon were keenly aware that physical prob- The scale was revised in 1908 and 1911. The lems could mimic psychological ones. Their experience at the laboratory school they set up revealed that 5 % of stu- newer versions were developed with larger sam- dents experienced academic problems merely because ple sizes, greater age, and socioeconomic ranges, they could not see the blackboard (Binet 1907 ). 11 Alfred Binet and the Children of Paris 159

Mental Orthopedics The Binet-Simon Scale Comes to the United States In 1905, Binet submitted a report in which he outlined his recommendations for special educa- The ultimate popularity of the Binet-Simon tion pedagogy. He was optimistic about opportu- Scale owes a large debt to the actions of the nities to help subnormal children improve their American psychologist Henry Herbert Goddard intelligence, and he strongly disavowed the popu- (1866–1957). One year after Binet and Simon lar notion that intelligence should be viewed as a published the fi rst version of their intelligence fi xed and immutable quality. He later stated the test, Goddard accepted a position as Director of case this way: Research at the Training at a school for feeble- I have often realized, with great sorrow, the exis- minded children in Vineland, New Jersey. The tence of frequent prejudice against the educability United States did not possess a uniform system of intelligence. The well-known proverb that says: for defi ning, diagnosing, and classifying intel- “When one is stupid, it is for long” seems to be lectual disability, and most educators and physi- taken literally by some teachers … they don’t care about less-intelligent pupils, they don’t nourish cians depended on a highly subjective and any liking or respect towards them … Intelligence unreliable “we know it when we see it” approach. is not a unique, indivisible function, a particular Goddard was fairly confi dent in his own judg- essence, but it’s made up of the cooperation among ment in these matters, and he was convinced that all the minimal functions of discrimination, observation, retention, etc., whose plasticity and exten- most people who worked closely with disabled sibility have been verifi ed … As a consequence, persons could also be relied on to make “rather intelligence is susceptible t development; through accurate” intuitive judgments (Goddard 1908b , exercise, training, and, above all, method, one will p. 12). However, as a scientist, he would have be able to increase one’s attention, memory, judgment, and to become literally more intelligent than preferred an objective method, had one been before. (Binet 1909 /1973, pp. 100–102) available. But the major steps recently taken in France had not yet made their way across the Binet developed a series of cognitive exercises Atlantic (Zenderland 1998 ). he called “mental orthopedics,” which he believed For the next 2 years, Goddard experimented could raise children’s intelligence. A particular with several unsuccessful approaches to mental focus of these exercises was improving the sub- testing. In 1908, he took an extended trip to jects’ capacity to pay attention, since this seemed Europe to seek counsel with experts there. On to be fundamentally lacking in many children of one of these visits, he met a Belgian physician low intelligence. For instance, he advocated for and special educator named Ovide Decroly, who the use of fun games like “statue” in which chil- shared a copy of the Binet-Simon Scale. dren had to freeze until they were permitted to Intrigued, Goddard brought the test back to the move (Binet 1909 /1973; see also Fancher 1985 ). United States and tried the tasks with the stu- In 1907, he set up three experimental special edu- dents at the Vineland school. He discovered that cation classes where mental orthopedics could be the mental levels of the children generally cor- practiced. (The law mandating special education responded to the intuitive judgments made by would not go into effect for 2 more years.) It is himself and the other members of the Vineland notable, however, that he also advocated for edu- staff, thus providing evidence of criterion valid- cating intellectually normal and subnormal chil- ity (Goddard 1908a ). Soon thereafter, the dren together. He believed that this practice American Association for the Study of the would provide positive models for the slower- Feeble-Minded tentatively adopted Goddard’s learning children and healthy opportunities for classifi cation system as “the most reliable the faster-learning children to exercise virtues of method at present in use for determining the duty and solidarity (Binet 1909 /1973 Binet and mental status of feebleminded children” Simon 1908 ; see also Foschi and Cicciola 2006 ). (Rogers 1910 ). With this adoption, Binet’s 160 A. Esping and J.A. Plucker approach to intelligence testing became fi rmly entrenched in American society (Zenderland Binet’s Infl uence on Future 1998 ). Over the next few years, Goddard dis- Intelligence Tests tributed 22,000 copies of his English translation of the Binet-Simon test (Fancher 1985 ). It is an Binet contracted an illness and died in 1911 at the irony of history that the Binet test did not age of 54. His premature death cut short a prodi- become popular in France until the mid- 1900s, gious career in its prime. Even so, the legacy he when a French social worker who had spent left is staggering in its infl uence. Aside from the time in the United States brought a US version unparalleled accomplishment of the 1905 test and of the test back to France (Kaufman 2009 ; its subsequent revisions, Siegler ( 1992 ) notes the Siegler 1992 ). importance of Binet’s willingness to discuss frankly the virtues and limitations of his scale, and his progressive ideas about the malleable nature of The Binet Tests and US Immigration intelligence. These remain hot topics in the present Restriction day. His careful attention to empirical evidence— learned the hard way through embarrassing expe- Between 1890 and 1910, approximately 12 mil- riences in Charcot’s laboratory—distinguished lion immigrants attempted to enter the United him from contemporaries, like Goddard, who were States through the Ellis Island Checkpoint. more comfortable trusting subjective expert judg- Immigration critics warned that this generation ment. The Binet-Simon Scale has been translated was “less educated, more impoverished, and into dozens of languages and revised and adapted more culturally ‘alien’ than earlier groups of countless times by intellectual heirs who appreci- immigrants” (Zenderland 1998 , p. 263). To allay ated the originality and utility of the tasks. Even fears, Congress passed an 1882 law prohibiting though more recent approaches to intelligence “idiots” and “lunatics” from passing through the theory and testing vary considerably in their theo- gates. The law expanded in 1907 to include retical orientations and in their approaches to test- “imbeciles,” feebleminded persons, and persons ing, many of the items on Binet’s original scale with physical defects that might prevent them have stood the test of time (see, e.g., the enduring from sustaining themselves through respectable popularity of the Stanford-Binet assessments) and employment (Zenderland 1998 ). would seem familiar to twenty-fi rst-century test Goddard and his team were invited to Ellis takers and psychometricians. Island to help enforce these regulations; the Binet-Simon Scale proved central to his task. The procedure Goddard developed in 1912 was a References two-step process: one assistant would visually screen for suspected mental defectives as the Binet, A. (1880). De la fusion des sensations semblables. immigrants passed by (using the intuitive judg- Revue Philosphique, 10 , 284–294. Binet, A. (1883). Le Raisonnement dans les perceptions. ment purportedly developed through close con- Revue Philosophique, 15 , 406–432. tact over many years). Those who appeared Binet, A. (1887a). La vie psychique des micro-organismes suspect would then proceed to another location [The psychic life of microorganisms]. Revue where the other assistant would test them with a Philosophique, 24 (449–489), 582–611. Binet, A. (1887b). Le fetichisme dans l’amour. Revue variety of performance measures and a revised Philosophique, 24 , 142–167. version of the Binet Scale . The number of immi- Binet, A. (1890a). On the double consciousness. grants who were deported increased exponen- Experimental psychological studies . Chicago: The tially as a result of these screening measures Open Court. Binet, A. (1890b). Children’s perception. Revue (Zenderland 1998 ). Philosophique, 30 , 582–611. 11 Alfred Binet and the Children of Paris 161

Binet, A. (1890c). Studies on movements in some young Binet, A., & Passy, J. (1895). Nots psychologiques sur les children. Revue Philosophique, 29 , 297–309. auteurs dramatiques. L’Année Psychologique, 1 , Binet, A. (1890d). The perception of lengths and numbers 60–118. in some small children. Revue Philosophique, 30 , Binet, A., & Simon, T. (1905a). Sur la necessité 68–81. d’établir un diagnostic scientifi que des états inférieurs Binet, A. (1894). Contribution à l’étude du système de l’intelligence. L’Année Psychologique, 11 , nerveux sous-intestinal ds insects. [A contribution to 161–190. the study of the subintestinal nervous system of Binet, A., & Simon, T. (1905b). Méthodes nouvelles pour insects.]. Paris: Alcan. le diagnostic du niveau intellectual des anormaux. Binet, A. (1896/1977). Alterations of personality. In D. W. L’Année Psychologique, 11 , 191–244. Robinson (Ed.), Signifi cant contributions to the his- Binet, A., & Simon, T. (1908). The development of tory of psychology (Vol. V). Washington, DC: intelligence in the child. L’Année Psychologique, 14 , University Publications of America. 1–90. Binet, A. (1900). La suggestibilité . Paris: Schleicher. Binet. A., & Simon, T. (1916/1973). The development of Binet, A. (1903). L’Etude experimentale de l’intelligence . intelligence in childre n. Baltimore/New York: Paris: Schleicher. Williams & Wilkins/Arno Press. Binet, A. (1904a). La création littéraire: Portrait Delboeuf, J. L. R. (1880). Note in Revue Philosophique, psychologique de M. Paul Hervieu. L’Année 10, 644–648 Psychologique, 10 , 1–62. Delboeuf, J. L. R. (1886). Infl uence de l’education et de Binet, A. (1904b). Sommaire des travaux en cours a la l’imitation dans le somnambulisme provoqué. Revue societe de psychologie de l’enfant [Summary of the Philosophique, 22 , 146–171. work in progress at the society of child psychology]. Fancher, R. E. (1985). The intelligence men: Makers of L’Année Psychologique, 10 , 116–130. the IQ controversy . New York: W. W. Norton & Binet, A. (1905). New methods for the diagnosis of the Company. intellectual level of subnormals. L’Année Foschi, R., & Cicciola, E. (2006). Politics and naturalism Psychologique, 12 , 191–244. in the 20th century psychology of Alfred Binet. Binet, A. (1907). La valeur medicale de l’examen e la History of Psychology, 9 (4), 267–289. vision par les instituteurs. Bulletin de la Societe libre Goddard, H. H. (1908a). The Binet and Simon tests of pour l’etude psychologique de l’enfant, 40 , intellectual capacity. Training School Bulletin, 5 , 3–9. 146–163. Goddard, H. H. (1908b). The grading of backward chil- Binet, A. (1909/1973). Modern ideas about children dren. Training School Bulletin, 5 , 12–14. (trans: Heisler S.). Menlo Park, CA: Suzanne Kaufman, A. S. (2009). IQ testing 101. New York: Springer. Heisler. Plucker, J. A., & Esping, A. (2014). Intelligence 101 . Binet, A. (1911). Nouvelles recherches sur la mésure du New York: Springer. niveau intellectuel chez des enfants d’école. L’Année Rogers, A. C. (1910). Editorial: The new classifi cation Psychologique, 17 , 145–201. (tentative) of the Feeble-Minded. Journal of Psycho- Binet, A., & Delboeuf, J. L. R. (1886). Les diverses écoles Asthenics, 15 , 70. hypnotiques. Revue Philosophique, 22 , 532–538. Siegler, R. S. (1992). The other Alfred Binet. Binet, A., & Féré, C. (1885a). L’hypnotisme chez les hys- Developmental Psychology, 28 (2), 179–190. tériques: Le transfert. Revue Philosophique, 19 , 1–25. Taine, H. (1870/1872). On intelligence. New York: Holt & Binet, A., & Féré, C. (1885b). Hypotisme et responsabil- Williams. ité. Revue Philosophique, 19 , 265–279. Wolf, T. H. (1961). An individual who made a difference. Binet, A., & Féré, C. (1885c). La polarization physique. American Psychologist, 16 (5), 245–248. Revue Philosophique, 19 , 369–402. Wolf, T. H. (1964). Alfred Binet: A time of crisis. Binet, A., & Henneguy, L. (1894). La psychologie des American Psychologist, 19 (9), 762–771. grands calculateurs et jouers d’écecs . Paris: Hachette. Zenderland, L. (1998). Measuring minds: Henry Herbert Binet, A., & Henri, V. (1896). La psychologie individu- Goddard and the origins of American intelligence test- elle. L’Année Psychologique, 2 , 411–465. ing . Cambridge: Cambridge University Press. From Psychometric Testing to Clinical Assessment: 12 Personalities, Ideas, and Events That Shaped David Wechsler’s Views of Intelligence and Its Assessment

Mark Benisz , Ron Dumont , and John O. Willis

To many students of psychology, the name David on many different societies and cultural groups Wechsler is synonymous with cognitive testing. (Georgas et al. 2003 ). Despite the fact that many Although Dr. Wechsler died over 30 years ago, different reliable and valid cognitive assessment new editions and revisions of the tests derived tools are available for sale, and even though sig- from his original Wechsler-Bellevue Intelligence nifi cant advances have been achieved in the fi eld Scale (W-B; Wechsler 1939 ) continue to list him of neuropsychology, the Wechsler products con- as the sole author (e.g., Wechsler 2008 , 2012 , tinue to retain their popularity (Kaufman 2009 ; 2014 ) and new, but related, tests list him as senior Boake 2002 ). author (e.g., Wechsler et al. 2004 ; Wechsler and Why do psychologists continue to prefer Naglieri 2006 ). Wechsler’s contributions to the Wechsler scales over other tests? Certainly, the fi eld of psychometrics and intelligence testing go tests have excellent and evolving psychometric well beyond the assessment tool he fi rst designed qualities, a wealth of research and commentary, over 70 years ago. His test products are not only and a valuable history of interpretation. Ironically, among the most widely used and most widely though, one reason for the resistance to changing studied measures of cognitive ability (Wahlstrom from such heavy reliance on the Wechsler scales et al. 2012 ), but they have also been translated is likely because of the changes Wechsler initi- into over a dozen languages and standardized ated in the way contemporary psychologists think about and use tests. He was a strong advocate of elevating the tests from a purely psychometric approach to a more clinical approach (Kaufman * M. Benisz , Psy.D. ( ) 2009 ). To understand how Wechsler achieved this Kiryas Joel School District , 1 Dinev Road , Monroe , NY 10950 , USA change, it is necessary to understand Wechsler e-mail: [email protected] from a historical perspective. Like other great R. Dumont , Ed.D. psychologists, Wechsler was infl uenced by Department of Psychology , Fairleigh Dickinson researchers who preceded him and by the domi- University (TWH101) , 1000 River Road , nant schools of thought that typifi ed and Teaneck , NJ 07666 , USA infl uenced the culture of his time. As is often the e-mail: [email protected] case in science, serendipitous events also played J. O. Willis , Ed.D. a role in shaping Wechsler’s views of psychologi- Department of Education , Rivier University , 419 Sand Hill Road , Peterborough , NH 03458 , USA cal assessment. Wechsler has, in turn, had a pro- e-mail: [email protected] found impact on the manner in which modern

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 163 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_12, © Springer Science+Business Media New York 2015 164 M. Benisz et al. psychologists assess intelligence. Many of his effects, reaction times, and pressure thresholds views on intelligence were forward thinking and (Cattell and Farrand 1896 ). Although at the time continue to be relevant today. Cattell used the term mental tests to describe his measurements, these tests were, for the most part, quite different from tests that would later become People Who Infl uenced Wechsler the core of modern-day cognitive assessment. However, at least two of the tests, a test of mem- In the United States, at the turn of the twentieth ory, where the students had to remember and century, the fi eld of psychology began to change repeat strings of digits, and a timed “test of auto- from a science that studied consciousness to a maticity” that required the students to name col- more practical science that mirrored the ambi- ors, are very similar to tasks found on modern tions and attitudes of a pragmatic American cognitive tests. Along with developing the tests society. Much attention was focused upon the themselves, Cattell and Farrand also gathered new functionalist school of psychology that data on each of the students’ cultural and socio- shifted the considerations of psychologists away economic backgrounds, as well as aspects related from structuralism – the study of the structure of to their physical health. Cattell’s early work on the mind, or what the mind is – to functionalism test development and standardization gave rise to – the study of the functions of the mind, or what the fi eld of intelligence testing and must have the mind does (Schultz and Schultz 2012 ). The made quite an impact on Wechsler who studied functionalist school was directly related to the under Cattell at Columbia University. birth of applied psychology (which itself would In 1917, Cattell was fi red from Columbia give rise to clinical psychology), which took psy- University for opposing the United States’ con- chology out of the laboratory and placed it scription policy during World War I. Years later, squarely into the real world. To examine the Cattell successfully sued Columbia University, functions of the mind, psychologists focused on and in 1921, he used the money from that settle- the study of memory, perception, feeling, imagi- ment to start The Psychological Corporation, a nation, and judgment. In order to successfully company dedicated to fi nding solutions to indus- study these mental activities, psychologists of the trial problems via the science of applied psy- time needed to develop and then use objective chology (Pillsbury 1947 ). One can only speculate assessment methods. on the infl uence that Cattell’s diffi cult decision One major infl uence on Wechsler’s profes- with regard to involvement in World War I sional as well as personal life was James McKeen might have had on Wechsler’s own decision to Cattell. Cattell was the fi rst professor of apply for conscientious objector status when his psychology in the United States and, through his time came to serve in the Army. Later, after grad- work, helped establish psychology, once regarded uating from Columbia, Wechsler would turn to as a minor fi eld of study, or in some cases a pseu- Cattell to seek employment at The Psychological doscience, as a legitimate science. Corporation, the same company that would even- Cattell was interested in studying human abili- tually publish Wechsler’s intelligence scales. ties and combined his early laboratory training in Another major infl uence on Wechsler’s pro- Germany with his interest in statistical analysis, to fessional life was Robert Sessions Woodworth, a gather large amounts of data he could tabulate and student of Cattell’s and a scholar of psychology subject to statistical scrutiny (Cattell 1890 ). At the who also taught at Columbia University. Like turn of the nineteenth century, Cattell published Cattell, Woodworth researched mental tests and the results of his study conducted with Livingston in 1911, together with Frederic Lyman Wells, Farrand in which they assessed 100 students developed a version of the Substitution Test enrolled at Columbia University on an array of (Woodworth and Wells 1911 ), a form of which is measures that included tasks of sensorimotor familiar to the present-day psychologist as the 12 From Psychometric Testing to Clinical Assessment… 165

Digit Symbol, now Coding, subtest on the higher than CA, the child was considered to be Wechsler scales (Wechsler 1944 ). In 1916–1917, advanced. If the MA was lower than the child’s Woodworth delivered a series of lectures which CA, the child was considered to be behind or were later published under the title Dynamic “retarded.” Many of the items on the original and Psychology (Woodworth 1918 ). Woodworth revised Binet-Simon test are familiar to psychol- explained that his system of dynamic psychology ogists today. These include tasks such as unfi n- was concerned with the cause and effect relation- ished pictures, where the child has to fi nd the ships of human behavior. He wanted to under- important part that had been omitted in a picture, stand how humans learn and “what leads them to and repetition of fi gures, in which the child think and act as they do” (Woodworth 1918 , repeats a series of dictated digits, the length of p. 34). In the summer of 1917, Woodworth served which varies according to the age of the child. as chief psychological examiner at the naval base The Binet-Simon Intelligence Scales also in Brooklyn, NY, where his unit was responsible included a test which required the child to defi ne for administering group and individual tests to words, and another that required the child to over 1,000 men as part of a trial that would even- make change from 20 sous, a type of arithmeti- tually lead to the wide-scale testing of soldier cal reasoning. Each of these tasks is similar in recruits during World War I (Yerkes 1921 ). Of all nature to what Wechsler later included in his the people who shaped Wechsler’s persona and scale as the Picture Completion, Digit Span, destiny, Woodworth is perhaps the most impor- Vocabulary, and Arithmetic subtest (Boake tant for three reasons. First, Woodworth was 2002 ). Binet and Simon’s concept of intelligence Wechsler’s mentor at Columbia University, and was how well a person could adapt to a particular Wechsler completed both his master’s degree circumstance (Binet and Simon 1916a ). This and his doctorate under Woodworth’s tutelage. idea is refl ected in the types of tasks they chose Second, it is likely that Woodworth helped secure for their test. Wechsler a position in the Army as a psychologi- The Binet-Simon test was brought to the cal examiner during World War I (Wasserman United States in 1908 by Henry Goddard (1916 ), 2012). Third, it is likely that Woodworth steered who arranged for the test to be translated into Wechsler away from any aspirations of employ- English. An American revision of the Binet- ment in academia. Simon scale was developed in 1915 by Robert Cattell’s desire to design a meaningful and M. Yerkes et al., and was called the Yerkes- purposeful test was achieved a little more than a Bridges Point Scale. Instead of grouping the tests decade later by Alfred Binet. In 1904, in the by age level as in the Binet-Simon, the test items Republic of France, the Minister of Public were consolidated into subtests and arranged in Instruction sought to fi nd a means by which chil- order of diffi culty, with the examinee able to earn dren who were intellectually impaired could be points for each correct response. The point scale excluded from public education and moved to model was later adopted by Wechsler for use in special classrooms (Binet and Simon 1916b , his tests (Kaufman 2009 ). In his role as Chief of p. 9). In order to assist in this endeavor, Binet the Section of Psychology in the United States and his colleague Théodore Simon developed Army during World War I, Yerkes, had an indirect and published in 1905 and revised in 1908 a set impact upon Wechsler’s future career as a devel- of tasks, arranged in order of diffi culty, that were oper of psychological tests. administered individually to children. The test In 1916 Lewis Terman, a professor at Stanford produced a score called the “mental age” (MA) University and a pioneer in educational psychol- that was compared to the child’s actual chrono- ogy, published the Stanford Revision of the logical age (CA). If the derived mental age was Binet-Simon Scale , now universally known as the same as the child’s chronological age, the the “Stanford-Binet” test. He adopted William child was considered to be average. If MA was Stern’s ( 1912 ) suggestion that mental age 166 M. Benisz et al. divided by chronological age be used for an Wechsler’s Early Life intelligence quotient or IQ. The Stanford-Binet was an extremely popular test for many decades Wechsler’s early life did not presage his career, (Louttit and Browne 1947) and in its current edi- and he would have seemed an unlikely candidate tion is still in use today. Terman believed that a for fame. The youngest of seven children of good intelligence test could accurately assess Jewish parents, he was born in the Kingdom of abstract thinking, and therefore he emphasized Romania in 1896, during a time of severe eco- the use of verbal and language-based tests of nomic depression, which led to the enactment of arithmetical reasoning and abstract thinking anti-Semitic laws and the persecution of Jews (Terman 1921 ). (Matikainen 2006 ). At the time, Jewish residents Two other personalities that helped shape of Romania were not eligible for Romanian Wechsler’s view on intelligence were Charles citizenship and were thus excluded from many Spearman and Louis Thurstone. Spearman is vocational occupations which were open only to credited with having constructed the fi rst theory citizens (Iordachi 2002 ). Even Jewish children of intelligence (Kaufman 2009 ). Spearman were not immune from discrimination. As non- ( 1904 ) theorized that individual intelligence citizens they were barred from attending public could be represented by a unitary factor that he schools unless their parents paid prohibitively called g . He was able to demonstrate that differ- expensive tuition fees. These conditions led to a ent measures of mental abilities were all posi- mass exodus of Romanian Jews, who left the land tively correlated, which indicated that there was of their birth for more tolerant countries. Between an underlying global factor that represented 1900 and 1902, Wechsler’s parents and siblings intelligence. Spearman also theorized that cer- made their way to the United States and found tain mental tests would be better at measuring g refuge in New York City. Before Wechsler’s 11th than others. Wechsler referred to Spearman’s birthday, both of his parents had died, and he was discovery of g as “one of the great discoveries of raised by his brother, Israel Spaner Wechsler, psychology” (Wechsler 1944 , p. 6). Wechsler’s who would himself later become a renowned viewpoint as to the importance of g did not neurosurgeon. Wechsler attended City College of change over time. Kaufman (2009 , p. 45) writes New York, earning an A.B. in 1916, and did his that when Wechsler visited him in 1975, he told graduate work at Columbia University, earning a Kaufman’s students that “nothing is more master’s degree in 1917 and a Ph.D. in 1925. His important than g for understanding intelligence. fi rst published article (his master’s thesis) was a Global ability is the ability that underlies my study of memory in patients with Korsakoff’s IQ tests.” syndrome (Wechsler 1917a ). In his article, In contrast to Spearman’s unitary intelligence Wechsler lists the tests he used to evaluate the factor theory, Thurstone, a pioneer in the fi elds of participants in his study. What is noteworthy is psychometrics, believed that intelligence com- that Wechsler used a test he refers to as “auditory prised seven independent groups factors (Guilford memory span for digits” and the Knox Cube Test 1972 ). Thurstone’s work in factor analysis led him (a nonverbal memory test which employs sequen- to formulate a model of intelligence centered tial taps on four blocks), which he refers to as a around “Primary Mental Abilities” (PMAs) (e.g., “visual memory span for movement” (Wechsler Thurstone 1938). These included verbal compre- 1917a , p. 416). Versions of the auditory Digit Span hension, word fl uency, number facility, spatial are still found on the Wechsler Intelligence Scale visualization, associative memory, perceptual for Children, Fifth Edition (WISC-V; Wechsler speed, and reasoning. Thurstone’s model infl u- 2014) and the Wechsler Adult Intelligence Scale, enced hierarchical models of intelligence and Fourth Edition (WAIS-IV; Wechsler 2008 ). A would later infl uence the way Wechsler defi ned cube-tapping visual memory test (long absent intelligence (Matarazzo 1972 ) 12 From Psychometric Testing to Clinical Assessment… 167 from Wechsler scales) is included in the WISC-IV group and individual tests was reported to be 0.5, integrated (Wechsler et al. 2004 ) and in the which led Yerkes to decide that group examina- Wechsler Nonverbal Scale of Ability (WNV; tion was acceptable. From these trials, two group Wechsler and Naglieri 2006 ). It is also interesting administered tests were developed: the Army to note that in the Korsakoff syndrome study, Alpha, designed for English-speaking recruits, Wechsler performed a “qualitative” (Wechsler and the Army Beta, administered to recruits who 1917a , p. 417) as well as a quantitative “analysis were illiterate or had limited profi ciency in the of errors” (Wechsler 1917a , p. 423). This utiliza- English language. If a soldier was unable to pass tion and integration of separate measures of ver- the group examination, he was administered an bal and nonverbal abilities, coupled with an individual test such as the Stanford-Binet, the interpretive approach that combined both qualita- Yerkes-Bridges Point Scale, or a test that had been tive and quantitative analyses, would feature developed by Yerkes’ team called the Army prominently in Wechsler’s assessment tools sev- Performance Scale. This testing project, which eral decades later. would eventually result in the assessment of over 1.7 million soldiers, including over 83,500 individual assessments required a large number of World War I and Psychological men with knowledge of psychological testing to Testing in the Army administer these tests (Yerkes 1921 ). On June 5, 1917, the day before Wechsler’s In 1917 at the beginning of World War I, a meeting graduation with a master’s degree from Columbia of the Society of Experimental Psychologists was University, the US government required all men held at Harvard University. Yerkes, then president between the ages of 21 and 31 to register for the of the American Psychological Association, draft to fi ght in World War I. Wechsler, who had addressed the attendees. His goal was to see how turned 21 in January of that year, was no excep- American psychologists could unite and contrib- tion. Wechsler’s war-time service is not only ute to the war effort (Yerkes 1921 ). One of the noteworthy for how it infl uenced his later career, identifi ed needs of the Army was the assessment but it is also somewhat controversial. Wasserman of new recruits in order to know which tasks they ( 2012 ) suggested that Wechsler tried to evade the were capable of performing. Individual testing war-time draft and notes that “Given his efforts was not a practical solution for the Army as it to avoid military service in 1917, it might be required signifi cant amounts of time as well as considered ironic that the skills he acquired and highly trained test administrators. A protocol contacts he made during his military service for testing large groups of recruits needed to be would shape his career in assessment and test developed. After receiving the necessary approv- development” (p. 33). Specifi cally regarding als from the Army, Yerkes assembled a team of Wechsler’s draft status, Wasserman wrote: experts to develop a battery of tests that could be administered to large groups of men. Most of the After the U.S. Congress declared war on Germany in April, 1917, Wechsler ( 1917b) completed his tests were adapted from existing assessments. required registration for the draft, listing himself as The team evaluated tests using criteria such as a “Consciencious [sic ] Objector” and as an alien speed of testing, speed of scoring, minimal amount who was a citizen of Romania, who was disabled of writing, and (perhaps surprisingly for the time) by “Near Sightness” [ sic] and “Physical Unfi tness.” To the item asking about his occupation he wrote the test’s indifference to the amount of schooling a “Am student in school of philosophy.” Wechsler’s recruit had previously received. Yerkes (1921 ) draft registration thus used multiple methods to credited Arthur S. Otis for devising and adapting avoid being drafted—claiming status as a consci- the majority of these tests. The tests were tried out entious objector, claiming exemption from military service by reason of alien (noncitizen) status, and several times, and group scores were correlated claiming physical defi ciencies that would disqual- with individual scores. The correlation between ify him for military service. . . . Wechsler’s status 168 M. Benisz et al.

as a noncitizen native of Romania, some 15 years previously fi led a petition for naturalization with after his arrival in the United States also put him at the “Supreme Court of Kings County, Brooklyn, risk. As an alien, he could not be drafted, but he could be deported. (2012 , pp. 31–32) NY” on March 6, 1918. Since this civilian petition or “fi nal papers” could only be fi led three Wasserman based these statements on infor- years after fi ling fi rst papers, it seems likely that mation derived from a copy of Wechsler’s draft Wechsler fi led his declaration of intent sometime registration card (Wechsler 1917b ). Careful around 1915 and thus became subject to US draft examination of the draft card and of the selective regulations in 1917. Wasserman also made the service laws in effect at that time does not fully assumption that Wechsler fi lled out his own reg- support Wasserman’s assumptions. Although he istration card. This does not seem likely. The was not a citizen of the United States, Wechsler entire draft card appears to have been completed was most certainly subject to the draft under by the registrar. The draft card comprised two the existing draft laws. The Selective Service sides, a front and a back. The back part of the Regulations of 1917 (U.S. Offi ce of the Provost form is titled “Registrar’s Report,” fi lled out by Marshal General 1918 , p. 52) stated that “when the registrar; it included observations that the an alien has declared his intention to become a registrar made about the prospective recruit. It is citizen, regardless of how long ago, he is still here that the registrar, not Wechsler, noted that liable to draft, even though he has not in the Wechsler had possible disqualifying disabilities meantime applied for fi nal papers.” In 1917, the such as nearsightedness and physical unfi tness. It process for becoming a US citizen had two steps. seems likely that these were intended to be the First the alien would declare his intention to objective observations of the registrar, as the reg- become a citizen. This process was known col- istrar had to certify that his own answers were loquially as fi ling “fi rst papers” (DeSipio 1987 ). true. Several years later, the alien would fi le a petition One of the answers on the card can be mis- for naturalization, so-called fi nal papers. On this read easily because of the handwriting, which point, it is interesting to note that on the Wechsler- may be why Wasserman assumed that Wechsler Bellevue Intelligence Scale Form II (Wechsler fi lled out his own form. For item 7 “What is your 1946), Question 8 of the Comprehension subtest present trade, occupation or offi ce?” Wasserman is, “Why does the United States require that a noted that Wechsler answered “am student in person wait at least two years from the time he school of philosophy” (Wasserman 2012 p. 31). makes application until the time he receives his Upon careful examination, we believe that the fi nal citizen papers?” Listed as one of the highest answer actually reads “A.M. Student in school of score answers is, “Gives them a chance to prove Philosophy” where A.M. is likely an abbrevia- their sincerity and desirability.” tion of artium magister , or master of arts—the It is almost certain that Wechsler had already degree that Wechsler was completing at the time fi led his declaration of intent before the war of the registration. Wechsler, in fact, was broke out. Item 4 on Wechsler’s draft registration awarded the degree the very next day (June 6, card asks: “Are you (1) a natural-born citizen, (2) 1917) at the Columbia University commence- a naturalized citizen, (3) an alien (4) or have you ment ceremony. Furthermore, a careful examina- declared your intention (specify which)?” The tion of Wechsler’s signature on the card shows answer that is written on the card is “1st paper.” that it does not match the cursive handwriting Further proof that he had declared his intention used to complete the rest of the document, and, to become a citizen can be found in his petition additionally, the color of the ink used for the sig- for naturalization, fi led one day after Wechsler nature seems different from the rest of the form. entered the Army on May 12, 1918 (Wechsler It is possible that Wechsler was a bona fi de con- 1918). On the form, it notes that Wechsler had scientious objector, willing to serve his adopted 12 From Psychometric Testing to Clinical Assessment… 169 country but requesting to do so in a noncomba- 1925 with Woodworth as his mentor. Much of tant role. It is equally possible that Wechsler the research that Wechsler had done at the claimed to be a conscientious objector in order Sorbonne was used in his doctoral dissertation. to escape the possibility of live combat. There is, One of the research issues Wechsler sought to however, no evidence that favors either of these explore was whether there was evidence for possibilities nor is there any evidence that the existence of a general emotional factor, Wechsler tried to avoid (noncombatant) military similar to the general intelligence factor iden- duty. tified earlier by Spearman. Wechsler, however, Wechsler did enter the Army in 1918 and did not find any evidence to support this factor while waiting induction at Camp Yaphank in (Matarazzo 1972 ). Long Island, NY received some preliminary After graduating from Columbia, Wechsler training in the administration and scoring of the was unable to fi nd steady employment. We can Army Alpha Test (Matarazzo 1972 ). In May of speculate that some avenues of employment 1918, he attended basic training at Fort Greenleaf, might not have been open to Wechsler because of Georgia, and in August was assigned to Fort his religion. In those years, careers in academia Logan, Texas, where he was part of the were not always open to Jews (Schultz and psychology unit (Yerkes 1921 ). His responsibili- Schultz 2012 ). Even Wechsler’s mentor, Robert ties at Fort Logan consisted primarily of adminis- S. Woodworth, was not immune to the prejudices tering individual psychological tests, one part of of his times. In 1929, Woodworth made it clear to the mass psychological examining of recruits that a Jewish student named Daniel Harris that Harris occurred during World War I. It is in the Army could not become his assistant on account of that Wechsler’s main ideas about the nature of Harris’s religion. Woodworth advised Harris to psychological evaluation were formulated seek career opportunities outside of academia (Boake 2002 ). (Winston 1996 ). It is interesting to speculate about how different the history of intelligence testing might have been Postwar Experiences and Training had Wechsler been able to obtain a career in academia. Perhaps Wechsler might have settled into a At the end of the war, while soldiers stationed in long, distinguished professorship with many pub- Europe waited to return home, some were able to lications but nary a test to his name. participate in higher education courses such as After short stints at various locations (includ- law and medicine. This program was known as ing 5 years in private practice), Wechsler eventu- the American Expeditionary Forces (AEF) ally found employment at The Psychological University (Cornbise 1997 ). Although it is unclear Corporation as an unsalaried employee. During exactly how, Wechsler was able to take advantage his time there, he performed a study, funded by of this program. He transfered to France and later the newspaper the New York World, that mea- to the United Kingdom, where he worked with sured the intelligence of women, using a sample notable contributors to the emerging science of of chorus girls (Edwards 1974 ). He also devel- human intelligence such as Charles Spearman and oped a test that measured the intelligence and Karl Pearson, both of whom are renowned for alertness of taxi drivers for the Yellow Cab their research into statistical analysis. Company of Pittsburgh. The test employed Wechsler obtained a 2-year fellowship to the mechanical instruments that Wechsler developed Sorbonne in France and there studied the “psy- himself. Wechsler had already patented a chophysiology of emotions” (Boake 2002 ). He machine he called the photogalvanograph that returned to the United States in 1922 and com- measured the “variations in the electrical conduc- pleted his doctorate at Columbia University in tivity of the human or animal skin every time the 170 M. Benisz et al. individual is subjected to an emotion” (Wechsler loaded on verbal items that sometimes produced 1924). Wechsler criticized those who would use scores that did not refl ect the real-life intelligent the photogalvanograph for determining a per- functioning of the examinee. Wechsler was aware son’s guilt of a crime (Jones and Wechsler 1928 ). of some of the problems that existed with earlier At the time, some utilized the machine with a test tests because of his fi rst-hand experience testing that included reading a list of words to a suspect. soldiers during World War I. In 1935, Wechsler The list was purposefully designed to include argued that the mental age or IQ assigned to an some words that were meant to trigger an emo- examinee often did not accurately describe the tional response. Jones and Wechsler were able to functionality of that in individual in real life. show that the test results lacked reliability if the Wechsler understood that a person could score procedures were not standardized properly, poorly on a test and yet function adequately in including the careful location of the trigger society. Wechsler related an early experience he words in the list. Wechsler’s forward-thinking had in the Army in which he evaluated a 28-year- approach to test validity and reliability would old, white, Oklahoman soldier who had failed the come to good use when he developed his own Army Alpha and Beta tests. He administered intelligence test. both a Stanford-Binet and a Yerkes-Bridges Point Scale, and the man obtained a mental age of eight yet was able to function perfectly well as a sol- The Development of the Wechsler- dier in the Army. Before joining the Army, the Bellevue Intelligence Scale man worked as an oil driller, earning enough money to support his family. Wechsler also stated Wechsler was hired in 1932 by Bellevue Hospital that the tests were completely inadequate in in New York, where he eventually became the appropriately measuring the true mental abilities hospital’s chief psychologist, remaining until he of foreign-born adults and African-Americans retired in 1967. In 1932, Wechsler wrote a short (1935a ). paper describing what he saw as the advantages The Wechsler-Bellevue Intelligence Scale was of the Army Alpha test over the Stanford-Binet, published in 1939 and it contained several inno- namely, the examiner’s ability to analyze the vative features (Kaufman 2009 ). Wechsler used subtests of the Army Alpha test to determine standard scores, which he referred to as deviation individual “special abilities and disabilities”. IQ scores, acknowledging that he borrowed this Here is the fi rst time that analysis of strengths concept from other tests (Wechsler 1949a ). He and weaknesses is mentioned. In 1935, Wechsler set the mean score, somewhat arbitrarily published a book he considered to be one of his (Kaufman 2009 ) at 100 with a probable error of best works, The Range of Human Capacities . In 10 or standard deviation of 15, which was (inten- this book, Wechsler used Army Alpha data tionally or not) fairly close to the middle of the among other sources to argue that abilities peaked range of standard deviations of the ratio IQs at a certain age and then began to decline. He found on the revised Stanford-Binet (Terman and also argued very strongly that psychologists had Merrill 1937 ; see McNemar 1942 ). The deviation overestimated the range of variations among IQ score offered several distinct advantages over individuals and that human beings were actually the existing ratio IQ scores. As opposed to the surprisingly similar. ratio IQ score, the statistical meaning of a score By the time Wechsler began developing his on Wechsler’s test did not vary from year to year. fi rst intelligence test during the 1930s, he had For example, a score of 115 would be in the 84th identifi ed several key problems with existing percentile at any age and a score of 85 would tests that he felt needed to be addressed. He always be in the 16th percentile. If an examinee believed that the existing tests were heavily maintained roughly the same ability level on the 12 From Psychometric Testing to Clinical Assessment… 171 test compared to peers, the examinee’s score Almost all of the tasks that Wechsler chose would not vary just as a function of age. The for his original test were in fact adapted or score was meaningful, regardless of age. adopted from other tests; primarily those Deviation IQ scores were also better suited to Wechsler was intimately familiar with from his measure developing cognition in children as cog- experiences as an examiner in the Army. nition does not increase uniformly as children Kaufman (2009 ) notes that “the similarity of age (e.g., the difference between the average cog- Wechsler’s original set of subtests to the tasks nition of a 1-year-old and 3-year-old is greater used to evaluate recruits, soldiers, and offi cers than the difference between a 15-year- old and an during World War I is striking.” In fact, Wechsler 18-year old). Another advantage of the deviation ( 1944) stated this explicitly. He did not even IQ was that mental ages and ratio IQs were espe- change the names of many of the tests he bor- cially meaningless for adults and required select- rowed. Wechsler’s goal in developing the ing an arbitrary maximum chronological age to Wechsler-Bellevue was not to create new sub- use for older examinees and ways of designating tests or tasks but rather to select the best avail- mental ages above that level. able measures of intelligence and combine them Wechsler also made use of the subtest point- into a standardized, norm-referenced battery. scale method that had appeared on the Yerkes- Wechsler ( 1944 ) gives the source or sources for Bridges Point Scale as opposed to the age scale most of his tests. For example, the Information that appeared on the Stanford-Binet. The and Comprehension subtests were both adapted Wechsler-Bellevue also corrected the overem- directly from the tests developed for use by the phasis on verbal tests that appeared on tests the Army. Five of the ten items on the original Stanford-Binet and similar tests. Wechsler Comprehension subtest were taken directly, weighted the verbal and nonverbal (performance) some with only slight modifi cations, from the tests more or less equally (Kaufman 2009 ). Comprehension test of the Army Individual Finally, Wechsler used census data to create an Examination, while at least three other items unbiased normative sample matching age, gen- were taken from the Army Alpha (Yerkes 1921 ). der, and occupation to US census data (Wechsler Items on the Arithmetic subtest were also 1944). Although his urban sample could not be adapted from the Army Individual Examination. matched exactly with rural occupations, Wechsler Digit Span was derived from the Binet scales. used Yerkes’s (1921 ) report to fi nd the IQs of Although Wechsler ( 1944 ) described in detail people with specifi c urban occupations that were how he developed the word list for his optional similar to the IQs of agricultural workers. This vocabulary test, it too was derived from the allowed Wechsler to substitute urban occupa- Stanford-Binet test that was used in the Army. tions for rural ones. For the time, these methods Wechsler’s Block Design was derived from a of standardization and norming, carried out test constructed by Samuel Kohs called Kohs personally by Wechsler and colleagues without Block test (Kohs 1920 ). The Army Performance the support of a publisher, were quite advanced Scale had a somewhat similar test called Cube (Kaufman 2009 ). Construction. Wechsler ( 1944) notes that he The Wechsler-Bellevue was divided into two changed the colors of the Kohs Block test, scales, a Verbal scale and a Performance scale. The which had red, blue, yellow, and white blocks, subtests on the Verbal scale were Information, to a simpler version using just two colors, red Comprehension, Arithmetic, Similarities, and Digit and white. This change might have been inspired Span, with Vocabulary serving as an alternate test. by the Army’s Cube Construction test in which The subtests on the Performance scale were the blocks had only two colors, red and natural Digit Symbol, Picture Completion, Block Design, wood (Yerkes 1921). Picture Arrangement also Picture Arrangement, and Object Assembly. appeared on the individually administered 172 M. Benisz et al.

Army Performance Scale, although it had origi- intelligent behavior and improved the value of the nally been developed for the Army Beta test. psychometric instrument as a clinical tool. Subsequent to Wechsler’s death, the publisher of Wechsler adapted its pictures for his own test. revisions of his batteries yielded to political cor- Both Picture Completion and Digit Symbol rectness and removed virtually all of the clinically were also borrowed from the Army Beta test. charged items, the very ones that Wechsler believed Digit Symbol (Coding) had been created by would assess aspects of EI [emotional intelligence]. (p. 260) Otis and was based upon several versions of an older test called the substitution test, including the one designed by Woodworth (Yerkes 1921 ). Wechsler changed the administration time of Theory of the Wechsler-Bellevue Digit Symbol from 120 s to 90 s, although for Test the Wechsler Adult Intelligence Scale, Third Edition (WAIS-III), the allotted time was Wechsler did not base his tests on any hierarchi- restored to two minutes (Wechsler 1997 ). cal theories of intelligence, although he clearly Object Assembly was derived from the Army saw the subtests as measuring different aspects Performance Scale. of overall general intelligence or g . The most Wechsler believed that his tests were primarily recent editions of the Wechsler scales (Wechsler to be used as clinical instruments, despite the fact 2008 , 2012 , 2014) have all been updated to that they could also provide important quantita- more accurately refl ect current psychological tive information. Part of Wechsler’s success and and neuropsychological research as well as the popularity of his instruments can be attributed some aspects of a three-stratum, hierarchical to his ability to look forward and remain ahead of theory of intelligence known as the Cattell- the curve by developing tests that met the Horn-Carroll (CHC) theory (Flanagan et al. demands of the professionals that used them. 2013 ). Confi rmatory actor analysis supports the However, more important than the tests he devel- fi t of the Wechsler Preschool and Primary Scale oped is the way in which Wechsler transformed of Intelligence, Fourth Edition (WPPSI-IV; the “fi eld of IQ interpretation from psychometric Wechsler 2012 ), WAIS- IV, and WISC-V into measurement to clinical assessment ” (Kaufman this model (Lichtenberger and Kaufman 2013 ; 2009, p. 37). Wechsler’s view of intelligence test- Raidford and Coalson 2014; Wechsler 2014 ). ing was that it was only one part of personality However the original theory behind the Wechsler- testing. He believed that an individual evaluation Bellevue and subsequent Wechsler scales is was not complete without an assessment of non- based on the author’s defi nition of intelligence intellective factors such as temperament, moral- (Wechsler 1939 ): ity, and social values (Edwards 1974 ; Wechsler 1943 , 1950). He included items in his scales that The aggregate or global capacity of the individual to act purposefully, to think rationally and to deal would give the examinee an opportunity to effectively with his environment. It is global express rational thinking, purposeful thought, because it characterizes the individual’s behavior and effective problem solving. For example, as a whole; it is aggregate because it is composed because he felt it was a clinically useful test for of elements or abilities which, though not entirely independent, are qualitatively differentiable. (p. 3) diagnosing low mental ability, Wechsler included Digit Span in the original Wechsler-Bellevue According to Wasserman and Kaufman (in even though it was a poor measure of overall press ), Wechsler believed that this defi nition intelligence (Wechsler 1944 ). According to served to encompass the various theories that Kaufman and Kaufman (2001 ): had infl uenced him. So when he speaks of “The aggregate or global capacity,” he could certainly Wechsler embraced the inclusion of items with clinical content in his test batteries, believing that have been referring to Spearman’s general intel- they enhanced the more complete measurement of ligence factor. His emphasis on intelligence 12 From Psychometric Testing to Clinical Assessment… 173 being “composed of elements or abilities” is help of an interpreter. The results of the test con- similar to Thurstone’s primary mental abilities. fi rmed Goddard’s suspicions: the immigrant was When Wechsler refers to the capacity “to think mentally defi cient. Goddard remained unswayed rationally,” was he acknowledging Terman’s even when the interpreter told him that upon his emphasis on intelligence being related to the own arrival to the United States, he himself would capacity for abstract thinking? Finally Wechsler’s have been unable to answer many of the items on defi nition includes the concept of dealing effec- the test (Schultz and Schultz 2012 ). tively with one’s environment, which pays trib- The Army test data that had been gathered ute to Binet’s notion of intelligence being the during the war were made available to psy- ability to adapt to different circumstances. chologist Carl Brigham, an assistant professor at Princeton University. In 1923, he published A Study of American Intelligence , summarizing The Controversial Aspects the data of the various Army tests. The fore- of Intelligence Testing word to the book, written by Yerkes, warned that the United States could not “afford to Wechsler was a staunch defender of the utility ignore the menace of race deterioration or the and necessity of proper intelligence testing, evident relations of immigration to national something he referred to as an “intelligent test” progress and welfare” (Brigham 1923 ). The data (Wechsler 1966 ). In a letter to the New York showed that the admittance of “intellectually Times (Wechsler 1949b ), he noted that IQ tests inferior” immigrants, defi ned as those immi- were reliable and valid and were more “defi - grants from non- Nordic countries, contributed to nitely diagnostic” than an electrocardiogram. the decline of American intelligence. According However, intelligence testing was controver- to Brigham, the solution was to heavily restrict sial in Wechsler’s day and remains controversial immigration to the United States, and in 1924 the today (Kaufman 2009 ). Part of the reason for US Congress did enact legislation that sharply this controversy can be attributed to some of curtailed levels of immigration (Kamin 1974 ). Wechsler’s contemporaries who advocated the It is worthwhile to correct a misperception use of intelligence tests in support of their that has recently appeared in the literature opinions on race, eugenics, and immigration. regarding the lower scores of non-native English For example, in the fi rst chapter of the Stanford- speakers on the Army Beta. Ortiz et al. ( 2012 ) Binet manual published in 1916, Terman was wrote: explicit about his goals for the Stanford-Binet In examining his data on nearly 1.75 million test. It was his hope that the routine testing of American men, Yerkes noted that the average raw “feebleminded” individuals would result in some score on the Beta for native English speakers was a type of government action to limit their ability to stout 101.6, which classifi ed them as Very Superior (Grade A). In contrast, the average raw score for reproduce. For Terman, the primary use of the non-native English speakers (who could also not intelligence test was to identify “mental defec- read English) came in at only 77.8, which classi- tives.” The link between crime, vice, “industrial fi ed these individuals as Average (Grade C). For ineffi ciency,” and people with limited mental Yerkes, as well as the contingent of other notable psychologists working with him (e.g., Carl abilities was unequivocal. Brigham, David Wechsler, Lewis Terman), the Goddard advocated for the use of mental tests results confi rmed their own beliefs—that immi- to identify the feebleminded among prospective grants, particularly those from certain countries immigrants to the United States. In 1912, and from lower classes, were merely displaying their inherited lack of intelligence. (pp. 537–538) Goddard visited Ellis Island and administered the Binet test to an immigrant that he perceived as Ortiz et al. are correct in noting the manner in being feebleminded. The immigrant spoke no which some of psychologists misinterpreted the English, and the test was administered with the Army data as a result of their prejudiced views 174 M. Benisz et al. regarding immigrants. There is no evidence to period of a decade. The authors examined the IQ suggest that Wechsler had the same beliefs scores of 103 participant in the BEIP, eight years regarding the intelligence of immigrants as did after they entered the program. All of the children Yerkes, Brigham, and Terman. He was a junior participating in the study had come from deprived member of psychology staff in the Army and had home settings. The researchers found that the chil- yet to become a notable psychologist. In fact, as dren who had participated in the early intervention an Eastern European immigrant, persecuted program had signifi cantly higher scores than a because of his religion, Wechsler would likely control group of deprived children who had not have disagreed with the beliefs of his older con- participated in the program. In their methods sec- temporaries. Wechsler developed his test in part tion, the authors report that the participants’ IQs as an attempt to correct some of the fl aws that were measured using the WISC-IV ( 2003 ). It existed with earlier tests, such as the emphasis seems almost poetical that over a century after he on verbal tests. He defended intelligence tests was forced to leave Romania, his native land, one against accusations of cultural and racial bias. of Wechsler’s tests was used as part of a program As mentioned previously, Wechsler believed that there to improve the quality of life of deprived on the whole people are not very different from children. In light of the results achieved by Fox each other (Wechsler, 1935b). Furthermore, et al. (2011 ), Wechsler’s (1966 ) words in the New Wechsler was careful to note that he had not York Times Magazine (“The I.Q. is an Intelligent included a sample of African-Americans when Test”) provide yet another example of his forward- he developed the Wechsler-Bellevue and cau- thinking approach to intelligence testing: tioned against using the test with that popula- It is true that the results of intelligence tests, and of tion. (Wechsler 1944 ). others, too, are unfair to the disadvantaged, In 2011, Fox et al. published the results of a deprived . . . but it is not the I.Q. that had made study from the Bucharest Early Intervention them so. The culprits are poor housing, broken homes, a lack of basic opportunities, etc. etc. If the Project (BEIP). The goal of this project was to various pressure groups succeed in eliminating introduce early intervention services and foster these problems, the I.Q.’s of the disadvantaged will care into Romania and to study the results over a take care of themselves. 12 From Psychometric Testing to Clinical Assessment… 175

Appendix

Fig. 12.1 The front and back of David Wechsler’s World was eligible for the draft. It is likely that the A.M. in the War 1 registration card. On June 5, 1917, all men ages response to question 7 stands for artium magister, or mas- 21–31 in the United States were required to register to ter of arts. From the National Archives and Records determine eligibility for induction into the military. The Administration. In the public domain response to question 4 “1st paper” indicates that Wechsler 176 M. Benisz et al.

Fig. 12.1 (continued) 12 From Psychometric Testing to Clinical Assessment… 177

Fig. 12.2 View of Dr. David Wechsler seated at a table, galvanograph that Wechsler patented in 1924 (Image conducting a test on a patient. The date of the photo is courtesy of The Lillian and Clarence de la Chapelle unknown. The machine might be a version of the photo- Medical Archives at NYU)

Cattell, J. M. (1890). Mental tests and measurements. R e f e r e n c e s Mind, 15 , 373–381. Cattell, J. M., & Farrand, L. (1896). Physical and mental Becker, K. A. (2003). History of the Stanford-Binet intel- measurements of the students of Columbia University. ligence scales: Content and psychometrics. (Stanford- Psychological Review, 3 (6), 618–648. doi: 10.1037/ Binet intelligence scales, ﬁ fth edition, Assessment h0070786 . Service Bulletin No. 1). Itasca: Riverside. Retrieved Cornbise, A. E. (1997). Soldiers-scholars: Higher educa- from http://www.riverpub.com/products/sb5/pdf/ tion in the AEF, 1917–1919 . Philadelphia: American SB5_ASB_1.pdf Philosophical Society. Benson, N., Hulac, D. M., & Kranzler, J. H. (2010). DeSipio, L. (1987). Social science literature and the natu- Independent examination of the Wechsler adult intelli- ralization process. International Migration Review, gence scale—Fourth edition (WAIS-IV): What does 21 (2), 390–405. the WAIS-IV measure? Psychological Assessment, Edwards, A. J. (1974). Selected papers of David Wechsler . 22 (1), 121–130. New York: Academic. Binet, A., & Simon, T. (1916a). New methods for the Flanagan, D. P., & Kaufman, A. S. (2009). Essentials of diagnosis of the intellectual level of subnormals. In WISC-IV assessment (2nd ed.). Hoboken: Wiley. H. H. Goddard (Ed.), The development of intelligence Flanagan, D. P., Ortiz, S. O., & Alfonso, V. (2013). Essentials in children (the Binet-Simon Scale) (trans: Kite, E. S.; of cross-battery assessment (3rd ed.). Hoboken: Wiley. pp. 37–90). Baltimore: Williams & Wilkins. (Original Fox, N. A., Almas, A. N., Degnan, K. A., Nelson, C. A., work published 1905). & Zeanah, C. H. (2011). The effects of severe psycho- Binet, A., & Simon, T. (1916b). Upon the necessity of social deprivation and foster care intervention on cog- establishing a scientiﬁ c diagnosis of inferior states of nitive development at 8 years of age: Findings from intelligence. In H. H. Goddard (Ed.), The development the Bucharest Early Intervention Project. Journal of of intelligence in children (the Binet-Simon Scale) Child Psychology and Psychiatry, 52 (9), 919–928. (trans: Kite, E. S.; pp. 9–36). Baltimore: Williams & doi: 10.1111/j.1469-7610.2010.02355.x . Wilkins. (Original work published 1905). Georgas, J., Weiss, L. G., van de Vijver, F. J. R., & Boake, C. (2002). From the Binet–Simon to the Wechsler– Saklofske, D. H. (2003). Culture and children’s intel- Bellevue: Tracing the history of intelligence testing. ligence: Cross-cultural analysis of the WISC–III . San Journal of Clinical and Experimental Neuropsychology, Diego: Academic Press. 24 , 383–405. Goddard, H. H. (Ed.). (1916). The development of intelli- Brigham, C. C. (1923). A study of American intelligence . gence in children (the Binet-Simon Scale) (trans: Kite, Princeton: Princeton University Press. E. S.). Baltimore: Williams & Wilkins. 178 M. Benisz et al.

Guilford, J. P. (1972). Thurstone’s primary mental abili- Leipzig: J. A. Barth. Retrieved from http://openlibrary. ties and structure-of-intellect abilities. Psychological org/books/OL24489667M/Die_psychologischen_ Bulletin, 77 (2), 129–143. doi: 10.1037/h0032227 . Methoden_der_Intelligenzprüfung_und_deren_ Iordachi, C. (2002). Citizenship and national identity in Anwendung_an_Schulkindern Romania: A historical overview. Regio: Minorities, Terman, L. M. (1916). The measurement of intelligence . Politics, Society: English Edition, 1 , 3–34. Boston: Houghton Miffl in. Jones, H. E., & Wechsler, D. (1928). Galvanometric tech- Terman, L. M. (1921). Intelligence and its measurement: nique in studies of association. American Journal of A symposium–II. Journal of Educational Psychology, Psychology, 40 (4), 607–612. 12 (3), 127–133. doi: 10.1037/h0064940 . Kamin, L. J. (1974). The science and politics of I.Q . Terman, L. M., & Merrill, M. A. (1937). Measuring intel- Potomac: Erlbaum. ligence . Boston: Houghton-Miffl in. (Forms L and M Kaufman, A. S. (2009). IQ testing 101. New York: Springer. of the Stanford-Binet intelligence scale). Kaufman, A. S., & Kaufman, J. C. (2001). Emotional Thurstone, L. L. (1938). Primary mental abilities intelligence as an aspect of general intelligence: What (Psychometric monographs, Vol. 1). Chicago: University would David Wechsler say? Emotion, 1 (3), 258–264. of Chicago Press. doi: 10.1037/1528-3542.1.3.258 . U.S. Offi ce of the Provost Marshal General. (1918). Keith, T. Z., Fine, J. G., Taub, G. E., Reynolds, M. R., & Selective Service Regulations Prescribed by the Kranzler, J. H. (2006). Higher order, multisample, President under the Authority Vested in Him by the confi rmatory factor analysis of the Wechsler intelli- Terms of the Selective Service Law (Act of Congress gence scale for children—Fourth edition: What does it Approved May 18, 1917, with Supplementary and measure? School Psychology Review, 35 , 108–127. Amendatory Acts and Resolutions) (2nd ed.). Kohs, S. C. (1920). The block-design tests. Journal of Washington, DC: GPO. Experimental Psychology, 3 (5), 357–376. doi: 10.1037/ Wahlstrom, D., Breaux, K. C., Zhu, J., & Weiss, L. G. h0074466 . (2012). The Wechsler preschool and primary scale of Lichtenberger, E. O., & Kaufman, A. S. (2013). Essentials intelligence—Third edition, the Wechsler intelligence of WAIS-IV assessment (2nd ed.). New York: Wiley. scale for children—Fourth edition, and the Wechsler Louttit, C. M., & Browne, C. G. (1947). The use of psy- individual achievement test—Third edition. In D. P. chometric instruments in psychological clinics. Flanagan & P. L. Harrison (Eds.), Contemporary intel- Journal of Consulting Psychology, 11 , 49–54. lectual assessment: Theories, tests, and issues (3rd Matarazzo, J. D. (1972). Wechsler’s measurement and ed., pp. 224–248). New York: The Guilford Press. appraisal of adult intelligence (5th ed.). New York: Wasserman, J. D. (2012). A history of intelligence assess- Oxford University Press. ment: The unfi nished tapestry. In D. P. Flanagan & Matikainen, S. (2006). Great Britain, British Jews and the P. L. Harrison (Eds.), Contemporary intellectual international protection of Romanian Jews, 1900– assessment: Theories, tests, and issues (3rd ed., 1914: A study of Jewish diplomacy and minority pp. 3–55). New York: The Guilford Press. rights . Jyväskylä: University of Jyväskylä. Wasserman, J. D., & Kaufman, A. S. (in press). Biography McNemar, Q. (1942). The revision of the Stanford-Binet of David Wechsler. In R. L. Cautin & S. O. Lilienfeld scale: An analysis of the standardization data . Oxford: (Eds.), Encyclopedia of clinical psychology . Malden: Houghton Miffl in. Wiley Blackwell. Ortiz, S., Ochoa, S. M., & Dynda, A. M. (2012). Testing Wechsler, D. (1917a). A study of retention in Korsakoff with culturally diverse and linguistically diverse popu- psychosis. Psychiatric Bulletin, 2 , 403–451. lations: Moving beyond the verbal-performance Wechsler, D. (1917b). World War One Selective Service dichotomy into evidence-based practice. In D. P. System draft registration cards, 1917–1918 Flanagan & P. L. Harrison (Eds.), Contemporary intel- (Administration. M1509, Roll: 1765679, Draft Board: lectual assessment: Theories, tests, and issues 96). Washington, DC: National Archives and Records (pp. 526–552). New York: The Guilford Press. Administration. Pillsbury, W. B. (1947). Biographical memoir of James Wechsler, D. (1918). Petition for naturalization, David McKeen Cattell 1860–1944 . Washington, DC: Wechsler. United States District Court of the Northern National Academy of the Sciences. District of Georgia, Rome Division. Morrow: National Raiford, S. E., & Coalson, D. L. (2014). Essentials of Archives and Records Administration. WPPSI-IV assessment. Hoboken: Wiley. Wechsler, D. (1924). U.S. Patent No. 1,503,401. Schultz, D. P., & Schultz, S. E. (2012). A history of mod- Washington, DC: U.S. Patent and Trademark Offi ce. ern psychology (10th ed.). Belmont: Wadsworth. Wechsler, D. (1932). Analytic use of the Army Alpha Spearman, C. (1904). “General Intelligence,” objectively examination. Journal of Applied Psychology, 16 (3), determined and measured. The American Journal of 254–256. doi: 10.1037/h0072688 . Psychology, 15 (2), 201–292. Wechsler, D. (1935a). The concept of mental defi ciency Stern, W. (1912). Die psychologischen Methoden der in theory and practice. Psychiatric Quarterly, 9 , Intelligenzprüfung und deren Anwendung an 232–236. Schulkindern . [The psychological methods of testing Wechsler, D. (1935b). The range of human capacities . intelligence and their application to school children.] Baltimore: The Williams and Wilkins Company. 12 From Psychometric Testing to Clinical Assessment… 179

Wechsler, D. (1939). The measurement of adult intelli- Wechsler, D. (2008). Wechsler adult intelligence scale gence . Baltimore: Williams and Wilkins. -fourth edition (WAIS-IV) . San Antonio: Pearson. Wechsler, D. (1943). Nonintellective factors in general Wechsler, D. (2012). Wechsler preschool and primary intelligence. Journal of Abnormal and Social scale of intelligence-fourth edition (WPPSI-IV) . San Psychology, 38 , 101–103. Antonio: Pearson. Wechsler, D. (1944). The measurement of adult intelli- Wechsler, D. (2014). Wechsler intelligence scale for chil- gence (3rd ed.). Baltimore: The Williams and Wilkins dren-fi fth edition: Technical and interpretive manual . Company. Bloomington: Pearson. Wechsler, D. (1946). The Wechsler-Bellevue intelligence Wechsler, D., & Naglieri, J. A. (2006). Wechsler nonver- scale, form II. Manual for administering and scoring bal scale of ability (WNV) . San Antonio: Pearson. the test . (WB-II). New York: The Psychological Wechsler, D., Kaplan, E., Fein, D., Kramer, J., Morris, R., Corporation. Delis, D., & Maerlender, A. (2004). Wechsler intelli- Wechsler, D. (1949a). Wechsler intelligence scale for chil- gence scale for children-fourth edition-integrated dren (WISC) . New York: The Psychological Corporation. (WISC-IV Integrated). San Antonio: The Psychological Wechsler, D. (1949b). Precision instrument [Letter to the Corporation. editor], New York Times Magazine, p. SM4. Retrieved Winston, A. S. (1996). “As his name indicates”: from http://query.nytimes.com/mem/archive/pdf?res= R.S. Woodworth’s letters of reference and employ- FB0F17F73958107A93C0A9178BD95F4D8485F9 ment for Jewish psychologists in the 1930s. Journal Wechsler, D. (1950). Cognitive, conative, and non- of the History of the Behavioral Sciences, 32 (1), intellective intelligence. American Psychologist, 5 , 30–43. 78–83. Woodworth, R. S. (1918). Dynamic psychology, by Robert Wechsler, D. (1952). The range of human capacities (2nd Sessions Woodworth . New York: Columbia University ed.). Baltimore: The Williams and Wilkins Company. Press. Wechsler, D. (1966). The I.Q. is an intelligent test. New Woodworth, R. S., & Wells, F. (1911). Association tests. York Times Magazine . p. 191. Retrieved from http:// The Psychological Monographs , 13 (5). doi: 10.1037/ query.nytimes.com/mem/archive/pdf?res=F20A15FE h0093064 . 3F55117B93C4AB178DD85F428685F9 Yerkes, R. M. (Ed.). (1921). Memoirs of the National Wechsler, D. (1997). Wechsler adult intelligence scale— Academy of Sciences: Vol. 15. Psychological Third edition (WAIS-III) . San Antonio: The examining in the United States Army . Washington, Psychological Corporation. DC: US Government Printing Offi ce. Wechsler, D. (2003). Wechsler intelligence scale for Yerkes, R. M., Bridges, J. W., & Hardwick, R. S. (1915). children-fourth edition (WISC-IV) . San Antonio: The A point scale for measuring mental ability . Baltimore: Psychological Corporation. Warwick & York. A.R. Luria and Intelligence Defi ned as a Neuropsychological Construct 13

Dana Princiotta and Sam Goldstein

One of Luria’s many acclaimed works, The Gedenkschrift to the Father Working Brain (1973a ), was written by only 30 of Neuropsychology years into clinical neuropsychology’s gestation (Cole et al. 2006 ). The immense magnitude of July 16, 2002, marked an important date in the Luria’s legacy is not fully appreciated by most world of neuropsychology. This centennial day cel- scholars because of the political and linguistic ebrated the signifi cant contributions of the Russian deterrents surrounding Luria’s work. Despite neuropsychologist Alexander Romanovich Luria, publishing extensively over the course of 50 born in 1902. A world- renowned scientist, Luria’s years, a complete biography encompassing all theories continue to excite experts more than 30 published works does not exist (Cole et al. 2006 ). years after his death. By the 1980s, following Rationales for this are associated with challenges Luria’s death (August 14, 1977), a survey of in physically locating his publications and in neuropsychologists found that he was revered as translating works written in Russian. In Luria’s “number 1” among the ten founders of neuro- 50-year contribution to neuropsychology, he psychology (Akhutina and Pylaeve 2011 ). authored more than a lifetime’s worth of work. Underestimating this legacy in a single chapter is Nearly 40 years following his death, we continue an unfortunate inevitability when asked to chron- to refl ect on Luria’s philosophies. icle Luria’s unequivocal impact on the practice of Born to Jewish parents in Kazan Russia, psychology and neuropsychology worldwide. Luria’s parents pursued careers in medicine and Procuring the title “father of neuropsychol- dentistry despite political tensions in Russia. ogy” necessitated contributions in both the psy- Luria’s father (Roman Albertovich Luria) was a chological and medical realms as neuropsychology physician at the University of Kazan; his mother was growing to be a recognized fi eld of study. was a dentist. Luria’s sister followed the familial calling and pursued psychiatry. Luria’s formal pursuit of higher education began at the age of 16 D. Princiotta , Ph.D. (*) when he was accepted at Kazan State University. 1845 West Orange Grove Road , Against his father’s wishes, Luria pursued psy- Tucson , AZ 85704 , USA chology. During his studies at the University, he e-mail: [email protected] started the Kazan Psychoanalytic Association. S. Goldstein , Ph.D. He graduated in 1921 at the age of 19. Luria com- University of Utah , 230 South 500 East, Suite 100 , Salt Lake City , UT 84102 , USA pleted his studies while the Russian Revolution e-mail: [email protected] was underway. In 1924, Luria was introduced to

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 181 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_13, © Springer Science+Business Media New York 2015 182 D. Princiotta and S. Goldstein

Lev Semionovich Vygotsky, setting the stage for Luria expounded further and proposed that children a lifetime collaboration and the inauguration of are fi rst introduced to linguistic commands by Luria’s 50-year career. adults and then in turn adopt “self- commands” beginning as commands spoken aloud and then internally; thus, language mediates human behav- The Vygotsky Circle iors (Kaczmarek 1999 ). “The development of any type of complex conscious activity is at fi rst Much of Luria’s earlier documented work arose expanded in character and requires a number of under the guidance of Vygotsky. Luria, Vygotsky, essential aids for its performance; not until later and a team of scholars including medical special- does it gradually become condensed and con- ists, neuroscientists, psychologists, and educational verted into an automatic (motor) skill” (Luria specialists collaborated to investigate the processes 1973a ; Kotik-Friedgut 2006 , p. 48). and psychology involved in an individual’s matura- Luria examined the developmental differences tion as it related to culture (Luria, Cole, & Cole, as they relate to nature and nurture with his 1979 ; Cole 2005a ). This collection of scholars famous twin studies in the 1930s. He studied would become known as the Vygotsky Circle identical and fraternal twins at the Medico- (Luria, Cole, & Cole, 1979 ). Exploiting the terms Genetic Institute in Moscow. Luria believed that historical, cultural , and instrumental , the circle genetic contributions to behavior manifested aspired to “discover the way natural processes such expressly during tasks requiring natural cognitive as physical maturation and sensory mechanisms processes rather than tasks relying on “culturally become intertwined with culturally determined mediated processes” (Cole et al. 2006 , p. 83). processes to produce the psychological functions Working with fi ve sets of identical twins, Luria of adults” (Luria, Cole, & Cole, 1979 , p. 43). Much utilized differing levels of instructions for twin of this work was facilitated in the early years by the dyads. The results suggested almost equal abili- Institute of Psychology in Moscow. ties in the tasks requested by the examiners; how- The term “cultural-historical psychology” was ever, differences were apparent in their language one of the prominent earliest contributions from and the ability to repeat the task from memory 18 the collective Vygotsky Circle. Ideas, theories, months later (Bowden 1971 ). This implied that and research fl ourished under Luria, Vygotsky, functions could be considered acquired, as the and Alexey Leontiev’s guidance (Cole 2005a ). genetic factor was held constant in this study The Vygotsky Circle paid tribute to the interac- (Bowden 1971 ). tion of nature/nurture and natural/cultural factors Subsequent studies focused on the develop- contributing to the development of the mind, with ment of speech in 5-year-olds with minimal particular focus on the cultural, historical, and speech development. A twin that was encouraged instrumental aspects of psychology (Kotik- to verbalize his/her thoughts and desires devel- Friedgut 2006). They aimed to “discover the way oped speech much quicker than a twin confi ned natural processes such as physical maturation to an “indifferent” adult, thereby highlighting the and sensory mechanisms become intertwined importance of the social component necessary to with culturally determined processes to produce acquire language (Bowden 1971 ). This was best the psychological functions of adults” (Luria, summarized by Kotik-Friedgut in 2006: “We Cole, & Cole, 1979 , p. 43). always speak about things we see. We never Within a cultural-historical framework, the speak about things we did not see. Prior to Luria’s underpinnings of Vygotsky’s work supported way of thought, the world was confi ned to con- Luria in speculating that cognitive processes in crete thinking; after Luria had his say, we were humans are organized in a way that thought and all more of the thinking that the ability to think language are closely related, suggesting that spo- abstractly was due to schooling and culture” ken language refl ects thoughts (Kaczmarek 1999 ). (Kotik-Friedgut 2006 , p. 44). 13 A.R. Luria and Intelligence Defi ned as a Neuropsychological Construct 183

Hence when he tried to think, his mind had to World War II and the Emergence detour around these scorched areas and employ of Neuropsychology other faculties with which to learn and try to recover some lost skills” (Luria 1972 , p. 158; In the late 1930s, the Vygotsky Circle dispersed Kolb and Whishaw 2009). This period became as German soldiers invaded the Soviet Union. The the breeding ground for Luria’s systematic infl uence of the Vygotsky Circle resurfaced fol- approach to the brain and mental functions forming lowing the end of World War II. Due to mounting the backbone of neuropsychology (Cole 2005a ). tensions in Soviet Russia in the 1930s, Luria In 1959, Luria described a second case study embarked on a degree from medical school and focused on a war wound suffered by V., an offi cer focused much of his attention toward the apha- of the Polish army. V. was wounded in the left sias. It is not surprising given Luria’s intense occipital-parietal region during the war (Luria interest in the acquisition of language and the 1959 ). Although speech and vision improved and social environment as demonstrated with the V. recovered in these areas after 6 months, V. Vygotsky Circle that Luria would turn his atten- complained of “attacks of giddiness and nausea tion toward the aphasias. Luria was fascinated as well as photophobia and epiphora” (Luria with the aphasias, particularly in the classifi cation 1959, p. 439). V. demonstrated defects of visual of sounds, speech production and in writing asso- perception and oculomotor ataxia, among other ciated with aphasias (Kolb and Whishaw 2009 ). diffi culties. World War II yielded a forum in The aftermath of World War II yielded numer- which medical and psychological scholars would ous case studies by Luria of individuals suffering come to study the brain more intimately than from various forms of aphasia (Bowden 1971 ). before, especially as it relates to traumatic brain For example, Luria discovered that those indi- injuries. “To signify the combination of these two viduals suffering from oral apraxias could manip- enterprises, the ‘neurological’ and ‘psychologi- ulate speech sounds with the utilization of cal,’ the term neuropsychology was coined” observing their own movements of the lips and (Cole et al. 2006 , pp. 157, 158). tongue in a mirror, thus utilizing an “environ- Following the war, Luria’s work was inter- mental component of the reorganized functional rupted when he was asked to leave the Institute of system” (Bowden 1971, p. 413). Those who Neurosurgery due to his Jewish background. study Luria’s work closely recognize his associa- Eventually, he was invited to return. He contin- tion with traumatic aphasia. Luria devoted a book ued his study of neuropsychology in the 1950s at on this topic, outlining case studies of hundreds the Institute of Neurosurgery until his time of of individuals suffering from localized brain inju- death (Cole 2005a). For the fi eld of neuropsy- ries during World War II (Cole et al. 2006 ). “It chology to advance, Luria stated that neuropsy- was during the war and its aftermath that neuro- chologists needed to focus on a more thorough psychology became a full-fl edged science” (Cole understanding of the “neurophysiological mech- et al. 2006 , p. 140). anisms” that were responsible for operating brain One such case was The Man with a Shattered structures (Cole 2005a ). World. A World War II veteran named Saletsky suffered from brain damage from a bullet entering the posterior left hemisphere intersections of Topography, Localization, the parietal, occipital, and temporal cortexes and Functioning of the Brain (Kolb and Whishaw 2009). Luria would come to follow Saletsky’s trials and tribulations and prog- In the early half of the nineteenth century, phre- ress for 26 years. Saletsky never returned to a nology, a then popular philosophy developed by premorbid level of functioning (Kolb and a German physician, claimed that certain brain Whishaw 2009). Stated by Luria, “the damaged areas had localized and specifi c functions (Kolb areas of the cerebral cortex could not be restored. and Whishaw 2009 ). Under the guidance of 184 D. Princiotta and S. Goldstein phrenology, Franz Joseph Gall concluded that One of the major research questions at that the size of the brain areas was a signifi cant indi- time was whether a complex functional system cator as to human behaviors (Kolb and Whishaw could be localized. Luria did not suggest that this 2009 ). Now considered pseudoscience, neurolo- complex system could be grouped and located in gists and physiologists in the latter part of the zones of the cortex either (Luria 1973a ). Rather, nineteenth century continued to develop their he suggested that different areas of the brain must understanding of the brain. be involved in this system and not necessarily Although neurologists and psychiatrists of the neighboring each other nor even located near one 1880s were able to develop functional maps of another (Luria 1973a ). At this point in time, two the cerebral cortex (Cole et al. 2006 ), our under- major camps were developing revolving around standing was greatly enhanced half a century Luria’s work, those favoring a localized view of later in the 1930s with Luria’s assistance. “The the brain and those favoring an integrated discovery that a complex form of mental activity system. can be regarded as the function of a local brain The terms “function” and “localization” have area aroused unprecedented enthusiasm in neuro- come under great scrutiny as a result of Luria’s logical science. Within a short time many other advancements in the fi eld of neuropsychology. brain centers for intellectual functions were While neurologists may argue one point about found…” (Cole et al. 2006, p. 210). Our current function and localization, a neuropsychologist conceptualization of the topography of the brain may argue another. In the words of Luria himself: as well as localization of function was made pos- “Supporters of the fi rst (narrow localization) sible by Luria’s signifi cant contribution. approach started with the viewpoint that both Luria elaborated, “To say that the human brain elementary and higher (mental) functions must operates as a whole means to make simultane- be viewed as an immediate function of narrowly ously a correct and an erroneous statement. It is limited parts of the brain; therefore they found it correct because the most complex forms of possible to speak of zones in which such phe- human actions require the participation of all nomena as motor and sensory images of words, brain systems; it is erroneous because we can the function of writing or counting, are localized , hardly admit that the Human Brain—this highest and considered that the loss of these functions is point of Evolution—works as an undifferentiated an unequivocal symptom of damage in a corre- whole and that the quality of its work depends sponding zone of the brain cortex. Supporters of exclusively on the active mass of its excited tis- the second (anti-localization) approach, out- sue” (Luria 1969b , p. 9). The complexity of this wardly beginning from the opposite conception, statement would fuel much of Luria’s work in the in fact share the principal position of their oppo- 1930s and beyond. nents” (Luria 1964 , p. 5). Multiple concepts existed in the 1950s related In further illustrating his tenets of localization to brain functioning—the most popular entailed and functioning, Luria equated the meaning of the matching a specifi c mental function with a spe- term function with that of bodily organs, includ- cifi c cortical area or matching all brain functions ing the function of the liver or pancreas, further to all behaviors (Cole et al. 2006). Luria preferred arguing that the function of a particular brain area to focus on a complete functional system rather cannot be likened to the functioning of an organ than an isolated function (Luria 1973a ). A com- (Luria 1973a ). To think that most mental pro- plete functional system allows for the conceptu- cesses could be using “isolated or even indivisible alization of multiple components working faculties, which can be presumed to be the direct together. “This means that all these apparently so function of limited cell groups or to be localized widely different functions incorporate a common in particular brain areas” cannot be true (Luria factor, and it allows an approach to be made to 1973b , p. 29). “Furthermore , these functional sys- the more intimate analysis of the structure of psy- tems are not in ‘narrow zones of the cortex’ but chological processes” (Luria 1973b , p. 42). must be located throughout the brain, regardless 13 A.R. Luria and Intelligence Defi ned as a Neuropsychological Construct 185 of location from another team member of the surface brain structures and deeper brain structures functional system” (Luria 1973b , p. 31). (Cole et al. 2006 , p. 159). A discovery that the brain Luria believed that one could not reduce this structures responsible for cortical tone are actually question to a simple concept in which a distur- housed in the subcortex and brain stem (i.e., the bance of a mental function is the direct result of structures infl uence tone and are regulated by the the destruction of a specifi c part of the brain— structures) radically altered the trajectory of neuro- thus providing confi rmation that the function is psychology (Luria 1973a ). localized in the damaged portion of the brain (Luria 1973b , p. 34). “It should be apparent that if the operation of intellectual processes is Higher Mental Functions thought of in terms of functional systems instead of discrete abilities, we have to reorient our ideas Over the course of his impressive career, Luria about the possibility of localizing intellectual was engaged in diverse case studies of individuals functions … our solution has been to think of the with extraordinary abilities and defi cits alike. functional system as a working constellation of While we have briefl y touched upon two of activities with a corresponding working constella- Luria’s famous case studies demarcating defi cits, tion of zones of the brain that support the activi- the case of S. chronicles a case of greater faculties” (Cole et al. 2006 , p. 141). This might be one ties. In the Mind of a Mnemonist ( 1968 ), Luria of the biggest distinguishing features between the described one of his most famous case examples, human and animal brain (Luria 1973b , p. 31). The S., a man with fascinating abilities in the area of meaning of localized brain damage as it relates to memory. S. fi rst came to Luria after S.’ employer these higher mental functions began for Luria noted that he never took notes during meetings through the Vygotsky Circle (Cole et al. 2006 ). (Kolb and Whishaw 2009 ). Remarkably, S. was Luria did not make the mistake of attempting able to visualize stimuli mentally and recall stimuli to map out the precise locations with specifi c by reading from an internal “photocopy” of the higher cognitions occurring when he explained original (Kolb and Whishaw 2009 ). Interestingly, his three functional units of the brain. To do so S. also met diagnostic criteria for synesthesia, an would suggest that parts of the brain were func- ability to perceive a stimulus of one sense as the tioning independently. In other words, one could sensation of a different sense (e.g., tasting colors) not select a cognitive task and assign the task as (Kolb and Whishaw 2009 ). What allowed S. to only relying on one type of processing or ability create an internal photocopy of the original? How (Luria 1973a ; S. Goldstein, personal communica- do we begin to understand higher mental function, July 13, 2013). In Luria’s words, “…perceptions like S.’ extraordinary memory? tion of memorizing gnosis, and praxis, speech In 1966, Luria described stages of functions and thinking, writing, reading and arithmetic that were imperative for the development of intel- cannot be regarded as isolated or even indivisible ligence and executive functions (S. Goldstein, faculties…” (Luria 1973b , p. 29). personal communication, July 13, 2013; Luria is perhaps most remembered for his teach- Vygotsky, Veer, & Valsiner 1994 ). Only a few ings in the organization of functional brain systems years later, the systemic-dynamic approach of including energizing, coding, and planning and the brain organization of “higher mental functions” cultural contribution of the environment (Cole was refi ned through the work of the Vygotsky 2005a)—the key point aimed at the methodologi- Circle (Kotik-Friedgut 2006 ). Luria described cal conjunction of both theory and practice (Cole interconnected levels that assisted in explaining 2005a, p. 40). The essential tenets of the functional the relationship between the brain and behavior structure and brain organization, or higher mental and neurocognitive disorders. These levels functions, began with Vygotsky and continued included the structures of the brain, the functional with Luria (Akhutina and Pylaeve 2011 ). Rather organization (based on structure), syndromes than taking a “horizontal” viewpoint, Luria focused and impairments arising in brain disorders, and our attention to “vertical” perspectives of both the clinical methods of assessment (Kaczmarek 186 D. Princiotta and S. Goldstein

1999; S. Goldstein, personal communication, Providing further indication for these three July 13, 2013). units and their associated neuropsychological Recall the cultural-historical theory in which abilities and location in the brain, Luria stated Vygotsky and Luria proposed that environmen- that the brain stem, diencephalon, and medial tal interactions were particularly salient in the regions of the cortex constitute the fi rst func- brain and behavior. Expanding on Vygotsky’s tional unit (attention-arousal system) (Luria work, Luria suspected that higher mental func- 1973a ). More specifi cally, Luria was describing tions (e.g., abstraction, memory, and attention), a the midbrain, medulla, thalamus, and hypothala- term that will be repeated frequently in this mus. Working together, these organs maintain chapter, have an origin in social matters (e.g., appropriate cortical tone (Luria 1973a ). “Recent language and thought processes) and are “com- formulations of these regions suggest that some plex and hierarchical in structure” (Luria 1973b , structures at the level of the diencephalon and p. 31; Kotik-Friedgut 2006; Goldstein, personal medial regions have reciprocal connections to the communication, July 13, 2013). This statement cortex through a variety of subcortical circuitries, supports a theory in which human conscious potentially infl uencing a wide range of behav- activity is developed with the utilization of iors” (Koxiol 2009 ). external aids or tools (Kotik-Friedgut 2006 ). The occipital, parietal, and temporal lobes are Furthermore, “At the same time, the process of associated with the second unit in the medial internalization in the development of higher temporal portions positioned posterior to the mental functions takes place under the infl uence central sulcus (Semrud-Clikeman, & Fine 2008 ; of a specifi c cultural context, thus shaping and Goldstein, personal communication, July 13, moderating the process of development and the 2013). The predominant role of the second unit functioning of these basic cognitive abilities” is to process and retain external information (Kotik-Friedgut 2006, p. 43). As demonstrated in such as sensory reception and integration. Here the case study of S., the optimal outcome is effi - the sensory modality does correspond to a par- ciency in cortical functioning in the areas of ticular part of the brain (e.g., auditory stimuli to attention, memory, intelligence, executive func- the temporal lobe) (Luria 1980 ). The second unit tion, and language. was under the guidance of three guidelines pro- Luria’s particularly salient points are included posed by Luria: fi rst, the makeup of cortical in his books, Higher Cortical Functions in Man zones does not indeed remain the same during (1980) and The Working Brain (1973b). Luria development; second, the specifi city of cortical ( 1973b , p. 43) proposed that the brain is com- zone functioning decreases with development; prised of three functional units, each of which are and, third, with development, an increase in lat- “hierarchical in structure and consist of at least eralization occurs (Luria 1980 ). three cortical zones built one above the other.” In The third functioning unit is responsible for other words, each “unit” is further divided into the regulation and evaluation of behavior, includ- three cortical zones, arranged vertically. Coined ing self-monitoring (Luria 1980 ). The third func- primary, secondary, and tertiary cortexes, Luria tional unit, or the prefrontal area of the frontal described the cortical areas as working together. lobes, “synthesizes the information about the As sensory information travels from the fi rst to outside world … and is the means whereby the second to third zones, information in the tertiary behavior of the organism is regulated in confor- zone may then be processed in the amygdala or mity with the effect produced by its actions” the paralimbic cortex for memory and emotional (Luria 1980, p. 263). This unit manipulates the processing (Kolb and Whishaw 2009). In The most complex components of human behavior Working Brain, Luria described a complex func- such as consciousness, personality, voluntary tional system in which these three contributing activity, conscious impulse control, and linguistic units worked in constellation to “create” higher skills (Das 1980 ; S. Goldstein, personal commu- mental activities (Luria 1973a ). nication, July 13, 2013). Here we would expect to 13 A.R. Luria and Intelligence Defi ned as a Neuropsychological Construct 187

fi nd the unit responsible for planning and execu- zone is primarily responsible for units utilizing tive functioning, namely, (McCloskey, Perkins, neurons to receive impulses from sensory organs. & Van Divner 2009 ; S. Goldstein, personal com- Put differently, the primary zone in the motor munication, July 13, 2013). Of the three units, strip of the frontal lobe focuses on motor output the fi rst and the third are the most related to one (Kolb and Whishaw 2009 ; S. Goldstein, personal another; however, the second and the third unit communication, July 13, 2013). The second zone rely on the fi rst condition for collaboration has neurons employed to enable incoming excita- efforts (Luria 1973a ). tion to be moved along to the tertiary zones (Luria In other words, a person is permitted or able 1973a ). This zone sequences motor activity and through these processes along with already speech production, for example. This tertiary acquired knowledge to navigate their fl uid envi- zone is then responsible for the organization and ronments (Luria 1973a; S. Goldstein, personal integration of excitation received by sensory communication, July 13, 2013). structures and rearranging stimuli into a linear Luria premised his hierarchical model on order (Luria 1973a ). Damage to the frontal three main points: regions is particularly troublesome in altering the 1. Information is processed serially, one step at a behaviors and executive functions of individuals, time, by the brain. Information thus travels due to its infl uence and connections with other from sensory receptors → thalamus → primary cortical and subcortical areas of the brain includ- cortex → secondary cortex → tertiary cortex ing the thalamic, hypothalamic, and limbic sys- (Kolb and Whishaw 2009 ). tems (Kolb and Whishaw 2009 ). 2. Serial processing is hierarchical, with Luria would focus a large portion of his career to increased complexity at each level. understand the contribution made by units and 3. “Our perceptions of the world are unifi ed and zones of the brain to understand which mental activ- coherent entities” (Kolb and Whishaw 2009 , ities are encompassed under the umbrella of which p. 267). brain systems (Luria 1973b, p. 103). This is high- Stated by Luria, these three units represent (1) lighted by one of Luria’s salient quotes: “Finally I muscle tone/walking, (2) processing/storing discussed the chief sources of our knowledge of the external information, and (3) programming/regu- cerebral basis of mental activity and I showed that lating/verifying mental activity (Luria 1973a , of these three sources—the comparative anatomy of p. 43). Luria argued that every form of “con- the brain, methods of stimulation and methods of scious” activity was an output of the combined destruction of its individual areas—with respect to efforts of these three functional units working in the analysis of the functional organization of the unison (Luria 1973a ). Luria was very fascinated human brain it is the last which is evidently the most by studying the interactions and specifi c contribu- important” (Luria 1973b , p. 103). tions of these three functional units (Luria 1973a ). Subsequently, Daniel Felleman and David Van Essen would further refi ne this hierarchical model Disturbance of Higher Mental to include hierarchically organized cortical areas Functions but with more than “one area occupying specifi c position relative to other areas, but with more than Thus far, we have focused much of our effort one area allowed to occupy a given hierarchical toward describing the “function” from Luria’s level” (Kolb and Whishaw 2009 , p. 267). They perspective. It is appropriate to now turn to dys- coined this model a “distributed hierarchical sys- function from Luria’s perspective. As Luria has tem” in which the number of areas expands as we stated, “In order to learn more about human cog- go up the hierarchy (Kolb and Whishaw 2009 ). nitive functions we must study both their unfold- Beyond functional units, Luria provided an ing and disruption” (Kaczmarek 1999 ). explanation for subdivisions he called zones One cannot assume that studying brain (Luria 1973a ). These zones are known as the pri- lesions will provide comprehensive, generaliz- mary, secondary, and tertiary zones. The primary able answers. Inherently, brain lesions destroy 188 D. Princiotta and S. Goldstein various areas of the brain, “not just one narrowly lobes than other animals, maturing between 4 and localized group of nerve cells” (Luria 1973b , 7 years of age in humans (Luria 1973a , p. 187). p. 104). “The initial hopes for using disturbance The distinction between the frontal lobes of pri- of higher mental functions for local diagnosis of mates and non-primates lies within voluntary brain damage began to appear unfounded, and attention. After all, “Man lives in a constantly the possibility of using psychological symptoms changing environment, and these changes, which for local diagnosis became a controversial issue” are sometimes unexpected by the individual, (Luria 1964 , p. 4). Expounded by Luria, the brain require a certain level of increased alertness” is now functioning under “pathologically changed (Luria 1973a , p. 55). A disruption to the frontal conditions,” meaning that some elements of the lobes may lead to “substantial disturbances in brain may be completely destroyed while others the fl ow of intellectual processes” (Luria and function as normal—making it very diffi cult to Tsvetkova 1964 , p. 97). “These patients do not study the brain of those under “pathologically make a programme, there is no consequent real- changed conditions” and apply this to the neuro- ization of an original plan, and actions take on the typical population or even to other individuals character of change trials, which easily fall under with brain injuries (Luria 1973b , p. 104). the infl uence of immediate impressions or perse- Furthermore, “Nothing can be more mistaken verations. If the results achieved do not match than such an idea and such an attempt to localize with the original intention, the mistakes made are the symptom of apraxia (and consequently the not recognized and not corrected” (Luria and function of praxis) in a narrow area of the cortex” Tsvetkova 1964 , p. 107). (Luria 1973b , p. 35). And, “As we have seen, a Along with aforementioned case studies local brain lesion does not lead to the direct loss derived from the aftermath of World War II, Luria of a particular mental condition; this was the and colleagues investigated a patient with a verifi ed view held by the supporters of narrow localiza- arachnoidal endothelioma (meningioma) at the tions” (Luria 1973b , p. 103). Burdenko Institute of Neuropsychology, patient Luria believed that any sustained damage to a Zav in 1962 (Luria et al. 1964 ). With complaints pinpointed cortical area in earlier childhood of forgetfulness, headaches, and nausea, Zav had would result in “a relatively elementary basis of diffi culties following through with instructions mental activity, and it unavoidably causes a sec- during her neuropsychological evaluation. Once ondary ‘systemic’ effect, or an underdevelop- language was more complex, her speech behav- ment of higher structures built on these ior was replaced by perseveration that disrupted elementary functions” (Luria 1973b ; Kotik- her when attempting to answer questions (Luria Friedgut 2006, p. 48). et al. 1964 ). Luria and colleagues provided a However, in adults, Luria argued, damage to wealth of information regarding Zav’s progression “higher zones” produces just the opposite through extensive neuropsychological examina- effect—elementary functions would deteriorate tion. “Such deviation from the assigned instructions depending on the secondary or higher forms of and the substitution of a simple response was activities (Luria 1973a ; Kotik-Friedgut 2006 , shown with simultaneous complete retention of p. 48). However, he did not suggest that higher the instruction in the patient’s verbal system. mental functions are “built up as a second story As with the other tasks, the process of relating over elementary processes. Rather, higher mental her actions to the instruction was hampered. functions are the product of the marriage of ele- Recognition of errors was almost impossible” mentary functions into the new system under new (Luria et al. 1964 , p. 267). guidance” (Luria 1973b , p. 43). Overall, much of Zav’s ability to carry out One distinction between humans and animals actions was dependent upon the complexity of applies to the frontal lobes of the brain—the last the instructions (Luria et al. 1964). “At fi rst features of the cerebral hemispheres to be com- glance one may suppose that they preserve all the pletely formed. Primates possess larger frontal basic functions of the Human Brain. But this is 13 A.R. Luria and Intelligence Defi ned as a Neuropsychological Construct 189 not the case. Attentive observation shows how processes. Animals do not use means, their deep are the disturbances in the regulation and behavior is not mediated by means, tools, or control of the conscious behavior of these signs, and it is all unmediated, natural patients” (Luria 1969b , p. 12). Further, “The sug- behavior, whereas human behavior is always gestion, which has been detailed elsewhere, is mediated (tool- or means-using) or indirect that frontal lobe resection limits the subject’s structure, social by origin and voluntary (or ability to organize and reorganize his behavior conscious) by the modes of work. when fl exibility is demanded, and especially 3. This indirect, means-using behavior is medi- when external cues to support such organization ated via speech—the most important system are wanting …the patient with frontal lobe of tools or signs in human history. Language destruction is often unable to fulfi ll instructions, (or speech) has not only its semantic function, is unable to inhibit impulsive reactions or to hold function of categorization of impressions, but back the tendency towards fi xed repetition of its pragmatic or regulatory (or controlling) movement” (Luria et al. 1964 , p. 258). functions as well. By using language, man Luria postulated, “This means that symptoms overcomes the direct infl uences of environ- of disturbance of any higher cortical function ment, and his behavior becomes no more fi eld may be used for local diagnosis of brain damage, linked but is goal or plan linked. but that such a diagnosis can be carried out only 4. The indirect, tool (sign)-using behavior starts a when there is qualitative analysis or evaluation of new form of cortical work: human cortex is no the symptom. Such evaluation of the symptom is more a complex of work of different zones, the fundamental task of neuropsychology” (Luria organized by infl uences of the centrencephalic 1964 , p. 13). system (or natural drives): it becomes a historically organized, plastic functional system where language plays a decisive organizing A Synopsis of Luria in the Final role. That is why higher cortical functions of Decade man have to be evaluated as functional systems of cortical zones, linked by the leading role of Luria corresponded with colleagues throughout the language as a decisive means of behavior. world, questioning their conceptualization and 5. The disorganization of the brain functions fol- understanding of the extra-cortical organization of lowing local brain lesions is in no way a par- brain functions (Kotik-Friedgut 2006 ). Six years tial defi cit (destruction of a spherical local before his death, Luria corresponded with Professor function) and no more a total lowering of the Douglas Bowden in the United States. This cor- general brain activity. It results in a disorgani- respondence, dated February 20, 1971, from zation of functional brain systems, each time Souhanovo, Russia, was included in Dr. Bowden’s resulting from a defect of a basic factor Meta-principles in Luria’s Neuropsychology . (according the locus of the lesion)—bringing Excerpts of this correspondence appear below series of primary symptoms and resulting in a (adopted and transcribed as it appears in Dr. series of secondary symptoms or functional Bowden’s work). These excerpts very well summa- (systemic) results. rize Luria’s thoughts in his fi nal years: 6. The basic goal of neuropsychology is neither 1. There are higher cortical (or psychological) a pure description nor a direct reduction to a functions specifi c to human being and not physiological issue but careful psychological existing in animals. These specifi cally qualifi cation of the symptom (i.e., singling human psychological processes (or func- out the underlying factor and then a descriptions) derive from social sources, i.e., tool- tion of systemic results of the destruction or using social behavior of man. elimination of this factor), that is, the real 2. The most specifi c feature of these higher pro- way to the neuropsychological diagnostics of cesses is that they are tool- or means-using brain injury. 190 D. Princiotta and S. Goldstein

brain structures as described by Luria are A Post-Luria World described in terms of planning, attention, and simultaneous and successive processes or abili- After laboring in the fi eld of neuropsychology for ties (Luria 1966 , 1973b , 1980 ; Das and Naglieri much of his adult life, Luria passed away on 1997). More popularly known as the PASS August 14, 1977, at the age of 75 from cardiac theory, these processes are based on Luria’s neu- arrest. Luria’s research continued directly and rodevelopmental model of stages of higher indirectly through his pupil named Alfred Ardila maturation (S. Goldstein, personal communica- among others (Cole et al. 2006 ). Continuing tion, July 13, 2013). The CAS has recently been work under the tenets of cross-cultural neuropsy- revised for a second edition. chology, Ardila is one of many scholars perpetu- We previously outlined the three functional ating the work of Alexander Luria (Kotik-Friedgut units, which apply to PASS theory: attention uses 2006 ). A centennial celebration of Luria’s birth the fi rst unit (i.e., midbrain, medulla, thalamus, was not necessary to broadcast his true impact on and hypothalamus), simultaneous and successive the world. His contributions to the fi eld of neuro- the second, and planning the third (Luria 1973b ; psychology are immeasurable. Our current con- Koxiol 2009; S. Goldstein, personal communica- ceptualization of extra-cortical organization of tion, July 13, 2013). Simultaneous processing higher mental functions is one of Luria’s most refers to the integration of stimuli into interrelated notable principles (Kotik-Friedgut 2006 ). groups or a whole (e.g., following multistep Because of Luria’s infl uence, the fi eld of neu- instructions) (Das 1994 ). Successive processing ropsychology has progressed rapidly in recent involves sequencing of stimuli in a serial order years. He shaped the process of learning how we (e.g., decoding unfamiliar words and speech artic- think, learn, and solve problems. “In terms of ulation) (Das 1994 ). In a sense, the roles of suc- assessment techniques, Luria’s methods are qual- cessive and simultaneous processing are reversed. itative and fl exible; he seeks links in functional Planning, associated with the fi rst unit, is a pro- systems, his methods are clinical-theoretical and cess in which an individual evaluates solutions to case oriented” (Cole 2005a , p. 35). By contrast, problems. Attention, also associated with the fi rst North American neuropsychologists rely on psy- function, entails a process in which an individual chometric, actuarial, quantitative, group studies” focuses on particular stimuli (Das 1994 ). (Cole 2005a , p. 35). Furthermore, “By the early 1980s, neuropsychology was no longer confi ned to a few elite laboratories, and the new fi eld of Concluding Remarks clinical neuropsychology blossomed in the clinics and hospitals. Since that time, three factors Alexander Luria was active in the fi eld of neuro- have enhanced the rate of change in neuropsy- psychology up to his passing. His work would chological assessment: functional imaging, not be forgotten. His unfi nished publications cognitive neuroscience, and managed health- would later be published by colleagues as care” (Kolb and Whishaw 2009 , p. 806). Paradoxes of Memory (Cole et al. 2006 ). One of the many assessment tools in the fi eld “Depicted by four approaches, Luria outlined of neuropsychology today founded on Luria’s the derivation of human characteristics, human theories is the Cognitive Assessment System consciousness, psychology as a biological sci- (Naglieri & Das, 1997). The CAS was developed ence, and psychology as a unique science” (Cole by J.P. Das and Jack Naglieri. The CAS attempts 2005a, p. 37). In perhaps his last published arti- to measure the functional units described by cle before his death, Luria continued his craft in Luria. Employed with children aged 5–17, the both identifying and distinguishing psychologi- CAS tracks children as they mature, measuring cal approaches to better confi rm the tenets of his their executive functioning through the years own philosophies (Cole 2005a , p. 35). One of the (Das 1980 ; S. Goldstein, personal communica- most poignant statements made by Luria on this tion, July 13, 2013). The functional dimensions of subject was, “As I have said already, any human 13 A.R. Luria and Intelligence Defi ned as a Neuropsychological Construct 191 mental activity is a complex functional system References effected through a combination of concertedly working brain structures, each of which makes Akhutina T.V., & Pylaeva N.M. (2011). L.Vygotsky, its own contribution to the functional system as a A. Luria and Developmental Neuropsychology. whole” (Luria 1973b , p. 38). Psychology in Russia: State of the Art(4), 155–175. In the famous words of Luria before his death Bowden, D. M. (1971). Keystone to Luria’s neuropsychology. Skolepsykologi, 8 (6), 409–416. in 1977: “In this respect neuropsychology is Cole, M. (2005). A. R. Luria and the cultural-historical merely the most complex and newest chapter of approach in psychology. In T. Akhutina, J. Glozman, neurology, and without this chapter, modern clin- L. Moskovich, & D. Robbins (Eds.), A. R. Luria and ical neurology will be unable to exist and develop. contemporary psychology: Festschrift celebrating the centennial of the birth of A. R. Luria (pp. 35–41). New Neuropsychology now has a fi rm foothold in York: Nova Science Publishers. clinical neurology and neurosurgery. This fact is Cole (2005b) Alexander Luria. Retrieved from http:// a source of great satisfaction to the author who, luria.ucsd.edu/index.html together with his colleagues, has spent a good Cole, M., Levitin, K., & Luria, A. (2006). The autobiography of Alexander Luria: A dialogue with the making of deal of his life in an effort to make neuropsychol- mind . Mahwah: Lawrence Erlbaum Associates. ogy an important practical branch of neurology. Das, J. P. (1980). Planning: Theoretical considerations It gives him confi dence that his scientifi c life has and empirical evidence. Psychological Research (W. not been spent in vain, and that new and impor- Germany), 41, 141–151. Das, J. P. (1994). Serial and parallel processing. tant prospects lie ahead for neurology, and for Encyclopedia of Human Intelligence, 964–966. those important divisions of neurology—the top- Kaczmarek, B. L. J. (1999). Extension of Luria’s psycho- ical diagnosis of local brain lesions and the reha- linguistic studies in Poland. Neuropsychology Review, bilitation of these patients” (Luria 1973a , p. 344). 9 (2), 79–87. Kolb, B., & Whishaw, I. Q. (2009). Fundamentals of neu- Those familiar with Luria’s work would likely ropsychology . New York: Worth Publishers. express immense gratitude for his contributions Kotik-Friedgut, B. (2006). Development of the Lurian and assure him that his scientifi c life was surely approach: A cultural neurolinguistic perspective. not spent in vain. Close disciples of Luria’s work Neuropsychology Review, 16 , 43–52. Koxiol, L.F., & Budding D.E. (2009). Subcortical struc- cannot begin to measure the salience of Luria’s tures and cognition: Implications for neuropsycho- work as it relates to the current times. logical assessment. New York: Springer Books. The modern tenets of neuropsychology are Luria, A. (1932a). The nature of human confl icts . Reprint. built upon Luria’s teachings, much as a religion’s New York: Liverlight, 1976. Luria, A. R. (1932b). The nature of human confl icts—or philosophies are built upon a book of God, such emotion, confl ict, and will: An objective study of disor- as the Bible. As Luria discussed the ever-chang- ganisation and control of human behaviour . ing climate of an individual’s environment, so New York: Liveright Publishers. too, the climate of neuropsychology will con- Luria, A. R. (1959). Disorders of ‘simultaneous perception’ in a case of bilateral occipito-parietal brain injury. tinue to evolve. Having said that, the basic tenets Brain, 82 (3), 437–449. of neuropsychology as provided by Luria will Luria, A. R. (1963). Restoration of function after brain prevail in secure establishment as the foundation injury . Oxford: Pergamon Press. of neuropsychology. As stated by Luria a few Luria, A. R. (1964). Factors and forms of aphasia. Reprinted from Ciba Foundation Symposium on years prior to his death, “Thus we are still very Disorders of Language, 1964, pp. 143–161. far from the solution of our basic problem—the Luria, A. R. (1968). The mind of a mnemonist (trans. Neuropsychological Organization of Man’s Solotaroff, L.). New York: Basic Books. Conscious Action, and we can only look forward Luria, A.R. (1969a). The origin and cerebral organization of man’s conscious action: an evening lecture to the with envy and hope to the work of the next XIX International Congress of Psychology. London. generations of Psychologists who will one day Luria, A. R. (1969b). Neuropsychology in the local diag- take our place and bring to a successful end the nosis of brain damage. Reprinted from Cortex, Volume work we have only started” (Luria 1969a , p. 20). 1, 1964, pp. 3–18. 192 D. Princiotta and S. Goldstein

Luria, A. R. (1970). Traumatic aphasia: Its syndromes, psy- Luria, A. R., & Tsvetkova, L. S. (1964). The programming chology, and treatment . The Hague: Mouton de Gruyter. of constructive activity in local brain injuries. Luria, A.R. (1972). The man with a shattered world Neuropsychologia, 2 , 95–107. (L. Solotaroff, Trans. ). New York: Basic Books. Luria, A. R., Pribram, K. H., & Homskaya, E. D. (1964). Luria, A. R. (1973a). The working brain: An introduction An experimental analysis of the behavioral distur- to neuropsychology (trans: Haigh, Basil). London: bance produced by a left frontal arachnoidal end- Penguin Books Ltd. othelioma (meningioma). Neuropsychologia, 2 , Luria, A. R. (1973b). The working brain. New York: Basic 257–280. Books. McCloskey, G., Perkins, L., Van Divner, B. (2009). Luria, A. R. (1976). The cognitive development: Its cul- Assessment and Intervention for Executive Function tural and social foundations . Cambridge, MA: Diffi culties. Routledge, New York. Harvard University Press. Naglieri, J. A., & Das, J. P. (1997). Cognitive Assessment Luria, A. R. (1980). Higher cortical functions in man System. Administration and scoring manual. (2nd ed.). New York: Basic Books. (Original work Interpretive handbook. Itasca, IL: Riverside. published 1966.) Semrud-Clikeman, M. & Fine, J. (2008). Brief and com- Luria, A. R., Cole, M., & Cole, S. (1979). The making of prehensive neuropsychological assessment of children mind: A personal account of Soviet psychology. and adults with learning diffi culties and disabilities. In Cambridge, Mass: Harvard University Press. R.D’Amato (Ed.). Neuropsychology of learning Luria, A. R. (1987). The man with a shattered world: The (pp. 245–291). Mahway, NJ: Lawrence Erlbaum. history of a brain wound. Cambridge, Mass: Harvard Vygotsky, L. S., Veer, R., & Valsiner, J. (1994). The University Press. Vygotsky reader . Oxford, UK: Blackwell. Intelligence: Defi ned as Neurocognitive Processing 14

Tulio M. Otero

and researchers may defi ne intelligence in Introduction dissimilar ways. As a result, their fi ndings may produce confl icting results. Failure to consider Intelligence does not exist. Although this is a the underlying defi nition of intelligence makes strange statement, it is a fundamental truth that accurate interpretation of collected data a tenu- must be understood before understanding what ous process. Pluker and Esping ( 2014 ) present intelligence is and how we defi ne it in this chapter. defi nitions of intelligence by 12 prominent his- Intelligence is not a tangible entity that can be mea- torical and 7 contemporary theorists. Of interest sured in the same way physicists measure matter. is that only one researcher and theoretician, JP Rather, intelligence is a hypothesized “phenome- Das, defi ned intelligence as the “sum of all cog- non.” Its ontology, etiology, and scale are inferred nitive processes” (p. 19). through indirect means. We assume that intelli- Intelligence as used in this chapter refers to a gence is found in the brain, yet we cannot locate it. subset of psychological processes that involve It is not a discrete, embodied force radiating from cognition. Neuropsychological and neuroana- the brain that can be measured directly with sophis- tomical studies have elucidated that neuropsy- ticated apparatus. There is nothing physical to set a chological assets and defi cits in cognitive ruler next to and claim, “this is how much intelli- processes can facilitate or impair specifi c types of gence is here” (e.g., Thorndike 1997 ). learning and performance. An examination of Human intelligence is a psychological con- anatomy also illustrates how and why some neu- struct defi ned by many infl uential philosophers, rocognitive processes are interrelated. Basic psy- researchers, and theoreticians at different points chological processes are important for any in history. Thorndike ( 1997 ) described a con- individual to be able to interact effectively with struct as a defensible collection of separate, the environment, to learn from formal instruction quantifi able qualities and attributes that, when and from experience, and to adapt to new situa- taken together, form a measurable exemplifi cations. Our brain takes in and processes new infor- tion of a multifaceted, hypothesized abstraction. mation in such a manner that typically the right Considering the defi nitions of intelligence as a frontal lobe systems are engaged until new infor- construct is of signifi cant importance. Clinicians mation is learned, assimilated, and accommodated into our existing knowledge base or becomes routinized (Goldberg 2009 ). Once * T. M. Otero , Ph.D. ( ) information or knowledge becomes readily The Chicago School of Professional Psychology , 325 N. Wells St. , Chicago , IL 60654 , USA familiar, greater activation of the left frontal e-mail: [email protected] systems appears to come “online” for handing this

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 193 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_14, © Springer Science+Business Media New York 2015 194 T.M. Otero and is known as the novelty-routinization principle C., Dum, R. P., & Strick, P. L. 2010 , 2013 ; (Goldberg 2009 ). This novelty-routinization Bostan and Strick 2010 ; Koziol et al. 2011 ). principle exemplifi es a dynamic view of the The importance of understanding neurocogni- brain function that provides us with one way of tive processing allows us to understand not only understanding the changing underlying neuro- the dynamic nature of the brain but also allows us anatomical and neuropsychological events that to understand differences in learning and skill occur during learning and skill development development. Technologies such as magnetic (Koziol 2014 ). These hemispheric asymmetries resonance imaging (MRI), functional magnetic are thought to support different styles of informa- resonance imaging (fMRI), positron emission tion processing. tomography (PET), computerized tomography The right hemisphere processes information (CT), and diffusion tensor imaging (DTI) have in a simultaneous holistic manner, whereas the reduced the need for neuropsychological tests to left hemisphere processes information in an ana- localize and access brain damage. Although these lytic, successive manner. This allows for the technologies are favored for investigating the acquisition of harmonizing information about the structural and functional dynamics of the brain, it world. Hence, these two modes of processing is these authors’ opinion that the understanding provide more robust information, different than and assessment of neurocognitive processes by what would be obtained from one type of pro- studying patterns of neurocognitive strengths and cessing in isolation. The brain is far more com- weaknesses in developmental, psychiatric, psy- plex than this simple dichotomy, however. The chosocial, and learning disorders are best great neuropsychologist Alexander Luria, in sev- achieved through formal assessment procedures. eral of his writings (Luria 1966 , 1980 , 1982 ), By addressing both brain functions and environ- maintained that the brain is complex and that no mental factors intrinsic in complex behaviors, part of it functions without the cooperation of such as thinking, reasoning, planning, and the other parts. Thus, Luria viewed the brain as a variety of executive capacities, clinicians are able functional mosaic, meaning that various parts to offer needed services to children with a variety interact in different combinations to apply vary- of learning, psychiatric, and developmental dis- ing combinations of cognitive processing abili- orders. Several neuropsychological tests play an ties (Luria 1973 ). Thus, Luria contended that important role in identifying the neurocognitive there is no area of the brain that functions without processes, or abilities, necessary for effective input from other areas. Integration of processing thinking, learning, and behaving, while also abilities is a key principle of brain function within allowing for judgments regarding the integrity of the Lurian framework. the brain. Psychologists of different specialties During the past decade, ideas about the func- may use standardized instruments to collect tional specialization of brain regions have dra- information and derive inferences about brain- matically evolved. According to Johnson and behavior relationships. Traditional neurocogni- colleagues (2005 ) and Friston ( 2002 ), functional tive testing and evaluation takes a cortical-centric specialization can be defi ned as the degree of approach to understanding brain-behavior rela- information processing specifi city of an identi- tionships. Neuropsychological tests can also be fi ed brain region for a particular cognitive ability utilized as one way to assess the integrity of or facet of cognitive operations. However, as cortical-subcortical functional networks such as Luria pointed out, brain regions obviously do not the fronto-striatal system, among others (Koziol function in isolation. The functional architecture 2009). Consistent with current functional con- of the brain is characterized by reciprocal connec- ceptualizations of the brain, we believe both cor- tional brain profi les of the cerebro-cortical, corti- tical and subcortical networks are important for cal-basal ganglia, cerebro-cerebellar, and basal basic neuropsychological processes to manifest ganglia-cerebellar circuitry systems (Bostan, A. effi ciently. 14 Intelligence: Defi ned as Neurocognitive Processing 195

A conceptualization of human cognitive of fl exible and interactive subcomponents that functioning like the one described by A. R. Luria are mediated by equally fl exible, interactive, can guide the development of assessment tools. neural networks (Luria 1962 , 1980 ). These cog- Such tools should not only evaluate the underly- nitive functions are theorized as three separate ing neurocognitive processes necessary for effi - but connected “functional units” that provide cient thinking and behavior but also provide for four basic psychological processes. The three the development of effective interventions and brain systems are referred to as “functional” units address the question of prognosis. because the neuropsychological mechanisms work in separate but interrelated systems. In other words, multiple brain systems mediate Neuropsychological Theory complex cognitive functions. For example, mul- and PASS Processes tiple brain regions interact to mediate attentional processes (Koziol, Joyce, & Wurglitz, 2014 ). The Luria’s theoretical account of dynamic brain executive functions subserved by the third func- function is perhaps one of the most complete tional unit, as described by Luria, regulate the (Lewandowski and Scott 2008) and current attentional processes of the fi rst functional unit in research in the area of large neuro-networks con- sustaining the appropriate level of arousal and tinue to lend support to his original observations vigilance necessary for the detection and selec- (Koziol 2014 ). Luria conceptualized four unifi ed tion of relevant details from the environment. levels of brain-behavior relationships and neuro- Consider the case of response inhibition; the cognitive disorders that the clinician needs to executive function of inhibition allows a student know: the structure of the brain, the functional to resist or inhibit responding to salient, but irrel- organization based on structure, syndromes and evant, details on a task. Response inhibition impairments arising in brain disorders, and clini- allows the student to sustain focus, over time, on cal methods of assessment (Korkman 1999 ). His task relevant features. theoretical formulations, methods, and ideas are The brain systems described above are con- articulated in works such as Higher Cortical sistent with the four psychological processes Functions of Man (1966 , 1980 ) and The Working identifi ed by the PASS theory (Naglieri and Das Brain (1973 ). Luria viewed the brain as a func- 1997b ), and this amalgamation of processing tional medley, the parts of which interact in dif- abilities is a key principle of brain function within ferent combinations to subserve cognitive the Lurian framework. Cognition and behavior processing (Luria 1973 ). Cognition and behavior result from an interaction of complex brain activ- then result from an interaction of complex brain ity across various areas. Naglieri and Das (1997a , activity across various areas. Luria’s (1966 , 1973 , b) used Luria’s work as a base to redefi ne intelli- 1980) research on the functional aspects of brain gence from a multi-ability perspective. The PASS structures formed the basis for the development theory has strong empirical support (see Das of the PASS theory (planning, attention, simulta- et al. 1979 , 1994b) and since the publication of neous, successive processing), initially described the Cognitive Assessment System (CAS) and by Das et al. ( 1994) and operationalized by Cognitive Assessment System-Second Edition Naglieri and Das (1997a , b ) in the Cognitive (CAS-2) (see Naglieri 2012 ; Naglieri and Assessment System (CAS), and most recently in Conway 2009 ; Naglieri and Otero 2011 ). the Cognitive Assessment System-Second Luria ( 1973) stated “each form of conscious Edition (Naglieri et al. 2014a , b ). activity is always a complex functional system From a Lurian framework, cognitive func- and takes place through the combined working of tions, such as attention, executive functions, lan- all three brain units, each of which makes its own guage, sensory perception, motor function, contribution” (p. 99). In other words, the four visuospatial facilities, and learning and memory, processes form a “working constellation” (Luria are multifaceted capacities. They are composed 1966 , p. 70) of cognitive activity. Thus, a child or 196 T.M. Otero adult can use different combinations of the four suggest some structures at the level of dience- psychological processes in conjunction with their phalic and medial regions have reciprocal knowledge and skills to perform a task. Although connections to the cortex through a variety of effective functioning is achieved through the large-scale brain circuitries (Koziol and Stevens appropriate combination of all processes as 2012) potentially infl uencing a wide range of demanded by the task, each process is not equally behaviors (Koziol 2009 ). involved in every task. For example, reading comprehension may predominately involve one process, while reading decoding can be strongly First Functional Unit dominated by another (Das et al. 1994b ). Or basic math calculation may require more of one Attention is a basic component of intelligent process, while math-reasoning tasks may require behavior involving allocation of resources and a different cognitive process. For example, learn- effort. Arousal, attention, effort, and capacity are ing basic math calculation operations may ini- concepts that have a complex relationship and tially require a basic step-by-step approach. importance for understanding behavior. When a Understanding math-reasoning problems, how- person is required to pay attention to only one ever, requires holding multiple elements of the dimension of a multidimensional stimulus array, task in memory, surveying the elements, and the inhibition of responding to other (often more making decisions about these elements before salient) stimuli and the allocation of attention to solving the problem. Das and Naglieri and their the central dimension are required. Luria stated colleagues used Luria’s work as a blueprint for that optimal conditions of arousal are needed defi ning the basic neuropsychological processes before the more complex forms of attention that underlie human performance (Naglieri involving “selective recognition of a particular 2003 ). Their efforts represent the fi rst time that a stimulus and inhibition of responses to irrelevant specifi c researched neuropsychological theory stimuli” (Luria 1973 , p. 271) can occur. This way was used to provide an alternative conceptualiza- of conceptualizing attention is analogous to such tion of human intelligence. contemporary models as Mirsky and Duncan’s ( 2001 , 2003 ; Koziol 2014 ) in which focus, shift , sustenance, and stabilization of attention are nec- Three Functional Units Described essary before complex learning can take place. Moreover, the second and third functional units Luria (1973 ) provided considerable evidence for can operate effectively only after individuals are the neuropsychological processes associated suffi ciently aroused and their attention is ade- with each of the three functional units and their quately focused. association with specifi c regions of the brain. Briefl y stated, these three functional units have been used by Naglieri and Das (1997b ) as the Second Functional Unit basis of planning (third functional unit), attention (fi rst unit), and simultaneous and successive The occipital, parietal, and temporal (particu- (second unit) cognitive processes. The brain larly medial temporal portions) lobes posterior stem, the diencephalon, and the medial regions to the central sulcus of the brain are associated of the cortex are the primary locations for the with the second functional unit. Information fi rst of the three functional units of the brain, the from the external world is received, processed, attention-arousal system, (Luria 1973). This unit and retained within this unit. Thus, the major is comprised specifi cally of the midbrain, medulla, function of the second functional unit is sensory thalamus, and hypothalamus. These structures reception and integration (Semrud-Clikeman work in concert to maintain the appropriate cor- and Teeter Ellison 2009 ). The areas of the sec- tical tone. Recent formulations of these regions ond functional unit correspond to their sensory 14 Intelligence: Defi ned as Neurocognitive Processing 197 modality: temporal for auditory stimuli, parietal structures forming the second functional unit can for tactile, and occipital for visual. This func- result in diffi culty across all academic areas. tional unit is hypothesized to be guided by three Tertiary zones process and integrate informa- functional laws: (1) during development, the tion from all sensory areas. This integration of makeup of cortical zones does not remain the information from various modalities occurs same; (2) the specifi city of function of the corti- through simultaneous processing. For example, cal zones decreases with development; and (3) some math involves the integration of both visual an increase in lateralization of function increases materials and knowledge of number quantity. with development (Luria 1980 ). This hierarchy Math reasoning, in the form of word problems, is further subdivided into zones. The primary may additionally involve the integration of gram- zones of these structural units employ modality- matical skills, analysis of auditory information, specifi c groups of neurons to receive impulses and the comprehension of auditory or written from the sensory organs, while the secondary material. Damage to these zones has been related zones of these structures surround the primary to lower measured intelligence and diffi culties zones with associative neurons which enable across several basic academic areas. incoming excitation to be conveyed to the ter- Simultaneous and successive processing are tiary zones. These tertiary zones are responsible subserved by the second functional unit. for integrating and organizing the excitation Simultaneous processing is a mental activity by arriving from the different sensory structures and which a person integrates stimuli into interre- converting the stimuli which are received in a lated groups or a whole. For example, in order specifi c linear order into simultaneously pro- to follow multistep directions, the relationships cessed groups (Luria 1974). The secondary among the different parts of what is said must be zones are involved in the input of data and inte- correctly understood. Reading unfamiliar words gration of information. These zones process that are initially diffi cult to decode and then are information sequentially and connect cross- later quickly and effortlessly recognized as a modally with several stimuli impinging on the whole word is another example of a task initially brain at a time. For example, reading is an inte- demanding successive processing, followed by gration of both visual and auditory material, and the effi cient simultaneous processes of reading mathematics is the integration of visual material the word as a whole unit. Children presenting with the knowledge of numbers and quantity. with diffi culties performing on tasks that require There are several secondary zones for auditory, learning new information by associating it with tactile, and visual information. The auditory sec- other information may have defi cits in simultane- ondary zone lies within the secondary regions ous processing. Diffi culty integrating visual within the temporal lobes and involves the analy- information may be a primary defi cit in children sis and synthesis of sounds and the sequential with nonverbal learning disabilities. These chil- analysis of phonemes, pitch, tone, and rhythm. dren have been found to do more poorly on The secondary tactile zone is within the parietal measures that demand simultaneous processing lobe and is involved in the recognition of complex such as tests on visual motor integration and tactile stimuli and two-point discrimination, for visual- perceptual skills compared to children example. The secondary visual zone borders the with ADHD and normally developing children primary visual cortex of the occipital lobe. Visual (Wilkinson and Sermund-Clikeman 2008 ). discrimination of letters, shapes, and fi gures are While simultaneous processing involves related to it. Traditional intelligence tests are working with interrelated stimuli, successive pro- hypothesized to measure some aspects of the sec- cessing requires work with stimuli in a specifi c ond functional unit (Semrud-Clikeman and Teeter serial order. This processing ability is required Ellison 2009 ). Because the second functional when a child arranges things in a strictly defi ned unit can be considered as the center for analysis, order, where each element is only related to those coding, and storage of information, damage to the that precede it and these stimuli are not interrelated. 198 T.M. Otero

Successive processing ability involves both the been implicated in higher-order processing such perception of stimuli in sequence and the forma- as attention, decision-making, and intelligence tion of sounds and movements in order. For (Kolb and Whishaw 2009 ). example, successive processing is involved in the Luria stated that “the frontal lobes synthesize decoding of unfamiliar words, production of syn- the information about the outside world … and tactic aspects of language, and speech articula- are the means whereby the behavior of the organ- tion. Following a sequence such as the order of ism is regulated in conformity with the effect pro- operations in a math problem is another example duced by its actions” (Luria 1980 , p. 263). The of successive processing. Most real-life situa- frontal lobes provide for the programming, regu- tions will require both of these processes to lation, and evaluation of behavior and enable a become activated to some degree. person to ask questions, develop strategies, and self-monitor (Luria 1973 ). Other responsibilities of the third functional unit include the regulation Third Functional Unit of voluntary activity, conscious impulse control, and various linguistic skills such as spontaneous One of the most extraordinary capacities we have conversation. The third functional unit provides as humans is our ability to refl ect and self-direct for the most multifaceted aspects of human behavior. This ability is frequently described behavior, including personality and conscious- using the term executive function (EF). Although ness (Das 1980 ). The fi rst and third functional many defi nitions of EF abound (Goldstein et al. units share a reciprocal relationship. The higher 2014), the concept of EF is intimately linked to cortical systems both regulate and work in col- the frontal lobes. The prefrontal areas of the fron- laboration with the fi rst functional unit while also tal lobes of the brain are associated with the third receiving and processing information from the functional unit (Luria 1980 ). The prefrontal cor- external world and determining an individual’s tex is well connected with every distinct func- dynamic activity (Luria 1973 ). It is both infl u- tional unit of the brain (Goldberg 2009 ). This unit enced by the regulatory effects of the cortex and is mostly responsible for output planning and infl uences the tone of the cortex. The ascending with most behaviors we typically consider as and descending systems of the reticular forma- executive functions. The third functional unit is tion enable this relationship by transmitting also further differentiated into three zones, with impulses from the lower parts of the brain to the the primary zone in the motor strip of the frontal cortex and vice versa (Luria 1973). Thus, damage lobe being concerned with motor output. The to the prefrontal area can alter this reciprocal secondary zone is responsible for the sequencing relationship, so that the brain may not be suffi - of motor activity and speech production, while ciently aroused for complex behaviors requiring the tertiary zone is primarily involved with sustained attention. A breakdown in any portion behaviors typically described as executive func- of the complex loop-like interactions between the tions. Damage to any of several areas of the fron- prefrontal, ventral brain stem and posterior cor- tal regions has been related to diffi culties with tex is likely to produce symptoms of attention impulse control, learning from one’s mistakes, defi cit (Goldberg 2009 ). delay of gratifi cation, and attention. Because the The psychological processes that are routed third functional unit has rich connections with in each of the functional units are linked. For the other parts of the brain, both cortical and subcor- PASS theory, this means that psychological pro- tical, there can be forward and backward infl u- cesses of attention and planning are strongly ences, to and from other regions such as the related because planning often has conscious cerebellar, thalamic, hypothalamic, and limbic control of attention. In other words, one’s lim- areas. Additionally, a growing body of evidence ited attentional resources are dictated by the plan points to a network of connected regions in the for one’s behavior. However, attention as well as adjacent frontal and parietal lobes, which have the other PASS processes are infl uenced by 14 Intelligence: Defi ned as Neurocognitive Processing 199 many variables other than planning. One of these Luria believed that a child’s cultural experi- infl uences is the environment. Novel encounters ence is a signifi cant infl uence on the functional within daily life demand we act in one way or units and also a necessary foundation that aids another. Several PASS processes can be involved the development of human cognition (Luria as we make judgments about similarities and dif- 1979 ). The organization of the brain into func- ferences between past situations and the present tional units also accounts for the interaction of demands, while hypothesizing possible out- cultural infl uences and biological factors within comes of our actions and as we select behaviors higher cognition. Luria ( 1979) notes “…the while acting on the environment. child learns to organize his memory and to bring it under voluntary control through the use of the mental tools of his culture” (p. 83). Kolb et al. Functional Units: Interactions (2003 ) also wrote that although “the brain was and Infl uences once seen as a rather static organ, it is now clear that the organization of brain circuitry is con- Luria believed no part of the brain works by stantly changing as a function of experience” itself, and therefore, his organization of the brain (p. 1). Various brain systems are highly modifi - into functional units was not an attempt to map able by experience and dependent on experience out the precise locations where specifi c areas of only during particularly sensitive time periods higher cognition took place, but rather to empha- and other systems remain capable of change by size that no cognitive task solely requires simul- experience throughout life (Neville 2006 ; taneous, successive, planning, or attention Neville and Stevens 2008). Similarly, Vygotsky processing or any other process, but rather, it is a (1976 ) described this interplay when he matter of emphasis. He stated, “…perception and described speech as a self-regulatory function. memorizing, gnosis and praxis, speech and think- Self-talk functions as self-guidance and regula- ing, writing, reading and arithmetic, cannot be tion, helps children think about their mental regarded as isolated or even indivisible ‘facul- activities and behaviors, and select courses of ties’…” (Luria 1973 , p. 29). That is, an attempt to action, and is the foundation for all higher cogni- identify a fi xed cortical locale for any complex tive processes (e.g., controlled attention, deliber- behavior is a mistaken endeavor. Instead, the ate memorization and recall, categorization, brain should be conceptualized as a functioning planning, problem- solving, abstract reasoning, whole comprised of units that provide purpose. self- refl ection). Stuss and Benson (1990 ) Activities such as reading and writing can be described this interplay as follows: evaluated and seen as constellations of activities The adult regulates the child’s behavior by com- related to specifi c working zones of the brain that mand, inhibiting irrelevant responses. His child support them (Luria 1979 ). This means that since learns to speak, the spoken instruction shared the brain operates as an integrated functional sys- between the child and adult are taken over by the child, who uses externally stated and often detailed tem, even a minor disruption in an area can cause instructions to guide his or her own behavior. By disorganization in the entire functional system the age of 4 to 4 ½, a trend towards internal and (Varnhagen and Das 1986 ). Thus, many behav- contract speech (inner speech) gradually appears. iors may be impacted by a disruption caused by a The child begins to regulate and subordinate his behavior according to his speech. Speech, in addi- lesion, damage, or underdeveloped structures. tion to serving communication thought, becomes a For example, lesions or damage to the prefrontal major self-regulatory force, creating systems of cortex, with its complex connections with other connections for organizing active behavior inhibit- areas of the brain including several subcortical ing actions irrelevant to the task at hand. (p. 34) areas, may result in affective dissociations, impaired executive functions, poor judgment and Culture infl uences the development of higher processing, or intellectual defi cits. cognitive functioning through a variety of dif- 200 T.M. Otero ferent channels. Luria (1979 ) emphasized the actions as success or failure relative to our importance of the frontal lobes in language, intentions” (Goldberg 2009 , p. 23). Planning organization, and direction of behavior and speech helps one to achieve goals through the develop- as a cultural tool that furthers the development ment of strategies necessary to accomplish tasks of the anterior brain region and self-regulation. for which a solution is required. Therefore, plan- Cultural experiences accelerate the use of planning is an essential ability to all activities that ning and self-regulation and the other cognitive demand the child or adult to fi gure out how to processes. Luria (1979 ) suggested that abstrac- solve a problem. This includes self-monitoring tion and generalizations are themselves products and impulse control as well as making, assess- of the cultural environment. Children learn, for ment, and implementation of a plan. Thus, plan- example, to selectively pay attention to items that ning allows for the generation of solutions, are pertinent through conversations and playful discriminating use of knowledge and skills, as interactions with adults. Even simultaneous and well as control of attention, simultaneous, and successive processes are infl uenced by cultural successive processes (Das et al. 1996 ). experiences (e.g., learning dances, poems, game The essential dimension of the construct of rules, and so on). Naglieri (2003 ) summarized planning as defi ned by Naglieri and Das ( 1997b ) research that showed that the infl uence of social is very similar to the description of executive interaction on children’s use of plans and strate- function provided by others (see Naglieri and gies resulted in improvements in performance on Goldstein 2006). For example, O’Shanick and academic tasks. Luria’s concept of functional O’Shanick (1994 ) describe executive functions units and their relationship to the larger sociocul- as including the abilities to formulate and set tural context provides the foundation for the goals, assess strengths and weaknesses, plan PASS theory. and/or direct activities, initiate and/or inhibit behavior, monitor current activities, and evaluate results. Executive functions include abilities to From Luria to PASS Theory formulate a goal, to plan, to carry out goal- of Intelligence directed behaviors effectively, and to monitor and self-correct spontaneously and reliably (Lezak The four processes in the PASS theory represent et al. 2012). McCloskey et al. ( 2009) identify a fusion of cognitive and neuropsychological two key dimensions that unify several diverse constructs including executive functioning (plan- defi nitions of executive functions. To some ning); selective, sustained , and shifting attention degree, all defi nitions address components that (attention); visual-spatial tasks (simultaneous); direct and cue other processes, and all address and serial features of language and memory functions that link activation to the frontal lobe (successive) (Naglieri and Das 2005 ). These four regions. These skills are essential for fulfi lling processes are more fully described in the sec- most daily responsibilities and maintaining tions that follow. appropriate social behavior. A variety of assess- The human ability to plan differentiates ment tools that have been proposed to assess humans from other primates. Planning is associ- executive functions often yield confl icting data ated with the prefrontal cortex. The prefrontal given the very broad defi nition of these functions cortex “plays a central role in forming goals and (e.g., for a review of this issue in the assessment objectives and then in devising plans of action of ADHD, see Barkley 2006). Planning in the required to attain these goals. The cognitive pro- PASS theory offers a more fi nite description that cesses required to implement plans, coordinate may be characterized as executive function. these activities, and apply them in a correct order Attention is a cognitive process that is closely are subserved by the prefrontal cortex. Finally, the connected to the orienting response. Attention, as prefrontal cortex is responsible for evaluating our ability, allows a person to demonstrate focused, 14 Intelligence: Defi ned as Neurocognitive Processing 201 selective cognition over time, with resistance (Naglieri 1999). This is most apparent in the to distraction. Attention occurs when a person inclusion of the verbal-spatial relationship sub- selectively focuses on particular stimuli and test in the CAS (Naglieri and Das 1997a ). inhibits responses to competing stimuli. The Typically, however, matrices tests have been process is involved when we must demonstrate included in the so-called nonverbal scales of focused, selective, sustained, and effortful activity. intelligence tests such as the Wechsler Nonverbal Focused attention involves concentration directed Scale of Ability (Wechsler and Naglieri 2006 ), toward a particular activity, and selective atten- the perceptual reasoning portion of the Wechsler tion is important for the inhibition of responses to Intelligence Sale for Children-IV (WISC-IV; distracting stimuli. Sustained attention refers to Wechsler 2003), the Stanford-Binet Fifth Edition the variation of performance over time, which (SB5; Roid 2003 ), the Naglieri Nonverbal Ability can be infl uenced by the varying amounts of Test (NNAT; Naglieri 1997 ), and the Kaufman effort required to solve a task. Brain structures Assessment Battery for Children, Second Edition within Luria’s fi rst functional unit, the reticular (KABC-2; Kaufman and Kaufman 2004 ) and as formation, allow one to focus selective attention a simultaneous processing test (Naglieri and Das toward a stimulus over a period of time without 1997a , b ). the loss of attention to other competing stimuli. Successive processing is relevant when work- The longer attention is needed, the more the ing with stimuli arranged in a defi ned serial order activity necessitates vigilance. Intentions and such as remembering or completing information goals mandated by the planning process control in compliance with a specifi c order. Successive attention, while knowledge and skills play an processing is typically an essential element integral part in the process as well. The attention involved with the serial organization of sounds, work of Schneider et al. ( 1984) and the atten- such as learning sounds in sequence and early tion selectivity work of Posner and Boies (1971 ), reading. Furthermore, successive processing has which relates to deliberate discrimination between been conceptually and experimentally related to stimuli, are similar to the way that the attention the concept of phonological analysis (Das et al. process is conceptualized. Planning processes 1994b ). When serial information is grouped into regulate a variety of other processes, including a pattern, however, (like the number 553669 attention. organized into 55-3-66-9), then successful repeti- Simultaneous processing is necessary for syn- tion of the string may be a function of another thesizing separate elements into a cohesive whole cognitive processes, such as planning (i.e., using or interrelated group. The ability to recognize the strategy of chunking) and simultaneous (orga- patterns as interrelated elements is made possible nizing the numbers into related groups). This by the parieto-occipital-temporal brain regions. method is often used by older children and can be Due to the substantial spatial characteristics of an effective strategy for those who are weak in most simultaneous tasks, there is a visual-spatial successive processing (see Naglieri and Pickering dimension to activities that demand this type of 2003). In clinical practice, we have observed that process. Conceptually, the examination of simul- young children with poor successive processing taneous processing is achieved using tasks that often have diffi culty following directions or com- could be described as involving visual-spatial prehending what is being said to them when sen- reasoning, found in progressive matrices tests tences are too lengthy. Teachers and parents often like those developed by Penrose and Raven misinterpret this weakness as a failure to compre- ( 1936). Simultaneous processing is not, however, hend or as a problem of attention. limited to nonverbal content, as demonstrated by Traditionally, intelligence is measured through the important role it plays in the grammatical verbal, nonverbal, and quantitative tests, yet the components of language and comprehension of PASS theory offers an alternative approach to word relationships, prepositions, and infl ections intelligence. This theory broadens the idea of 202 T.M. Otero what abilities should be measured and also Validity emphasizes the signifi cance of basic neurocognitive processes. Additionally, the functional units Using Luria’s neuropsychological framework of of the brain that encompass the PASS processes three functional units, Das ( 1972) and Das et al. are considered the building blocks of ability con- (1975 , 1979 , 1994b ) began the task of fi guring ceptualized within a neurocognitive processing out methods for measuring the PASS processes. framework. While the theory may have its roots These efforts included extensive analysis of the in neuropsychology, “…its branches are spread methods used by Luria, related procedures used over developmental and educational psychology” within neuropsychology, experimental research (Varnhagen and Das 1986 , p. 130). Thus, with its in cognitive and educational psychology, and connections to developmental and cognitive pro- related areas. Their work was summarized in cessing, the PASS theory offers an advantage in several books by Kirby ( 1984); Kirby and explanatory power over the notion of general Williams (1991 ); Das et al. (1994b ); Naglieri intelligence (Naglieri and Das 2002 ). (1999); and Naglieri et al. (2014a , b ), which provide considerable evidence that the PASS processes associated with Luria’s concept of the Measuring PASS Processes three functional units could be measured and that once measured, these processes have con- The PASS theory was operationalized by the CAS siderable reliability and validity. Their work also (Naglieri and Das 1997a ; Naglieri et al. 2014a, b ). demonstrated that there was signifi cant potential This instrument is thoroughly described in the for the application of the theoretical conceptual- CAS-2 interpretive handbook (Naglieri et al. ization of basic psychological processes. The 2014b ). Naglieri et al. (2014a , b ) generated tests remainder of this section will provide a summary to measure the PASS theory following a system- of relevant validity research on the PASS theory atic and empirically based test development pro- as operationalized by the CAS. gram designed to obtain effi cient measures of the processes for individual administration. The PASS theory was used as the foundation of the Relationship to Achievement CAS, so the content of the test was determined by the theory and not infl uenced by previous views One of the purposes of an ability test is to deter- of ability. The CAS-II (Naglieri et al. 2014a , b ) mine a child’s level of cognitive functioning that is a Lurian-based test of cognitive abilities and can then be used to anticipate performance in a processing that is highly predictive of academic number of contexts, such as school. Some have learning and very useful in identifying process- noted that the relationship between a test of abil- ing strengths and weaknesses. The CAS-II, ity and achievement is perhaps one of the most normed for ages 5–18, is supported by research important aspects of validity (Brody 1992 ; Cohen linking the CAS-II processes with specifi c et al. 1992 ; Naglieri and Bornstein 2003 ). For types of learning (Naglieri and Das 1997a , b ) and many years, researchers have studied the rela- research linking specifi c-processing defi cits with tionship between ability and achievement. Well- specifi c learning disabilities. The four CAS-II known IQ tests often include measures of composites are intended to measure planning, vocabulary, general information, and arithmetic, attention, simultaneous processing, and succes- as do tests of achievement. It is no surprise then sive processing. The CAS-II composites and all that the relationship between ability and intelli- of the subtests can be categorized as measures gence has been found to be about 0.55–0.60 of cognitive processing. The CAS-II does not (Brody 1992 ; Naglieri 1999). It has been argued, contain any measures of verbal knowledge or however, that a portion of the correlation between crystallized intelligence (see Chapter XXX for a traditional IQ tests and academic achievement thorough presentation of the CAS-II). tests is due to the similarity in content that exists 14 Intelligence: Defi ned as Neurocognitive Processing 203 between these two types of tests (Naglieri and Full Scale. These fi ndings suggested that the Bornstein 2003 ; Naglieri and Rojahn 2004 ). four PASS Scales individually and collectively Given that the CAS does not include test items correlate higher with achievement than the four that are typically part of traditional IQ tests such scales aggregated into the one Full Scale score. as vocabulary and arithmetic, how well does it Additionally, the predictive power of the combi- correlate with achievement? nation of the four PASS Scales was weakened Naglieri and Rojahn ( 2004) studied the when any one of the PASS Scales was excluded relationship between the PASS processing scores in the prediction equation (Naglieri and Rojahn of the CAS with the Woodcock Johnson-Revised 2004 ). This suggests that each of the PASS Scales Tests of Achievement (WJ-R; Woodcock and has additive value in predicting achievement and Johnson 1989 ) with a sample of 1,559 students further supports the notion of interrelated neuro- aged 5–17 years. The correlation between the cognitive processes within the Luria’s framework CAS Full Scale and the WJ-R Tests of of functional units. Achievement was 0.71 for the standard (all 12 subtests) and 0.70 for the basic battery score (eight subtests). These fi ndings provide evidence Relationship to Behavior for the construct validity of the CAS and more importantly suggest that basic psychological pro- Limited research has been conducted specifi cally cesses are strongly related to academic perfor- examining the relationship of PASS processes to mance as measured by this standardized test of behavior. Clinically, the connection between achievement. PASS processes and a child’s behavior is often Naglieri et al. (2006 ) compared the Wechsler observed. For example, successive processing Intelligence Scale for Children-Third Edition involving the ability to follow information in a (WISC-III; Wechsler 1991 ) to the CAS and the linear organization or chainlike progression will Woodcock-Johnson III Tests of Achievement exert a signifi cant impact on a child’s behavior. (WJ-III-ACH; Woodcock et al. 2001 ) with a sam- Planning processing involves the ability to focus ple of children aged 6–16 who were referred for one’s thinking, attend and screen out distractions evaluation due to learning problems. The correla- which is essential for children to play effectively tion of the WJ-III-ACH scores with the WISC-III with others on the playground, and interact with Full Scale IQ scores was 0.63 and 0.83 with the adults, as well as a variety of real-life tasks. CAS Full Scale. However, the CAS Full Scale Several researchers have examined the rela- scores correlations were signifi cantly higher tionship between the behavioral diffi culties seen (Naglieri et al. 2006 ). in children with ADHD and PASS profi le scores. The fi ndings provide evidence for the con- For example, Paolitto ( 1999 ) studied matched struct validity of the CAS and suggest that basic samples of ADHD and normal children. Children psychological processes are strongly correlated with ADHD earned signifi cantly lower scores on with academic performance and are especially the planning scale. Similarly, Dehn (2000 ) and important because the measures of the PASS pro- Naglieri et al. ( 2003) found that groups of chil- cesses do not include achievement-like subtests dren who met diagnostic criteria for ADHD (e.g., vocabulary and arithmetic). This provides earned signifi cantly lower mean scores on the considerable advantage and is especially impor- planning scale of the CAS. These results support tant for children who come from disadvantaged the view that ADHD involves problems with environments as well as those who have had a behavioral inhibition and self-control, which is history of academic failure. associated with poor executive control (Planning; Importantly, Naglieri and Rojahn ( 2004 ) also Naglieri and Goldstein 2006). These fi ndings found that prediction of achievement was slightly suggest that the PASS processing theory has util- higher for the four PASS Scales than the CAS ity for differential diagnosis, intervention, as well 204 T.M. Otero as response to intervention for behavioral prob- signifi cantly, while there was a concomitant lems (Naglieri 2003 , 2005). improvement on behavior as measured by rating The measurement of cognitive processes using scales. Mccrea (2009 ) studied three patients with the Cognitive Assessment System (CAS) has unilateral focalized stroke lesions longitudinally been utilized with individuals who suffer from on the CAS subtests at 1 month and 6 months traumatic brain injury (TBI). Due to the fact that post infarct, such that each patient functioned as cognitive impairments and defi cits are very their own baseline. Patient 1 with a left temporal common in individuals with TBI, the CAS is a pole lesion had a severe syntactic comprehension measure that can be used to assess the cognitive defi cit on sentence questions. Patient 2 had a rare processes of this population. According to Luria right anterior cerebral artery (ACA) aneurysm (1973 ), when one suffers severe brain damage, it culminating in an orbitofrontal syndrome and is likely that he or she will also experience impairments on expressive attention, word series, impairments in such processes as organization as well as a praxis-based fi gure ground reversal and planning (as cited in Gutentag et al. 1998 ). phenomenon on fi gure memory. Patient 3 suf- One of the main reasons as to why an assessment fered a right frontoparietal lesion with resulting tool such as the CAS would be particularly useful representational as well as elements of motor for the TBI population is because typical intelli- neglect and impairments on matching numbers, gence tests only yield results that refl ect one’s number detection, and receptive attention. Each general intelligence; they do not provide mea- patient’s lesions were all entirely consistent surement of basic psychological processes. For with the nature of cognitive neuropsychological example, defi cits in attention and planning that symptoms suggesting that the CAS subtests are interfere with the academic performance of chil- unique and also sensitive and specifi c to focal- dren with TBI (Savage and Wolcott 1994 ) must ized cortical lesions. be measured. Gutentag et al. ( 1998 ) studied The characteristics of the US population con- children with TBI showed defi cits on the CAS tinue to change with every census, and the need compared to a matched control group drawn from for fair assessment of children has become pro- the CAS normative population. Neurocognitive gressively more important. Traditional IQ tests defi cits were most pronounced in the attention have items that measure content that is dependent and planning domains and less severe in the on exposure to the dominant culture, language, simultaneous and successive domains (Gutentag and formal education. This content can create an et al. 1998 , p. 265). unfair disadvantage for many children, such as those living in non-English-speaking homes and impoverished environments. Reducing the Fairness amount of knowledge needed to correctly answer the questions on intelligence tests is a useful way The CAS has been increasingly used across cul- to ensure appropriate and fair assessment of tures and languages. In Spain for example, Perez- diverse populations. Some researchers have sug- Alvarez et al. (2006 ) used the CAS in a study gested that conceptualizing intelligence as a set assessing the effects of topiramato (a pharmaco- of psychological processes, such as the PASS logical treatment for epilepsy) on cognitive pro- theory as operationalized by the CAS, has utility cesses and behavior. The 35 patients ranging in for assessment of children from culturally and age from 5 to 15 years were assessed with the linguistically diverse populations because verbal CAS at baseline and again at 6 and 12 months. and quantitative skills are not included (Naglieri The parents were given behavior-rating scales at and Otero 2011 ; Naglieri et al. 2005 , 2007 ). each interval as well. At baseline, 6 and 12 Several researchers have found up to a months, patients had lower successive scores. 15-point mean difference between Blacks and At 12 months, planning scores had increased Whites on traditional tests of cognitive ability. 14 Intelligence: Defi ned as Neurocognitive Processing 205

Results for PASS processing tests have shown However, these tests fail to control for the level of only small differences between these groups. For language profi ciency (Harris and Llorente 2005 ). example, Naglieri et al. (2005 ) compared CAS In the case of ELL Hispanic children, Naglieri scores of a sample composed of 298 Black chil- et al. (2007 ) compared the English and Spanish dren and 1,691 White children. Controlling for versions of the CAS for bilingual Hispanic key demographic variables, regression analyses children. The children in this study earned very showed an estimated CAS Full Scale mean score similar CAS Full Scale scores, and defi cits in difference of 4.8, which is smaller than that found successive processing were found on both versions with traditional tests of ability. Another fi nding of the test. Importantly, 90 % of children who had was that correlations between the CAS scores a neurocognitive weakness on one version of the and WJ-R Tests of Achievement were very simi- CAS also had the same neurocognitive weakness lar for Blacks (0.70) and Whites (0.64; Naglieri on the other version of the CAS. Otero et al. et al. 2005 ). Naglieri et al. (2006 ) examined CAS (2013 ) examined the performance of referred scores for 244 Hispanic and 1,956 non-Hispanic Hispanic English Language Learners of varying children. They found that the two groups differed profi ciency levels (N = 40) on the English and by 6.1 points when the samples were unmatched Spanish versions of the CAS and found no signifi - samples, 5.1 with samples matched on basic cant differences between the Full Scale scores or demographic variables, and 4.8 points when in any of the PASS scales. Students earned their demographic differences were statistically con- lowest scores in successive processing regardless trolled. These fi ndings further demonstrate the of the language in which the test was adminis- utility of PASS theory as one way to fairly assess tered. These fi ndings suggest that the CAS may diverse populations. be a useful measure for Hispanic children with When evaluating the cognitive ability of underdeveloped English language profi ciency. English language learners, psychologists cur- These results suggest that the PASS scores rently have three practice options (Ortiz 2009 ). from both the English and Spanish version of the These methods are (a) modifi cations and adapta- CAS could be used as part of a comprehensive tions of the standardized administration and scor- evaluation. ing of the test; (b) the selection and use of specifi c tests or battery of tests that are of a nonverbal nature; and (c) the use of a more traditional native Conclusions language-based test (e.g., WISC-IV Spanish). Each method has its limitations and advantages. Although several defi nitions of intelligence have The fi rst method, modifying or adapting the tests been set forth for over a century, the defi nition of in terms of administration or scoring, violates intelligence as a constellation of neurocognitive standardization directly, resulting in error and in processes such as planning, attention, simultane- less reliability and validity of scores attained. ous, and successive processes is unique and theory The second method, and perhaps the most com- driven. There is a growing need for neurocognitive monly practiced (Ortiz 2009 ), involves the use of measures to evaluate and explain function, facili- a nonverbal battery or the administration of select tate prognosis, and most importantly guide inter- subtests that make up the PIQ (Figueroa 1990 ), vention. Luria’s PASS theory offers a blueprint or more recently, the PRI. This method, although for defi ning the basic neurocognitive processes it reduces the impact of language on test results, underlying human performance, behavior, and would not be helpful in cases in which the stu- intelligence. Appreciation and application of this dent’s dysfunction is actually language based processing model as a framework for assessment (such as reading or written language). The third provides psychologists with an essential tool nec- option, the use of a native language test, may essary to not just understand children’s learning seem to be the ideal option with ELL students. and behavior but to guide and develop effective 206 T.M. Otero intervention. The PASS theory as operationalized Friston, K. (2002). Beyond phrenology: what can neuro- by the CAS-2 provides a well- developed tool for imaging tell us about distributed circuitry? Annual Review of Neuroscience, 25, 221–250. assessment of the four basic psychological pro- Goldberg, E. (2009). The new executive brain: Frontal cesses described by Luria. The importance of lobes and the civilized mind . New York: Oxford assessing PASS neurocognitive processes cannot University Press. be overemphasized in light of the evidence of its Gutentag, S. S., Naglieri, J. A., & Yeates, K. O. (1998). Performance of children with traumatic brain injury use with various groups, including Hispanic on the cognitive assessment system. Assessment, 5 , English language learners. 263–272. Harris, J. G., & Llorente, A. M. (2005). Cultural considerations in the use of the Wechsler Intelligence Scale for Children-Fourth Edition. In A. Prifi tera, D. H. References Saklofske, & L. G. Weiss (Eds.), WISC-IV clinical use and interpretation: Scientist-practitioner perspectives Bostan, A. C., & Strick, P. L. (2010). The cerebellum and (pp. 382–416). San Diego: Academic. basal ganglia are interconnected. Neuropsychology Johnson, M. H., Griffi n, R., Csibra, G., Halit, H., Farroni, Review, 20 , 261–270. T., de Haan, M., Tucker, L. A., Baron-Cohen, S., & Bostan, A. C., Dum, R. P., & Strick, P. L. (2010). The Richards, J. (2005). The emergence of the social basal ganglia communicate with the cerebellum. In: brain network: Evidence from typical and atypical Proceedings of the national academy of sciences (Vol. development. Development and Psychopathology, 107, pp. 8452–8456). 17 , 599–619. Bostan, A. C., Dum, R. P., & Strick, P. L. (2013). Kar, B. C., Dash, U. N., Das, J. P., & Carlson, J. S. (1992). Cerebellar networks with the cerebral cortex and basal Two experiments on the dynamic assessment of plan- ganglia. Trends in Cognitive Sciences, 17 , 241–254. ning. Learning and Individual Differences, 5 , 13–29. Brody, N. (1992). Intelligence . San Diego: Academic. Kaufman, A. S., & Kaufman, N. L. (2004). Kaufman Cohen, R. J., Swerdlik, M. E., & Smith, D. K. (1992). assessment battery for children (2nd ed.). Circle Pines: Psychological testing and assessment . Mountain American Guidance Service. View: Mayfi eld Publishing. Kirby, J. R. (1984). Cognitive strategies and educational Das, J. P. (1972). Patterns of cognitive ability in nonre- performance . New York: Academic. tarded and retarded children. American Journal of Kirby, J. R., & Williams, N. H. (1991). Learning prob- Mental Defi ciency, 77 , 6–12. lems: A cognitive approach . Toronto: Kagan and Woo. Das, J. P. (1980). Planning: Theoretical considerations Kolb, B., & Whishaw, I. Q. (2009). Fundamentals of and empirical evidence. Psychological Research human neuropsychology (6th ed.). New York: Worth (W. Germany), 41 , 141–151. Publishers. Das, J. P., Kirby, J. R., & Jarman, R. F. (1975). Simultaneous Kolb, B., Gibb, R., & Robinson, T. E. (2003). Brain plas- and successive syntheses: An alternative model for ticity and behavior. Current Directions in cognitive abilities. Psychological Bulletin, 82 , 87–103. Psychological Science, 12 , 1–4. Das, J. P., Kirby, J. R., & Jarman, R. F. (1979). Korkman, M. (1999). Applying Luria’s diagnostic princi- Simultaneous and successive cognitive processes . pals in the neuropsychological assessment of children. New York: Academic. Neuropsychology Review, 9 , 89–105. Das, J. P., Naglieri, J. A., & Kirby, J. R. (1994). Assessment Koziol, L. F., & Budding, D. E. (2009). Subcortical struc- of cognitive processes . Needham Heights: Allyn & tures and cognition : implications for neuropsycho- Bacon. logical assessment. New York, NY: Springer. Das, J. P., Kar, B. C., & Parrila, R. K. (1996). Cognitive Koziol, L. F., & Stevens, M. C. (2012). Neuropsychological planning: The psychological basis of intelligent assessment and the paradox of ADHD. Applied behavior . Thousand Oaks: Sage. Neuropsychology: Child, 1(2), 79-89. Das, J. P., Parrila, R. K., & Papadopoulos, T. C. (2000). Koziol, L., Joyce, A.W., & Wurglitz, G. (2014). The neu- Cognitive education and reading disability. In ropsychology of attention: Revisiting the “Mirsky A. Kozulin & Y. Rand (Eds.), Experience of mediated Model”. Applied Neuropsychology: Child. 3(4), learning (pp. 276–291). Amsterdam: Pergamon. 297-307. Dehn, M. J. (2000). Cognitive assessment system perfor- Koziol, L. F. (2014). The myth of executive functioning: mance of ADHD children. Paper presented at the Missing elements in conceptualization, evaluation and annual NASP Convention, New Orleans. assessment . New York: Springer. Figueroa, R. A. (1990). Assessment of linguistic minority Koziol, L. F., Budding, D. E., & Chidekel, D. (2011). group children. In C. R. Reynolds & R. W. Kamphaus Sensory integration, sensory processing, and sen- (Eds.), Handbook of psychological and educational sory modulation disorders: Putative functional neu- assessment of children: Volume 1 intelligence and roanatomic underpinnings. Cerebellum, 10 , achievement (pp. 135–152). New York: Guilford. 770–792. 14 Intelligence: Defi ned as Neurocognitive Processing 207

Lewandowski, L., & Scott, D. (2008). Introduction to Naglieri, J. A., & Das, J. P. (2002). Practical implications neuropathology and brain-behavior relationships. In of general intelligence and PASS cognitive processes. R. C. D’Amato & L. C. Hartlage (Eds.), Essentials of In R. J. Sternberg & E. L. Grigorenko (Eds.), The gen- neuropsychological assessment: Treatment planning eral factor of intelligence: How general is it? for rehabilitation (2nd ed.). New York: Springer (pp. 855–884). New York: Erlbaum. Publishing Company. Naglieri, J. A., & Das, J. P. (2005). Planning, attention, Lezak, M. D., Howieson, D. B., Bigler, E. D., & Tranel, T. simultaneous, successive (PASS) theory: A revision of (2012). Neuropsychological assessment (5th ed.). the concept of intelligence. In D. P. Flanagan & P. L. New York: Oxford University Press. Harrison (Eds.), Contemporary intellectual assess- Luria, A. R. (1962) Higher Cortical Functions in Man . ment (2nd ed., pp. 136–182). New York: Guilford. Moscow University Press. Library of Congress Naglieri, J. A., & Goldstein, S. (2006). The role of intel- Number: 65-11340. lectual processes in the DSM-V diagnosis of ADHD. Luria, A. R. (1966). Human brain and psychological pro- Journal of Attention Disorders, 10 , 3–8. cesses . New York: Harper and Row. Naglieri, J. A., & Otero, T. (2011). Cognitive Assessment Luria, A. R. (1973). The working brain. New York: Basic System: Redefi ning intelligence from a neuropsycho- Books. logical perspective. In A. Davis (Ed.), Handbook of Luria, A. R. (1979). The making of mind: A personal pediatric neuropsychology (pp. 320–333). New York: account of Soviet psychology . Cambridge, MA: Springer. Harvard University Press. Naglieri, J. A., & Otero, T.M. (2012). The Cognitive Luria, A. R. (1980). Higher cortical functions in man (2nd Assessment System. In D. P. Flanagan and P. L. ed.). New York: Basic Books. Harrison (Eds.) Contemporary Intellectual Assessment Luria, A. R. (1982). Language and cognition. New York: (Third Edition) (pp. 379-399). New York: Guilford. Wiley. Naglieri, J. A., & Pickering, E. (2003). Helping children McCloskey, G., Perkins, L. A., & Van Diver, B. (2009). learn: Instructional handouts for use in school and at Assessment and intervention for executive function home . Baltimore: Brookes. diffi culties . New York: Routledge. Naglieri, J. A., & Rojahn, J. R. (2004). Validity of the Mccrea, S. M. (2009). A cognitive neuropsychological PASS theory and CAS: Correlations with achieve- examination of the Das-Naglieri cognitive assessment ment. Journal of Educational Psychology, 96 , system subtests: A report of three stroke cases studied 174–181. longitudinally during recovery. International Journal Naglieri, J. A., Goldstein, S., Iseman, J. S., & Schwebach, of Neuroscience, 119 (4), 553–559. A. (2003). Performance of children with attention Mirsky, A. F., & Duncan, C. C. (2001). A nosology of dis- defi cit hyperactivity disorder and anxiety/depression orders of attention. Annals of New York Academy of on the WISC-III and Cognitive Assessment System Science, 931 , 17–32. (CAS). Journal of Psychoeducational Assessment, 21 , Mirsky, A. F., & Duncan, C. C. (2003). Attention battery for 32–42. children: A systematic approach to assessment. In Naglieri, J. A., Salter, C. J., & Edwards, G. (2004). G. Goldstein & S. R. Beers (Eds.), Comprehensive hand- Assessment of ADHD and reading disabilities using book of psychological assessment, volume 1, intellectual the PASS theory and cognitive assessment system. and neuropsychological assessment . New York: Wiley. Journal of Psychoeducational Assessment, 22 , 93–105. Naglieri, J. A. (1997). Naglieri nonverbal ability test . San Naglieri, J. A., Rojahn, J. R., Matto, H. C., & Aquilino, Antonio: The Psychological Corporation. S. A. (2005). Black white differences in intelligence: Naglieri, J. A. (2003). Current advances in assessment A study of the PASS theory and cognitive assessment and intervention for children with learning disabili- system. Journal of Psychoeducational Assessment, 23 , ties. In T. E. Scruggs & M. A. Mastropieri (Eds.), 146–160. Advances in learning and behavioral disabilities Naglieri, J. A., Rojahn, J., & Matto, H. (2007). Hispanic Volume 16: Identifi cation and assessment (pp. 163– and non-Hispanic children’s performance on PASS 190). New York: JAI. cognitive processes and achievement. Intelligence, 35, Naglieri, J. A., & Bornstein, B. T. (2003). Intelligence and 568–579. achievement: Just how correlated are they? Journal of Naglieri, J. A., Goldstein, S., Delauder, B. Y., & Psychoeducational Assessment, 21 , 244–260. Schwebach, A. (2006). WISC-III and CAS: Which Naglieri, J. A., & Conway, C. (2009). The cognitive correlates higher with achievement for a clinical sam- assessment system. In J. A. Naglieri & S. Goldstein ple? School Psychology Quarterly, 21 , 62–76. (Eds.), A practitioner’s guide to assessment of intelli- Naglieri, J. A., Rojahn, J., & Matto, H. (2007). Hispanic gence and achievement (pp. 3–10). New York: Wiley. and non-Hispanic children’s performance on PASS Naglieri, J. A., & Das, J. P. (1997a). Cognitive assessment cognitive processes and achievement. Intelligence, 35 , system . Itasca: Riverside Publishing Company. 568–579. Naglieri, J. A., & Das, J. P. (1997b). Cognitive assessment Naglieri, J. A., Das, J. P., & Goldstein, S. (2014a). system interpretive handbook. Chicago: Riverside Cognitive assessment system (2nd ed.). Austin: Pro-Ed Publishing Company. Publishing Company. 208 T.M. Otero

Naglieri, J. A., Das, J. P., & Goldstein, S. (2014b). Savage, R. C., & Wolcott, G.F. (Eds). (1994). Educational Cognitive assessment system-second edition. dimensions of acquired brain injury. Austin, TX: Interpretive handbook . Austin: Pro-Ed Publishing PRO-ED Company. Schneider, W., Dumais, S. T., & Shiffrin, R. M. (1984). Neville, H. (2006). Different proﬁ les of plasticity within Automatic and controlled processing and attention. In human cognition. In Y. Munakata & M. Johnson R. Parasuraman & D. R. Davies (Eds.), Varieties of (Eds.), Processes of change in brain and cognitive attention (pp. 1–28). New York: Academic. development . New York: Oxford University Press. Semrud-Clikeman, M., & Teeter Ellison, P. A. (2009). Neville, H., & Stevens, C. (2008). Experience shapes Child neuropsychology: Assessment and intervention human brain development and function: A framework of neurodevelopmental disorders (2nd ed.). New for planning interventions for children at-risk for Jersey: Springer. school failure (Summary) . Annual meeting of the Stuss, D. T., & Benson, D. F. (1990). The frontal lobes and American Association for the Advancement of language. In E. Goldberg (Ed.), Contemporary psy- Science, Boston. chology and the legacy of Luria . Hillsdale: Lawrence O’Shanick, G. J., & O’Shanick, A. M. (1994). Personality Erlbaum Associates. and intellectual changes. In J. M. Silver, S. C. Thorndike, R. L. (1997). Measurement and evaluation in Yudofsky, & R. E. Hales (Eds.), Neuropsychiatry of psychology and education (6th ed.). Upper Saddle traumatic brain injury (pp. 163–188). Washington, River: Prentice Hall. DC: American Psychiatric Press. Varnhagen, C. K., & Das, J. P. (1986). Neuropsychological Ortiz, S. O. (2009). Bilingual-multicultural assessment functioning and cognitive processing. In J. E. Obzrut with the WISC-IV. In E. O. Lichtenberger & A. S. & G. W. Hynd (Eds.), Child neuropsychology, Vol. 1: Kaufman (Eds.), Essentials of the WISC-IV (pp. 295– Theory and research (pp. 117–140). New York: 309). New York: Wiley. Academic. Otero, T. M., Gonzalez, L., & Nagleri, J. A. (2013). The Vygotsky, L. S. (1976). Mind in society: Development of neurocognitive assessment of Hispanic English- higher psychological processes . London: Harvard language learners with reading failure. Applied University Press. Neuropsychology: Child, 2 (1), 24–32. Wechsler, D. (1991). Wechsler intelligence scale for Paolitto, A. W. (1999). Clinical validation of the Cognitive children (3rd ed.). San Antonio: The Psychological Assessment System with children with ADHD. ADHD Corporation. Report, 7 , 1–5. Wechsler, D. (2003). Wechsler intelligence scale for children Penrose, L. S., & Raven, J. C. (1936). A new series of per- (4th ed.). San Antonio: The Psychological Corporation. ceptual tests: Preliminary communication. British Wechsler, D., & Naglieri, J. A. (2006). Wechsler nonver- Journal of Medical Psychology, 16 , 97–104. bal scale of ability . San Antonio: Harcourt Assessment. Perez-Alvarez, P., Timoneda-Gallart, C., & Baus-Rosell, Wilkinson, A., & Sermund-Clikeman, M. (2008). Motor J. (2006). Topiramato y epilepsia a la luz del Das- speed in children and adolescents with nonverbal Naglieri Cognitive Assessment System. Revista learning disabilities. Paper presented at the Neurologia, 42 (1), 3–7. International Neuropsychological Society, Waikoloa, Pluker, J. A., & Esping, A. (2014). Intelligence 101 . Hawaii. New York: Springer. Woodcock, R. W., & Johnson, M. B. (1989). WJ-R Tests of Posner, M. I., & Boies, S. J. (1971). Components of atten- Cognitive Ability. Itasca, IL: Riverside Publishing. tion. Psychological Review, 78 , 391–408. Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Roid, G. (2003). Stanford-Binet ﬁ fth edition . Itasca: Woodcock-Johnson III. Itasca, IL: Riverside Riverside. Publishing. CHC Theory of Intelligence 15 Samuel O. Ortiz

The concept of intelligence is as ubiquitous as in a given context. But when a situation requires any other artifact of our culture. It is a word that a precise and scientifi cally validated defi nition, it is used often, readily accepted in meaning, and is alarming to see both lay people and profession- used to invoke explanations for all sorts of social als scramble for support like rats on a sinking and behavioral phenomena observed and encoun- ship. Indeed, perhaps the most natural reaction in tered on a daily basis. It is found frequently in all such cases is to invoke the one explanation pro- forms of popular and professional literature and vided by McNemar (1964 ) which is sure to discourse and is a common topic in introductory appease everyone, where he noted “no defi nition science texts across myriad specialties and related is required because all intelligent people know disciplines, particularly psychology. Even the what intelligence is—it is the thing the other guy term born from the idea of intelligence, “IQ” (or lacks” (p. 871). “intelligence quotient”), is no longer an acronym The focus of this chapter is not centered per se but an accepted word in all modern lan- directly on intelligence. The purpose here is to guage dictionaries including the one that even describe the history and development of what is governs the offi cial rules of the game, Scrabble. currently referred to as CHC theory, an acronym Unquestionably, we can point to few things in derived from the chronological order of contri- modern life that can claim to be so familiar, so bution by the surname of its three main develop- well known, and so much a part of our daily ers (i.e., Cattell, Horn, and Carroll). It is experience as the notion of intelligence and our impossible, however, to discuss the develop- presumed understanding of it. ment of the theory without addressing the That intelligence is so ingrained in what we debates surrounding the very notion and defi ni- know value may also have done it a considerable tion of intelligence, since any such theory must disservice. Consider, for example, we all readily necessarily provide an explanation of it. And admit that we know exactly what it refers to, despite offering what is considered perhaps the and thus rarely is anyone called upon to defi ne it most empirically validated and best supported or its usage in a given context. Even less fre- theory of human cognitive abilities to date quently is someone asked to defend the theoreti- (e.g., Schneider and McGrew 2012 ), CHC the- cal or empirical bases that would support its use ory is unlikely to be the fi rst mention in response to any query regarding conceptualizations of intelligence. Admittedly, there has been an * S. O. Ortiz ( ) extraordinary and precipitous increase in pub- St. John’s University , 8000 Utopia Pkwy , Jamaica , NY 11439 , USA lished and unpublished research regarding CHC e-mail: [email protected] theory since 2000 (Schneider and McGrew 2012 ),

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 209 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_15, © Springer Science+Business Media New York 2015 210 S.O. Ortiz but the reason why CHC theory remains in rela- Cattell’s beliefs were no different than the vast tive obscurity compared to other, more popular majority of his fellow colleagues at the time, notions of intelligence (e.g., Spearman’s g , particularly his British counterparts whose his- Gardner’s ‘multiple intelligences’) is an impor- torical class structure was justifi ed on the basis tant part of its history and extremely relevant to of hereditary power stemming from hereditary its development. Likewise, because the theory attributes—particularly intelligence. Coupled itself carries the very names of its constituent with a society that was still actively engaged in authors, discussion of the one cannot be accom- colonialism, albeit wrestling with the notion plished without discussion of the other. more so than before and searching for scientifi c support to justify the practice, it is not surprising that Cattell’s own early thoughts concerning CHC Theory and Cattell: Birth intelligence differed little from his mentors or and the Early Years others at the time. What is truly remarkable about Cattell, however, was his willingness to modify The basic formulations of what would eventually his position and accept the mistakes and failings become known as CHC theory began with that characterized some of his early beliefs and Raymond B. Cattell (1941 , 1943) whose aca- theoretical conceptions when the data suggested demic pedigree in psychology could be traced otherwise. For example, in 1997 when Cattell back to his dissertation mentor, Charles was nominated and selected for the prestigious Spearman, who himself was trained by Wilhelm APA Gold Medal for Lifetime Achievement in Wundt. Born in 1905, it is important to recall and the Science of Psychology, he was quickly understand the time in which Cattell was attend- accused of being a racist and condemned by a ing school and how he was infl uenced heavily not small faction who hinged their attack on Cattell’s only by his education and academic mentoring early thoughts on eugenics and theology as well but also the values and beliefs of his culture and as statements he made some 60 years prior and the prevailing zeitgeist. Having just missed mili- without any recognition of the manner in which tary service in World War I, Cattell entered his views evolved over the course of several University College in London in 1921 studying decades. In an open letter to APA, he declined the chemistry primarily but then moving into psy- award and, perhaps irrevocably heartbroken by chology for graduate studies in 1925 at Kings the unfair portrayal, died several months later. College. Cattell had the good fortune of working Whereas the label of racist might be fairly with Spearman and completing his dissertation applied to some psychologists whose formative right around the time Spearman was refi ning the years spanned the early 1920s to late 1940s, it is process of evaluating correlational data statisti- not accurate to do the same with Cattell. First, he cally in a manner that would eventually become was a product of the times, and many of those known as factor analysis. Needless to say, Cattell with similar beliefs and attitudes (e.g., C. C. was very well trained in this relatively new Brigham, Lewis Terman, H. H. Goddard) eventu- method and adopted not only the methodology ally recanted those beliefs formally in the face of from Spearman but a good deal of his philosophy better science, a mature perspective of life, or an as well. This included a strong belief in the intolerance for the ruse that drives various aspects genetic nature and determination of intelligence, of scientifi cally based racism. In fact, Cattell not ideas regarding the need for proactive policy with only threw over many of the positions which had regard to individual’s identifi ed with low intelli- become socially undesirable and untenable, he gence, and perhaps the most salient idea regard- actually departed in a signifi cant way from the ing his future career—a staunch acceptance of very idea used to support such discriminatory data analytic methods that provided a fi rst-order policies—the belief that there was one and only general factor which was believed to represent g one type of intelligence (i.e., general intelligence or general intelligence. It should be noted that or g ) and that it was supreme among all other 15 CHC Theory of Intelligence 211 so-called second-order factors that might be tests for use in evaluating adults, particularly identifi ed. That Cattell was able to break with those over the age of 20, and despite listing some what is the most important empirical foundation 44 popular tests currently in print, Cattell (1943 ) for scientifi cally driven racist beliefs not only denounces the lot as refl ecting a “dearth of tests” indicates that he was not afraid to relinquish such that “must come as a shock to most psycholo- inequitable ideas as fallacious but that he was gists, for it has been widely assumed that the also willing to challenge the very core of the momentum of real progress in intelligence test prevailing statistical ideology that he had been theory and practice which arose in the fi rst two taught by none other than Spearman himself. decades of this century as continued unabated Cattell’s application of factor analytic tech- through the ensuing 20 years” (p. 156). That tests niques and his early dissatisfaction with the idea have failed to demonstrate any useful progress of general intelligence or a single g began with a over the prior two decades is ascribed by Cattell very brief presentation at the APA annual confer- to problems with the fundamental theory of ence in 1941 published in the Psychological intelligence. Bulletin that may well have been prompted by the At this point in his career, Cattell had already renewed interest in the testing of military recruits begun dismissing many of the foundations of the as the USA entered World War II. In that paper genetic arguments for intelligence that had gone (Cattell 1941 ), he stated, “the notion of a general hand in hand with Spearman’s g . Cattell’s focus factor does not recommend itself so strongly in on adults sensitized him to the problems involved adult testing because adult tests are less saturated in evaluating intelligence free from the effects of with a general factor” (p. 592). Two things are schooling, verbal skills, and mathematical abili- noteworthy in his comments. First, that he is ties and the need to account for the decline in beginning to see that g is an inadequate explana- “speed” but not “power” of intelligence as one tion for intelligence at the adult level. And second, ages. Cattell was clearly dissatisfi ed with notions that intelligence must be viewed as a develop- regarding tests of intelligence and their concomi- mental process. Perhaps one of the greatest and tant lack of theory apart from the three basic rules most overlooked contributions to intelligence taken from three different fi elds of study. Clinical provided by Cattell (and still refl ected in research study of individuals with low intelligence pro- conducted on CHC theory today) is the attention vided the notion that intelligence involved the to the manner in which abilities develop, peak, capacity to think abstractly. Animal studies and maintain, or decline across the life span. Cattell analogs offered up the idea that intelligence was concerned himself with developmental issues refl ected by an ability to learn. And last, mea- to an extent that is extremely uncommon even surement in education generated the premise that today but which is fundamental to any defi nition. intelligence must involve the capacity to adapt Cattell’s movement away from g can also be seen means to ends. These three principles were reiter- in his admonition that answers to the inadequa- ated widely at the time and troubled Cattell cies of g in adults may rest with the work of greatly as evidenced by his citation of Wechsler Thurstone and Thomson, and he offered a practi- ( 1939 ) and Wells (1932 ) to wit: “Wechsler, in his cal solution that forms the very beginnings of Bellevue Scale manual, defi nes adult intelligence what would become known as Gf-Gc theory (one as ‘the aggregate or global capacity of the indi- of the precursors to CHC theory). His proposed vidual to act purposefully, to think rationally, and solution to measurement is better delineated in a to deal effectively with his environment.’ But later publication (Cattell 1943) where he begins Wells (p. 265), deploring the fact that Wechsler’s to attack tests on several fronts including inade- concept of intelligence involves adding up the quate psychometric properties, norm samples subtest scores into a single total, says: ‘The chief lacking representation for the general population, use of global scores is administrative.’” (p. 159). and overreliance on manual dexterity. For the Cattell argued forcefully that despite a wide array most part, however, he decries the paucity of suitable of published tests of intelligence, few authors 212 S.O. Ortiz attempted to provide positive statements about showing the most precipitous declines. It must the nature of intelligence, and those that did were have dawned on him quickly that if there is such no different than what had been offered even 40 a thing as a single general factor, the evidence of years prior. It was not so much that Cattell simply abilities that decline or do not decline did not sup- felt any general theory of intelligence was lack- port it. A new conceptualization must be created ing but more that there was very little support for to explain this phenomenon, and on page 178, he intelligence irrespective of the theory. He notes fi nally writes, for perhaps the fi rst time in formal that “those applied psychologists who have been discourse, that the difference between such abili- the most prolifi c designers and users of ‘intelli- ties can be expressed by the terms “crystallized” gence tests’ and who have so long and so uncriti- when implicating the former and as “fl uid” in cally accepted the sum of a hodgepodge of tests referring to the latter (Cattell 1943 ). In describing in the form of a single IQ measurement are now his “hypothesis of fl uid and crystallized ability,” swinging—on no better evidence—to the oppo- (p. 180), Cattell postulates that if sustained, his site extreme of demanding that tests should yield formulation would be critical to intelligence test- measurements of separate abilities” (p. 161). All ing in adults by requiring the specifi c measure- this is not to say that Cattell was completely dis- ment of both abilities rather than relying on enchanted with the notion of a general factor of notions of general intelligence. It should be intelligence. In discussing his opinions regarding understood as well that Cattell was not wholly the competing viewpoints of Spearman and dismissing Spearman’s g or attempting to devalue Thurstone, Cattell argues that Spearman did it and its utility. Rather, he was merely breaking it admit to fi nding certain group factors and that into two relatively equal pieces—both of which Thurstone also conceded that a general factor could be considered equivalent components to could be identifi ed. But he leans in favor of intelligence that should be viewed as distinct, yet Spearman by noting that he “introduces his group cohesive in a general sense. That he likely hoped factors to the reader with a cold and perfunctory to retain ties to his mentor, Spearman, is evident politeness, while Thurstone’s general factor is in the data and analyses Cattell uses to support his only permitted to enter society as a ‘second-order new theoretical formulation. Cattell took great factor’ after the ‘primary abilities’ have made off pains to tease out speed issues, noting that there with all of the actual test variance” (p. 170). Still, were various aspects of speed which were related Cattell emphasizes that factor analysis will not to the decline in fl uid abilities. Yet, he did not bring such a debate to an end but that it will per- attempt to incorporate speed as a factor in its own mit signifi cantly greater objectivity to the entire right, either because his analysis of the data did enterprise. He understood that disagreements in not permit it or he could not reconcile the prob- science will continue to stem from differences in lems a third factor introduces with respect to the how data are viewed and interpreted, but the arbi- idea that there should be only a single general trariness of the endeavor is removed and elevates one. Two parts of the same thing was likely the enterprise to the level that true science acceptable in principle, and Cattell even used demands. Graphical representations of Spearman’s designations (i.e., lower case gf and Spearman’s and Thurstone’s models are provided gc) to maintain the link to a single g . But three in Fig. 15.1 for comparative purposes. would likely have proven too much of an anom- Following his own admonitions and concerns aly, and the matter remained unaddressed until he about theory, Cattell introduced the notion that mentored a dissertation designed to examine his deterioration in intelligence in adults was not uni- theory by his student John Horn who had no form across all tests but differential, with the qualms about identifying the presence of addi- higher g -saturated tests (e.g., vocabulary, infor- tional factors that had been either conveniently mation, verbal comprehension) showing the overlooked by Cattell or simply hidden by his least decline across the life span and relatively own personal and professional biases. g- unsaturated tests (e.g., speeded tasks, abstract Despite his stated concerns about the glacial problem-solving, unfamiliar performances) pace at which intelligence theory had proceeded 15 CHC Theory of Intelligence 213

Spearman’s g

g First order general factor

Second order group factors S1 S2 S3

Thurstone’s PMA’s First order broad (but not general)factors

VWN S M P R

Second order general factor (g)

Fig. 15.1 Comparison of Spearman’s g and Thurstone’s primary mental abilities

prior to his explication of a two factor g f- gc various theoretical frameworks, primarily single model, Cattell himself allowed two decades of g, that had been carried over unchanged for the his own to pass before he put his theoretical past half century (Cattell 1963 ). In hindsight, framework to the test. It was not that he com- Cattell’s inability, reluctance, or other reason for pletely neglected the theory, and indeed he avoiding the necessary research may have also offered a few refi nements along the way (Cattell played a role in why his theoretical model was 1950 ; Cattell 1957a , b ). But evidence in support not readily adopted by the major test publishing of his theory lagged signifi cantly, and Cattell companies and why, despite the advancements in admitted that the bridging of the theory-practice thinking it accorded, was not used as a platform gap was not as simple as it might seem. Part of for the development or revision of current and the problem was likely the industrialization of future intelligence tests. The work of Cattell’s intelligence tests and testing which had provided student, John Horn, would begin to correct this an economic boom in its own right and resulted oversight but not without another three decades in greater diffi culties as well as added costs of relative obscurity. required for development. Nevertheless, Cattell persisted and after 20 years of theory, he turned his attention to empirical investigation in support CHC Theory and Horn: Headlong of his yet unsubstantiated model. The result was into Adolescence a publication that serves as the offi cial pro- nouncement of the birth of the “theory of fl uid In his landmark paper on g f-g c theory, Cattell and crystallized intelligence” and outlined a (1963 ) makes a notable reference to the unpub- well-developed and modern conceptualization of lished work of John Horn, the second major intelligence that contrasted starkly against the contributor to CHC theory. Cattell emphasized 214 S.O. Ortiz that investigations of g f-g c theory cannot be abilities including general visualization (or Gv) made piecemeal or relative to one ability but not and general speediness (or Gs).1 the other. He gleefully noted that “the necessary Figure 15.2 provides a side-by-side view of experimental conditions for constructive con- Cattell’s original g f- gc formulation and the modi- clusions [about the theory] are possible only if fi cations and conceptual differences suggested by the total theory is kept in focus, as is happily the Horn ( 1965 ). As is evident in the illustration, case in, for example, the recent work of Horn” Horn retained the basic defi nitions provided orig- (p. 2). Cattell does not cite Horn’s research in inally by Cattell in terms of Gf and Gc, including the bibliography, but the timing of the article the use of the term “intelligence” as part of the suggests that he was most likely referring to name for each one (i.e., fl uid intelligence and Horn’s (1965 ) dissertation which was in prog- crystallized intelligence). He viewed these as ress around that time. During an informal col- broad dimensions exhibited as individual differ- legial dinner and discussion among a number of ences involving various aspects of reasoning and so-called CHC enthusiasts, Horn related per- problem-solving in both, but without any require- sonal details on the history of his dissertation ment of pre-training, knowledge, or education in explaining that he had hoped to do something the former and as specifi cally refl ected as neces- quite different, and when it did not work out, sary “skills which constitute the collective intel- Cattell provided him with some data and recom- ligence of the culture and which are learned mended that he analyze it (J. Horn—personal under conditions of intensive acculturation” communication, March 1, 2007). He did, and (p. 309). The construction of Gc in this manner not surprisingly, the title of his thesis is quite on highlights an important dilemma in test construc- point, “Fluid and crystallized intelligence: A tion that seeks to base itself on sound theory. If an factor analytic study of the structure among pri- ability is defi ned as “acculturation” and specifi ed mary mental abilities” and indicates that just as to be the “result … of opportunity such as is Spearman had passed on the legacy of factor occasioned by special schooling and continued analysis to Cattell, so too did Cattell pass on his exposure to the culture in again,” the practice of affi nity for factor analysis to Horn, but not nec- stratifying normative samples primarily on the essarily their attachment to a single g or to two basis of age becomes problematic. Certainly, related g factors (g f-g c). education can be easily controlled as it is not Horn apparently got a very quick start in his diffi cult to ascertain the extent to which an indi- role as a contributor to Cattell’s newly formed vidual has received formal instruction over the g f- g c theory. He had already been studying with life span. But rarely is that information coupled Cattell and contributing some research in line with actual age in determining an appropriate with the theory when his dissertation provided a standard for comparison. More troubling, how- powerful and comprehensive test of the theory, ever, is the lack of attention paid to the process of much like Cattell had done in 1963, and inadver- cultural knowledge acquisition outside of school tent as his topic may have been, it set the course that Horn points out as being related to age. for much of the rest of his academic career. What Horn found, however, was that Cattell was essen- 1 It should be noted here that it is not clear why Horn chose tially “wrong” in suggesting that there existed to use capital letters in designating these abilities. Whereas two distinct but similar aspects of general intelli- it was not an uncommon practice in the literature at the time, it may have been a deliberate choice on Horn’s part gence because his new analyses, completed in the so as to refl ect his desire to eliminate all direct links to any manner specifi ed and approved by Cattell him- of the abilities being misconstrued as components of a self, provided support for at least four such pri- larger general factor, or g—an idea he was beginning to mary factors. In addition to what he referred to as accept as untenable even in his dissertation. Whatever the case, the capital letter designations have become the “Gf” and “Gc,” Horn suggested the data he col- default and preferred format within CHC theory and are lected also supported two other primary mental used throughout the remainder of the chapter. 15 CHC Theory of Intelligence 215

Cattell’s Original gf-gc model

Two second-order general factors representing g

gf gc

Ps Pc Ma VNR

First order primary factors First order primary factors related to g f related to gc

Cattell-Horn “Early” Gf-Gc

Gf Gc Gv Gs

I Mk Cs P

L EMS Cf Sc

R N

DFT V

Fig. 15.2 Comparison of Cattell’s g f- g c theory and the Cattell-Horn early Gf-Gc theory

Whereas individuals residing in the same country results of so-called “culture-fair” and “culture- under relatively similar conditions would be free” tests of intelligence which were intention- expected to share approximately the same oppor- ally designed to minimize the effects of prior tunities for learning about the culture in which schooling or the need for previous learning but they live, the same cannot be said for individuals showed only modest correlations to more com- whose opportunities for learning have been lim- prehensive tests and lower g -saturation on the ited in comparison to same age peers due to fac- whole (Cattell 1943 , 1963 ; Horn 1965 ). If Gc tors such as immigration, ethnic and cultural must be measured given its centrality to the con- differences in the home and community, and lan- cept of intelligence, a more precise delineation of guage differences. Moreover, in what manner can experiential differences in exposure and prior Gc be measured without the use of Gc-based learning, above and beyond age and education, stimuli? As Cattell so proudly announced, the will be crucial to the generation of fair and equi- theory of fl uid and crystallized abilities must be table standards. evaluated as a whole such that evaluation of one Horn’s (1965 ) analyses presented evidence in or the other alone undermined empirical efforts support of a general visualization (Gv) primary to support it. This had already been seen in the factor that he described as “the processes of 216 S.O. Ortiz imagining the way objects may change as they model was solidifi ed (Horn and Cattell 1966a , b ). move in space, maintaining orientation with As noted previously, it is a testament to Cattell’s respect to objects in space, keeping confi gura- ability to alter his beliefs in the face of compel- tions in mind, fi nding the Gestalt among dispa- ling evidence as it is clear that he was more than rate parts in a visual fi eld and maintaining a happy to let go once and for all the notion that fl exibility concerning other possible structurings there existed a general intelligence, whether of elements in space” (p. 310). Cattell had alluded comprised of a single g or two g’s (i.e., g f- gc). to elements of so-called visual processing in his His outright rejection of this theoretical notion is prior analyses but had mostly seen the issue as given special credence by its conspicuous and one related to the covariance observed in age- prominent placement as the very fi rst sentence related decline of visual perceptual tasks due to ever published jointly between he and Horn degradation of visual acuity. As a developmental (Horn and Cattell 1966a ) which stated rather psychologist, Cattell understood and recognized emphatically, “the theory of fl uid and crystallized the role that diminished vision played in attenuat- intelligence … seriously questions the notion that ing test performances that relied on visual- there is a unitary structure which can be desig- perceptual processes, and thus he was inclined nated general intelligence” (p. 253). Not content not to see it as an independent and distinct factor with upsetting the proverbial apple cart with this in its own right. Horn had no such diffi culty and single premise, they add that Gf-Gc theory also in his own elaborate style provided a strong argu- “questions the belief—often implicit, but ment for Gv including “quite defi nite support for expressed clearly in a recent article by McNemar the hypothesis of a general visualization dimen- (1964 )—that the conglomerate measured by sion spanning the facets of Vz [visualization], S combining subscores from a collection of intel- [spatial orientation], Cf [fl exibility closure], Cs lectual tests is the best estimate of intelligence” [speed closure], and DFT [adaptive fl exibility] (p. 253). In this regard, Horn and Cattell had and dipping into measures of Gf when these become more aligned with Thurstone’s (1946 ) involve Figural content ” (pp. 280–281). original position that: Horn (1965 ) presented similarly persuasive Instead of attempting to describe each individual’s evidence in support of general speediness (Gs) as mental endowment by a single index such as a a primary factor. He hypothesized that Gs “could mental age or an intelligence quotient, it is prefer- perhaps be an attribute indicating a state of test- able to describe him in terms of a profi le of all the primary factors which are known to be signifi - taking effortfulness, rather than a stable trait” and cant…. If anyone insists on having a single index subject to variation as a function of the nature of such as an I.Q., it can be obtained by taking an a given task (p. 310). Nonetheless, he indicated average of all the known abilities. But such an that Gs “is measured most purely in simple writ- index tends so to blur the description of each man that his mental assets and limitations are buried in ing and checking tasks which require little in the the single index. (p. 110) 2 way of complex relation-perceiving” and further admonished that “the function itself produces variance in the measure of most intellectual functions unless care is taken to cancel it out by mea- 2 As yet another example regarding the infl uence of aca- suring with both unspeeded and speeded tasks” demic genealogy, Horn co-mentored this chapter’s (p. 310). Elements of this speed factor included author’s own dissertation which examined various meth- Sc (speed copying), Wf (writing fl exibility), and ods of data aggregation in defi ning latent variables—that is, what is the best way to put two or more scores together P (perceptual speed). mathematically to represent a single psychological con- In a manner of speaking, Horn (1965 ) simply struct. Prior to the fi nal results, Horn predicted that a opened the fl oodgates with respect to the identifi - simple arithmetic average would emerge as the superior cation of primary factors, and as Cattell joined method, and in contrast to the manner in which many constructs and test score composites/clusters are calculated in him in publication, the affi rmation of the empiri- the present day, he was indeed correct every bit as much as cal support for an early Cattell-Horn Gf-Gc was Thurstone. 15 CHC Theory of Intelligence 217

Neither Horn nor Cattell was alone in the many times –ad nauseam some may think. Still, rejection of a general factor. But for Cattell, for those who care about evidence, there is lots of it. One can examine it, and when one does one whose life, education, and experiences were fi nds a drip, drip, drip of results from study after rooted in just such a notion, it must have been a study punching out huge holes in the belief that truly radical departure from his early founda- there is a g (somewhere) and demonstrating that g tions. On the other hand, Horn was well hasn’t been found and that it now seems unlikely that it will be found. In any case, if there is a g, we acquainted with the “British” school and its ide- have yet to fi nd it. ology, having learned it fi rsthand from both And there is no contrary evidence, no evidence Cattell and via his Fulbright work in Australia supportive of g. The only thing that gets treated as (1956–1957) and his position as a research asso- evidence is positive manifold of the intercorrelations among measures of cognitive abilities and a ciate at the University of London in 1972. Indeed, string of correlations with other variables that having “seen” the other side may have allowed refl ect this positive manifold. But this is evidence Horn to view the British perspective from a more that Thurstone showed many years ago does not objective angle and therefore sensitized him to its support a structural hypothesis of g, much less a developmental, genetic, neurological, educational, shortcomings and allowed him to see problems social, anthropological--n general, a construct and anomalies where others saw none. His insight validit--hypothesis. into the problems with a general factor was Recently, for example, McArdle has presented revealed in an interesting manner more recently no fewer than three studies showing that g does not work structurally, developmentally, and when he responded to a seemingly innocuous dynamically. Also recently is the evidence of post on the fi rst list serve ever created specifi cally Richard Roberts analyses of the Armed Services dedicated to the topic of CHC theory by Kevin Vocational Aptitude Battery (ASVAB)--he battery McGrew. Not long after its launch, a question used in the data analysis parts of the infamous Herrnstein & Murray “Bell Curve” study. In the arose regarding why the WJ-R ( Woodcock and H&M studies obeisance was given to g when in Johnson 1989 ) seemed to correlate so poorly fact the evidence of H&M’s own factoring indi- with the venerable Wechsler scales and whether cated no g. Roberts’ results elegantly demonstrate such modest correlations were indicative of a test this. (Here one might want to look at my review of H&M’s book, there also pointing out this problem. that did not actually measure intelligence, par- Also good reading on this point are the Haut et al. ticularly in the broad or general sense. Being a reports, book and papers). charter member of the list serve, Horn took the Just a couple of years back, Schonemann and a opportunity to address the issue—which at the whole host of responders to his work, concluded that there is no g. A little further back in history is time must have seemed that we had all returned Carroll’s monumental work where, as I point out in to a state of quiet acquiescence of Spearman’s g . several papers (again ad nauseum perhaps), there Because these are his words (previously unpub- are no fewer than 8 different general factors, all lished in their entirety), they merit inclusion in quite distinct, but still referred to as “the” (singular) general factor or g. Prior to that, reviewing this chapter at this very point and are offered here Jensen’s “Bias in Mental Testing,” Horn & for the benefi t of the reader. His remarks, as they Goldsmith found that what Jensen referred to as appear below, are quoted in their entirety, exactly “g” in one chapter of his book was most similar to as he wrote them on August 2, 1999. Gc, what he referred to as g in another chapter was similar to Gf and what he accepted as g in still The problem is that there is no g--.e., no single g. another chapter was essentially Gv. While Jensen’s Of course this is contrary to existing dogma. But work presents particularly stark examples of this dogma is dogma, not evidence, not something we chameleon-like interpretation of ability measure- want much of in science. It is an assumption ments, in fact he is simply doing what many others implicitly accepted, an assertion made so fre- do. But if one looks at the evidence, s/he will see quently, by so many who are assumed to be (and Gf, Gc, Gv, etc., have quite distinct construct valid- assume themselves to be) authorities, and made so ities –quite different relationships to neurological, uncritically that it is widely accepted as true. educational, vocational, genetic variables--n gen- (How could something said so often, so confi - eral the network of variables that provide a basis dently, so casually, by so many, and so many for understanding human capabilities. Going back smart, and informed people, not be true?) But the further yet there are the classic studies of El evidence adds up, as I have said now so many, Koussy ( 1935) and Rimoldi ( 1948) studies that 218 S.O. Ortiz

were steadfastly and beautifully designed to prove Of primary signifi cance in Horn’s comments the validity of g, but concluding--eluctantly, almost is the argument that attempts to distill measure- sadl--that the g hypothesis can not be supported. Then, too, more tangentially, there are the well- ment of a single general ability factor result only designed and well-executed studies of Gustafsson in the measurement of a distinct broad ability fac- (1984 , 85), Undheim (1976 ) and Undheim and tor, notably Gf, when language and cultural ele- Gustafsson ( 1987 ) showing that in batteries of tests ments are not used, or Gc, when they are. Horn designed for children the general factor was identical to the Gf factor. Relatedly, there are the results also makes reference to the version of the theory from our work showing that a battery very care- that became known as the “modern Gf-Gc” fully designed to provide evidence of one and only framework and which greatly expanded the num- one factor corresponding to Spearman’s g comes ber of primary (narrow) and secondary (or broad, very close to succeeding, as in Thurstone and Thurstone ( 1941 ), but the g that is indicated is Gf but not general) mental abilities included in the (and only Gf): as soon as other well-regarded indi- theory. That expansion was relatively rapid paced cators of intelligence, such as those of Gc, Gv, Ga, as in ( 1966a ), only a year after introducing two etc., are considered, that “g” factor (which is Gf) new broad, secondary factors in his dissertation disappears. So, old friends of NASP--and Jensen, (Gv and Gs), he adds a fi fth to the collection “F” Bouchard, Eysenck, Carroll, etc. (the list is long) (general fl uency) which was more of a process be aware that no battery of ability tests provides a that was refl ected in certain tasks, particularly measure of g, because there is no g, only conglom- those that required speed. The specifi c abilities erate, composites. One good reason why the composite score on the WJR may correlate at a lower subsumed under this broad factor included Fa level with this or that composite or other variable is (associated fl uency), Fi (ideational fl uency), and because the WJR is well designed to provide mea- Fw (word fl uency). By the early 1990s, Horn had sures of the different concepts of human cognitive expanded the broad abilities which now num- capabilities--hat in the vernacular is referred to as intelligenc--that, so far, have been indicated by bered ten in all and included as many as 80 or so research. It provides measures of Gv, Ga, Gs, SAR, primary (narrow) abilities subsumed by them TSR, Gq, as well as Gf and Gc, whereas the collectively. The usual suspects remained, Gf, Wechsler scales provide only Gc, a Gf-Gv mixture, Gc, Gv, and Gs, but F had become TSR (fl uency and a very weak SAR and Gs. The composite of the WJR is thus a broad mixture relative to the of retrieval from long-term storage), and another composites of other, more narrow, batteries. But memory component had been identifi ed, SAR even when the same elements of a composite (short-term apprehension and retrieval). In addi- appears in other batteries, the elements enter in dif- tion, three new broad abilities were incorporated, ferent proportions to the whole: the composites of different published tests differ not only in breadth, one very similar to Gv but relying on auditory but also in the proportions of different abilities that stimuli and perception, Ga (auditory processing); contribute to the composite measure. A broad com- one very similar to Gf and Gc, but specifi c to posite relative to a narrow one needn’t necessarily mathematical or quantitative knowledge (Gq); correlate at a lower (or higher) level with other variables, but it may. It depends on breadth of the and one very similar to speed (Gs) but labeled variable with which the composites are correlated correct decision speed (CDS) to indicate rapidity and on how well the components of a predictor in providing correct, not merely quick, responses composite match the components of the predicted in relatively simple comprehension, reasoning, or variable. In prediction of job performance on many jobs, for example, as in much of the research problem-solving tasks. A fi nal factor related to reviewed by Schmidt and Hunter (1992 ), the broad reading and writing skills, Grw, was also added composite of the ASVAB (and it’s descendents) as it became more commonplace to include pub- predicts better than most narrower composites. lished tests of academic achievement right along- Enough said. These points are not highly debat- able. One can have a quibble here and there, but side cognitive ability tests (a distinction that Horn basically the evidence at this point in history indicated was only a semantic issue and not a true doesn’t permit much deviation from the conclusion difference). As Horn continued to advance the that there is no g; the emperor is naked. theory, it began to be referred to as “modern” Sincerely, John Horn Gf-Gc theory and later as “extended” Gf-Gc theory. 15 CHC Theory of Intelligence 219

Gf Gq GcGsm Gv Ga Glr Gs CDS Grw (Stratum II) Broad Abilities (Stratum I) Narrow Abilities

Fig. 15.3 The Cattell-Horn “extended” Gf-Gc model

Regardless of the name, a general factor supreme vanished—that “the clinician may fi nd it helpful to all other broad and narrow abilities was, of to ‘cross’ batteries to obtain a set of measure- course, absent in each refi nement of the Gf-Gc ments required for a particular assessment” model. Figure 15.3 provides an illustration of the (p. 252). Coupled with his three prior points on extended Gf-Gc model delineated by Horn late in appropriate psychometric principles that must his career (Horn and Blankson 2007; Horn and govern test data (i.e., knowledge of the factorial Noll 1997 ). composition of each subtest in a battery, forma- Around 1988, having a rather a theoretical test tion of broad ability composite or cluster scores of intelligence under his belt already (i.e., drawn from two similar but qualitatively distinct Woodcock-Johnson Psycho-Educational Battery; narrow ability indicators, and avoidance of sub- Woodcock and Johnson 1977 ), Richard Wood- tests with mixed-factor loading that complicate, cock journeyed to the University of Southern if not obviate, explanations of test performance), California likely in search of better theoretical the foundations of what would become CHC guidance. His collaboration with Horn resulted cross-battery assessment (McGrew and Flanagan in not only the fi rst published test specifi cally 1998 ; Flanagan and Ortiz 2001 ; Flanagan et al. designed to operationalize modern Gf-Gc theory, 2007 , 2013 ) are evident. Ironically, the sugges- it also prompted the publication of an often over- tion that batteries could be “crossed” was not a looked and important article that further estab- new idea. Cattell (1943 ) himself, commenting on lished the validity and utility of the theory, the vast number of published intelligence batter- particularly as a platform for test development ies generated over the preceding 20 years, noted (Woodcock 1990 ) that not only affi rmed the that “it is at fi rst glance a suffi ciently impressive empirical support for modern Gf-Gc theory window display; indeed, the adult-testing psy- across a variety of published instruments, but chometrist may be led to consider himself richly also reintroduced a concept that had long since equipped with 44 tests” (p. 154). Using a variety 220 S.O. Ortiz of tests culled from a range of individual batteries was relatively commonplace and even neces- CHC Theory and Carroll: sary precisely because they lacked any real Coming of Age theoretical foundation. According to Atwell et al. (1941 ), “the verbal items of the Wechsler- Adolescence is often a period of time where Bellevue Intelligence Scale, supplemented by a concerns about self-identity become salient. And vocabulary test and alternate arithmetic ques- as the Cattell-Horn modern Gf-Gc theoretical tions from the Alpha test would be satisfactory framework matured, an important event occurred … it was expected that the Beta Block Counting that provided an additional foundation regarding test, the 11, 12, and 14 year levels of the Porteus the evolution of CHC theory into a form more mazes, and the third, fourth, and sixth designs of closely resembling its present incarnation. That the Wechsler-Bellevue Block Design test would event was the publication of John Carroll’s be relatively useful” (p. 898). Both Cattell’s and (1993 ) remarkable book Human Cognitive Horn’s analyses in support of the theory of fl uid Abilities: A Survey of Factor-Analytic Studies . and crystallized intelligence relied on test data In short, Carroll undertook examination of over generated across a range of tests and batteries. 50 years worth of data drawn from tests of intel- For unknown reasons, however, clinical practice ligence and cognitive abilities to create a single, with testing had settled largely into a test-kit- massive database with which he could perform driven modality where, apart from some of the his own factor analyses. Carroll was insistent practices of neuropsychologists, intelligence was upon the use of exploratory rather than confi rma- measured and determined almost exclusively by tory techniques because he wished to allow the the structure of a single test—a practice fraught “data to speak for themselves.” Whether he went with peril, as Cattell, Horn, and Woodcock all in with any preconceptions about the structure of noted, because it meant that the resulting intelli- intelligence or not is unknown, although his gence was not equivalent to the prevailing con- eventual insistence and fi rm arguments in favor ceptualization of intelligence but rather “an of a general factor may offer some insight into average of whatever has been chosen by the test his a priori beliefs. Nevertheless, Carroll dis- author to be included in that battery” (p. 250). tilled some 2,000 available datasets published Whereas the fi eld in general might have lost sight between 1983 and 1985 (approximately) down of this principle, cautions remained regarding the to about 460 which he believed met critical crite- inappropriate use of IQ and inaccurate percep- ria necessary to permit his analyses and required tions of the construct as eloquently stated by for the sake of eliminating data that were less Salvia and Ysseldyke, “different intelligence than trustworthy on various psychometric and tests are simply samples of behaviors. For that theoretical grounds. This included datasets that reason it is wrong to speak of a person’s contained a substantial number of variables, a IQ. Instead, we can refer only to a person’s IQ on sizable sample (greater than 100), a published a specifi c test … Because the behavior samples correlation or covariance matrix, and suffi cient are different for different tests, one must always descriptions of the sample and variables to per- ask, ‘IQ on what test?.” (p. 158). Such wisdom mit interpretation (Carroll 1997 ). has gone unheeded too long in applied psychol- Carroll drew from Cattell (1971 ) in his use of ogy, but the popularity and rise of CHC theory the term “stratum” to describe the difference and the relatively quick mustering by test devel- between fi rst-, second-, and third-order factors. opers in aligning themselves with current intelli- Because different datasets may contain informa- gence theory have reawakened an interest in tion that only permits construction of only a fi rst- understanding this important idea while at the and second-order factor, there had been some same time helping to de-emphasize the utility confusion about whether a general factor was and value of global intellectual scores. indicated at the second level or at the third. 15 CHC Theory of Intelligence 221

g (stratum III) (stratum General ability General

Gf Gc GyGv Gu Gr Gs Gt (stratum II) (stratum Broad abilities Broad (stratum I) (stratum Narrow abilities Narrow

Fig. 15.4 Carroll’s three-stratum model

Carroll sought to minimize such confusion by Carroll (1993 ) noted specifi cally that “the Cattell- implementing Cattell’s stratum notion in ways Horn model … is a true hierarchical model cov- that clearly delineated the nature of the ability or ering all major domains of intellectual functioning factor identifi ed at each of the three levels. Thus, … among available models it appears to offer the Carroll (1997 ) named the “fi rst-order factors most well-founded and reasonable approach to resulting from analysis of typical sets of psycho- an acceptable theory of the structure of cognitive logical datasets factors at the fi rst stratum, or abilities” (p. 62). Despite this ringing endorse- stratum I factors ” (p. 124; emphasis in original). ment, Carroll outlined specifi c areas in which his Likewise, “stratum II factors were second-order proposed three-stratum model differed from the factors from such data sets, and stratum III factors Cattell-Horn formulation and other similar theo- were third-order factors from such datasets” retical models. In doing so, it may have fostered (p. 124) and postulated that any ability must fall the impression that he was indeed, despite his at one of these three levels. reservations, proposing a newer and better model Carroll’s ( 1997 ) ultimate intent was to provide which should, in logical fashion, supplant the less extensions to and expansions of existing theoreti- correct extended Gf-Gc framework. Whatever cal models by providing a “provisional statement the reason, Carroll’s formulation has become about the enumeration, identifi cation, and struc- known in its own right as the three-stratum model turing of the total range of cognitive abilities of intelligence. Figure 15.4 provides an illustra- known or discovered thus far” (p. 124). If taken at tion of the model that Carroll proposed on the face value, Carroll’s work provides signifi cant basis of his analyses. support for modern Gf-Gc theory, above and At the narrow ability level, Carroll’s ( 1993 ) beyond other conceptualizations of intelligence. analyses revealed about 65 primary, fi rst-order 222 S.O. Ortiz factors. These are specifi c tasks measured rela- factor (which should have less signifi cance than tively purely and cleanly by a single task, most fi rst- or second-order factors) is also intention- often couched as a subtest within a larger battery ally placed at the top providing a clear indication of tests. The sheer number of abilities at this level of its superiority within a hierarchical structure. precludes inclusion of their names and defi ni- Moreover, it is positioned to the left side right tions in their entirety, and the reader is referred to above Gf and Gc, as a way of indicating that other sources for such information (e.g., Carroll those abilities bear the closest relationship to it 1997). At the broad ability level, Carroll’s (1993 ) by virtue of being the most g-saturated. Of analyses revealed only eight broad abilities—a course, one could argue, as did Cattell, Horn, result that was not surprising since he likely con- Thurstone, and others, that the general factors fi ned himself to intelligence tests. If the nomen- are simply Gf anyway and a mere duplication of clature applied by Carroll is loosely interpreted, the second-order factor. In 2003, he asserted that the resemblance to modern (or extended) Gf-Gc “researchers who are concerned with this struc- theory was quite good. For example, the broad ture in one way or another…. can be assured that abilities of Gf, Gc, Gv, and Gs were equivalent in a general factor g exists, along with a series of both defi nition and labels. The Cattell-Horn second-order factors that measure broad special model used Ga for auditory processing, whereas abilities” (p. 19). It seems impossible, perhaps Carroll used the designation “Gu.” A greater dif- even needless, to elaborate more on this debate. ference was evident in Carroll’s use of “Gy” for It is suffi cient to recognize that it is in fact a general memory and learning, in contrast to the debate, and not to assume as so many have done Horn-Cattell use of SAR (which began being erroneously, that g is an established scientifi c referred to as Gsm), as well as in the delineation fact. Likewise, if one wishes to dispense with g of “Gr” or broad retrieval ability which was altogether, it should only be in service to the pur- called TSR (which also began being referred to poses of evaluation which neither need nor as Glr) in the extended Cattell-Horn Gf-Gc require it. The peril in this regard is to succumb model. And fi nally, Carroll labeled his speed fac- wholly to one position or the other when the real- tor “Gt” referring to processing speed and RT ity is that neither, as of yet, is supported by any (decision speed) as one unifi ed broad factor com- incontrovertible evidence that makes either posi- pared to the Cattell-Horn formulation that split tion defensible. If applied psychologists con- the broad factor into two components: Gs, pro- tinue to demand g , or global, full-scale scores cessing speed, and CDS, correct decision speed. intended to represent general intelligence, then The most signifi cant and obvious difference test publishers will no doubt continue to offer it. in Carroll’s three-stratum model and the extended Whatever the case, it should be recognized that Horn-Cattell Gf-Gc model rests with the concept aggregating scores to produce a general index is of a general factor. As already discussed, Cattell a simple enterprise, perhaps too simple, and that had envisioned a general factor comprised of two because it can be so easily constructed, it may distinct components, and Horn had dispensed prove seductive for use in ways that cannot be with the notion of a general factor from the out- supported psychometrically and theoretically.3 set. Neither saw any rationale nor need for it, and neither factored their data in a way that supported it. For them, it was a mathematical non- necessity, a statistical artifact, and an illusion 3 I once had the opportunity to ask Richard Woodcock why that simply was not worth pursuing. Carroll, on his WJ-R contained a global intellectual ability (GIA) the other hand, appeared to care deeply about a score when the test was operationalized according to general factor. In adopting a three-stratum model extended Gf-Gc theory that did not specify such a general vs. a two-stratum model, he made a deliberate factor. His response was both enlightening and discourag- ing as he admitted, almost apologetically and reluctantly choice to create a level in which the general fac- that he could not have gotten the test published without tor could exist. In his diagrams, this third-order including it. 15 CHC Theory of Intelligence 223

cross- battery factor analyses drawn from subtest CHC Theory and an Integrated data generated by the DAS, K-ABC, KAIT, Framework: Maturing in Adulthood SBIV, WISC-III, WPPSI-II, WAIS-III, and WJ-R. Given the similarities of the two models, it Despite the apparent differences between the would seem that such a task did not present many extended Cattell-Horn Gf-Gc model and Carroll’s diffi culties, but that would be an incorrect char- three-stratum model, as well as the irreconcilable acterization. First, many of the subtests and bat- issue of g or not to g , a rather unusual thing hap- teries McGrew included had never been factored pened in the late 1990s. Whereas it would have in accordance with Gf-Gc theory, as had the been logical to reason that intelligence theorists WJ-R subtests. Due to the confi rmatory nature of and their personal frameworks would fracture the the analyses, expert classifi cations had to be landscape to the point that none of them would be made for many tests to provide a foundation for suitable for guiding test development, the very testing the relations among them. Second, not all opposite occurred, and attempts were made to batteries measured all broad abilities, and thus, consolidate and integrate them in one overarch- data on some abilities was likely to be rather lim- ing theoretical framework. Why this may have ited (e.g., only the WJ-R provided measures of occurred can probably be traced to several fac- auditory processing). But as Cattell had noted, tors. For example, neither extended Gf-Gc nor the theory should be tested in its totality, and it the three-stratum model was well recognized or would be necessary to ensure that adequate cov- popular in the mainstream discussions or mea- erage in terms of both breadth of abilities measurement of intelligence, so not a lot of research- sured as well as depth of abilities existed or else ers, let alone practitioners, paid them much the results might prove rather limited. attention. In addition, the WJ-R was making McGrew’s initial analysis resulted in the strong inroads, however, and beginning to craft extraction of nine broad abilities (one less than out a place of its own among the now relatively Cattell-Horn’s and one more than Carroll’s) small pantheon of comprehensive intelligence including Gc, Gv, Gq, Grw, Gsm, Gf, Gs, Glr, tests that had boiled down, more or less, to the and Ga. Given the limitations of the tests he Wechsler and Stanford-Binet scales, Kaufman used, he was only able to identify about 36 nar- Assessment Battery for Children, and Differential row abilities in the dataset. Some, like Gsm, Gf, Abilities Scales. Yet it still lagged signifi cantly and Gs, contained only two identifi ed narrow behind the others, particularly the Wechsler abilities, whereas others like Gc, Gv, and Grw scales in use and acceptance among applied psy- subsumed numerous narrow abilities, seven, six, chologists. As noted previously, Kevin McGrew and seven, respectively. Of particular note is that and Dawn Flanagan adopted the principles set in many cases (9 out of the 36), only one subtest forth by Woodcock ( 1990 ) and outlined the foun- from one battery was present to provide data for dations of cross-battery assessment (often a given narrow ability. This is not so much a lim- referred to as XBA; Flanagan and McGrew 1997 ; itation of McGrew’s study but more an indict- McGrew and Flanagan 1998 ) which placed ment on the narrowness of the factor structures extended Gf-Gc theory at its core. But perhaps upon which the tests had been developed. Later the key circumstance during this period was versions of the model contain ten broad abilities McGrew’s reconciliation and integration of the that include Gt (reaction time/decision speed) Cattell-Horn Gf-Gc and Carroll’s three-stratum which subsumes three additional narrow abili- model as a guide to bridging intelligence theory ties, but it is sometimes omitted since it is not with the practice of intellectual assessment currently measured by any major intelligence (McGrew 1997 ). McGrew specifi cally outlined a battery. In addition, some researchers have called “proposed comprehensive Gf-Gc framework” for a separation of Grw into its constituent that he generated via joint confi rmatory components, using the designations Grw-R for 224 S.O. Ortiz reading skills and Grw-W for writing skills sonal communication with both Horn and Carroll (Flanagan and Ortiz 2001 ; Flanagan et al. 2007 , in 1999 and that they had recommended the name 2013 ). McGrew asserts that these factors do not and that it be structured according to order of distinguish themselves in factor analytic studies contribution. and tend to be very highly correlated. The asso- And so was born CHC theory, which by this ciation between the two is quite clear as they are point continued to provide the long sought-after both symbolic aspects of language development. common taxonomy of abilities and nomencla- But because schooling specifi cally treats them ture that permitted further advancement by and teaches them as independent skills, there is a ensuring that all researchers and all studies were practical advantage in assessment that comes fi nally talking about the same things. What was from making this distinction. However, it has not Gf in one study or on one test was now compa- yet been fully incorporated into current CHC rable and relatively equivalent to Gf in another theoretical specifi cations as it presents a condi- study or another test. It is no small point to say tion that cannot be supported empirically. that the viability and success and ever growing Once again, it appears that fortuitous events popularity of CHC theory is rooted largely in reopened the fl oodgates regarding development the fact that it has created a landscape that has and advancement of the theory. In hindsight, the enhanced theoretical and psychometric develop- integrated model was only just beginning to blos- ment by clearly delineating factors, tests that som. Consider that in 1985, Horn, Carroll, and measure those factors, tests that do not measure McGrew were called upon as consultants regard- them, and language common to all that reduces ing the revision of the 1978. Schneider and confusion and ambiguity. McGrew referred to McGrew (2012 ) hails this “meeting of the minds” Carroll’s work, upon which much of the nomen- as “a moment where the interests and wisdom of clature of the integrated CHC model is based, as a leading applied test developer (Woodcock), the the “Rosetta stone” of the fi eld of intelligence. leading proponent of Cattell-Horn Gf-Gc theory With an extensive history of development and (Horn), and one of the preeminent educational increasingly widespread use in current research psychologists and scholars of the factor analysis and development, the momentum of CHC theory of human abilities (Carroll) intersected” came is such that it may sweep up many other lesser into a perfect alignment of psychometric stars models in its wake. This fact has not gone unno- that served, in his opinion, as “the fl ash point that ticed by test publishers either in that the fi rst resulted in all subsequent theory-to-practice decade of the twenty-fi rst century saw a rapid bridging events leading to today’s CHC theory convergence on CHC theory or radical revisions and related assessment developments” (p. 144). to the factorial structure of tests and their sub- Following his later integration, McGrew teamed tests to the point that it was described, albeit with Woodcock and Nancy Mather in 1999 to facetiously, as a miraculous and “sudden collec- develop and subsequently publish the Woodcock- tive psychometric epiphany” (Flanagan et al. Johnson III (Woodcock et al. 2001 ). By this time, 2007, p. 221). the integrated model had taken on a new name Without question, CHC theory has had a sig- that McGrew cites as having fi rst been presented nifi cant impact in terms of promoting theoretical in the WJ-R III technical manual (McGrew and developments in the area of intelligence and cog- Woodcock 2001 ) where the full name “Cattell- nitive abilities going on nearly a full century now. Horn- Carroll theory of cognitive abilities” And despite getting an extremely late start with appears and is described as “an amalgamation of respect to infl uencing the design and structure of two similar theories about the content and struc- tests used in measuring human cognitive abili- ture of human cognitive abilities” (p. 9). McGrew ties, it has made impressive gains in just the last further notes that by amalgamation, and by impli- two decades and shows little sign of slowing cation, the name was partially derived from per- down any time soon. Figure 15.5 provides an 15 CHC Theory of Intelligence 225

Gf Gq GcGrw Gsm Gv Ga Glr Gs Gt (Stratum II) (Stratum Broad Abilities Broad (Stratum I) (Stratum Narrow Abilities Narrow

Fig. 15.5 CHC theory “early” integrated model

illustration of the “early” integrated CHC model well be overthrown by a better and more accu- as formulated by McGrew (1997 ). rate theory at some point in the future. CHC theory offers just one view of the nature and specifi cs of human cognitive abilities, including CHC Theory—Aging Gracefully, intelligence, but it need not be seen as the fi nal A Summary of Sorts word on the matter. Nevertheless, its value as a useful and guiding paradigm is well established With incredible modesty and recognition of the and continues to inspire both confi dence and absolute purity of the scientifi c process, note interest in what might be accomplished with with all due sincerity that “the extended theory further refi nement. Consider that in the very lat- of fl uid and crystallized (Gf and Gc) cognitive est incarnation of CHC theory (Schneider and abilities is wrong, of course, even though it may McGrew 2012 ), fully 16 broad abilities are now be the best account we current have of the orga- specifi ed along with 81 narrow abilities. In addi- nization and development of abilities thought to tion to the ten abilities described previously, be indicative of human intelligence” (p. 41). Schneider and McGrew outline six new ones They further assert that “all scientifi c theory is including Gkn (domain-specifi c knowledge), wrong. It is the job of science to improve theory. Go (olfactory ability), Gh (tactile ability), Gp That requires identifying what is wrong with it (psychomotor ability), Gk (kinesthetic ability), and fi nd out how to change it to make it more and Gps (psychomotor speed). Despite this ele- nearly correct” (p. 41). It is unlikely that the gant expansion, none of the aforementioned CHC theory will ever be modifi ed to the point broad abilities are measured by any current bat- that it becomes completely correct, and it may tery which precludes their measurement and 226 S.O. Ortiz

Acquired Memory & Speed & Reasoning Knowledge Learning Sensory Motor Efficiency

Gf Gc GknGq Grw Gsm Glr Gv Ga Go Gh Gp Gk Gs Gt Gps (Stratum II) Broad Abilities (Stratum I) Narrow Abilities

Fig. 15.6 CHC theory “current” integrated model

limits assessment to the ten abilities noted pre- specifi c relationships between cognitive and aca- viously. Likewise, of the 81 narrow abilities, demic abilities that are quite useful in psychoedu- only 38 (just two more than before) can be mea- cational evaluations (Schneider and McGrew sured and evaluated by current tests. A simpli- 2012 ). While the theory- practice gap that Cattell, fi ed illustration of this “current” version of the Horn, and McGrew intended to bridge via refi ne- integrated CHC model is presented in Fig. 15.6 . ments to their theoretical models remains even Interestingly, much of the current discussion today, it is now a combination of tests that have regarding CHC theory does so without much not yet caught up to theory coupled with theory attention to the notion of general intelligence. that continues to grow and refuses to sit still. In There does not appear to be any deliberate attempt that regard, Cattell, Horn, and Carroll would all to include or exclude g in the latest integrated likely take heart in knowing that their burning CHC theoretical models, and its relative absence interest in furthering an understanding of intelli- may be more of a refl ection regarding interest in gence and human cognitive abilities has been broad and narrow abilities and their ability to pre- passed on intact to the current generation of psy- dict well specifi c areas of academic skill develop- chological scientists. CHC theory has grown up, ment above and beyond what is captured solely and it would be safe to assume that Cattell, Horn, by g (Keith and Reynolds 2010 ). Because the pre- and Carroll would likely be very proud to know ceding discussion has made it clear that g is nei- that their goals, their ideas, and their passion are ther necessary nor desirable in all cases, and still alive and kicking in CHC theory to the pres- because no test publisher is likely to release a test ent day. without including some global score, it will remain an option for any assessor who wishes to use it. Therefore, its inclusion within current the- References ory is not as compelling as it once might have Atwell, C. R., Bloomberg, W., & Wells, F. L. (1941). been, especially as research into the narrow abili- Psychometrics at an army induction center. New ties continues to demonstrate important and England Journal of Medicine, 224 , 898–899. 15 CHC Theory of Intelligence 227

Carroll, J. B. (1993). Human cognitive abilities: A survey Horn, J. L., & Noll, J. (1997). Human cognitive capabili- of factor-analytic studies. New York: Cambridge ties: Gf-Gc theory. In D. P. Flanagan, J. L. Genshaft, & University Press. P. L. Harrison (Eds.), Contemporary intellectual Carroll, J. B. (1997). Commentary on Keith and Witta’s assessment (pp. 53–91). New York: Guilford Press. hierarchical and cross-age confi rmatory factor analy- Keith, T. Z., & Reynolds, M. R. (2010). Cattell-Horn- sis of the WISC-III. School Psychology Quarterly, 12 , Carroll abilities and cognitive tests: What we’ve 108–109. learned from 20 years of research. Psychology in the Cattell, R. B. (1941). Some theoretical issues in adult Schools, 47 (7), 635–650. intelligence testing. Psychological Bulletin, 38 , 592. McGrew, K. S. (1997). Analysis of the major intelligence Cattell, R. B. (1943). The measurement of adult intelli- batteries according to a proposed comprehensive gence. Psychological Bulletin, 2 , 153–193. Gf-Gc framework. In D. P. Flanagan, J. L. Genshaft, & Cattell, R. B. (1950). Personality: A systematic theoretical P. L. Harrison (Eds.), Contemporary intellectual and factual study . New York: McGraw-Hill. assessment (pp. 151–179). New York: Guilford Press. Cattell, R. B. (1957a). Personality and motivation struc- McGrew, K. S., & Flanagan, D. P. (1998). The Intelligence ture and measurement. New York: Harcourt, Brace & Test Desk Reference (ITDR): Gf-Gc cross-battery World. assessment . Boston: Allyn & Bacon. Cattell, R. B. (1957b). The IPAT culture fair intelligence McGrew, K. S., & Woodcock, R. W. (2001). Woodcock- scales . Champaign: Institute for Personality and Johnson III technical manual . Itasca: Riverside. Ability Testing. McNemar, Q. (1964). Lost: Our intelligences. Why? Cattell, R. B. (1963). Theory of fl uid and crystallized American Psychologist, 19 , 871–872. intelligence: A critical experiment. Journal of Rimoldi, H. J. (1948). Study of some factors related to Educational Psychology, 54 (1), 1–22. intelligence. Psychometrika, 13 , 27–46. Cattell, R. B. (1971). Abilities: Their structure, growth, Schmidt, F. L., & Hunter, J. E. (1992). Development of a and action . Boston: Houghton Miffl in. causal model of processes determining job El Koussy, A. A. H. (1935). The visual perception of performance. Current Directions in Psychological space. British Journal of Psychology, Monograph Science, 1 , 89–92. Supplement, 20 , 1–80. Schneider, J., & McGrew, K. S. (2012). The Cattell-Horn- Flanagan, D. P., & McGrew, K. S. (1997). A cross-battery caroll model of Intelligence In D. P. Flanagan & P. L. approach to assessing and interpreting cognitive abili- Harrison (Eds.), Contemporary intellectual assess- ties: Narrowing the gap between practice and science. ment (3rd ed.). (pp. 99–143) New York: Guilford. In D. P. Flanagan, J. L. Genshaft, & P. L. Harrison Thurstone, L. L. (1946, February). Theories of intelli- (Eds.), Contemporary intellectual assessment gence. Scientifi c Monthly (pp. 101–112). (pp. 314–325). New York: Guilford Press. Thurstone, L. L., & Thurstone, T. G. (1941). Factorial Flanagan, D. P., & Ortiz, S. O. (2001). Essentials of cross- studies of intelligence (Psychometric Monographs, battery assessment . New York: Wiley. No. 2) . Chicago: University of Chicago Press. Flanagan, D. P., Ortiz, S. O., & Alfonso, V. C. (2007). Undheim, J. O. (1976). Ability structure in 10-11-year-old Essentials of cross-battery assessment (2nd ed.). children and the theory of fl uid and crystallized intel- New York: Wiley. ligence. Journal of Educational Psychology, 68 , Flanagan, D. P., Ortiz, S. O., & Alfonso, V. C. (2013). 411–423. Essentials of cross-battery assessment (3rd ed.). Undheim, J. O., & Gustafsson, J. E. (1987). The hierarchi- New York: Wiley. cal organization of cognitive abilities: Restoring gen- Gustafsson, J. E. (1984). A unifying model for the struc- eral intelligence through the use of linear structural ture of intellectual abilities. Psychological Bulletin, relations (LISREL). Multivariate Behavioral 61 , 401–404. Research, 22 , 149–171. Horn, J. L. (1965). Fluid and crystallized intelligence: A Wechsler, D. (1939). The measurement of adult intelli- factor analytic and developmental study of the structure gence . Baltimore: Williams & Wilkins. among primary mental abilities. Unpublished doctoral Wells, F. L. (1932). Army alpha revised. Journal of dissertation, University of Illinois, Champaign. Personality, 10 , 411–417. Horn, J. L., & Blankson, N. (2007). Foundations for a bet- Woodcock, R. W. (1990). Theoretical foundations of the ter understanding of cognitive abilities. In D. P. WJ-R measures of cognitive ability. Journal of Flanagan & P. L. Harrison (Eds.), Contemporary intel- Psychoeducational Assessment, 8 , 231–258. lectual assessment (2nd ed., pp. 41–68). New York: Woodcock, R. W., & Johnson, M. B. (1977). The Guilford Press. Woodcock-Johnson Psycho-Educational Battery . Horn, J. L., & Cattell, R. B. (1966a). Age differences in Chicago: Riverside Publishing. fl uid and crystallized intelligence. Acta Psychologica, Woodcock, R. W., & Johnson, M. B. (1989). The 26 , 107–129. Woodcock-Johnson Psycho-Educational Battery— Horn, J. L., & Cattell, R. B. (1966b). Age differences in Revised . Chicago: Riverside Publishing. primary mental abilities. Journal of Gerontology, 21 , Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). 210–220. Woodcock-Johnson III . Itasca: Riverside Publishing. Multiple Intelligences in the New Age of Thinking 16

Robert J. Sternberg

In this chapter, I describe two theories that involve the idea of “multiple intelligences.” The ﬁ rst is the theory of multiple intelligence (MI theory—Gardner 1983 , 1993 , 2006 ). The second is the theory of successful intelligence (Sternberg 1997 , 2003 , 2005 , 2009 , 2010 ).

other people; intrapersonal intelligence, used to Multiple Intelligences Theory understand oneself; and naturalist intelligence, used to understand the natural world. Howard Gardner (1983 , 1993 , 2006 ) has pro- Each intelligence is a separate system of func- posed a theory of multiple intelligences, accord- tioning, although these systems can interact to ing to which intelligence comprises multiple produce what we see as intelligent performance. independent constructs, not just a single, unitary Looking at Gardner’s list of intelligences, you construct. However, instead of speaking of mul- might want to evaluate your own intelligences, tiple abilities that together constitute intelligence perhaps rank ordering your strengths in each. (e.g., Thurstone 1938), this theory distinguishes In some respects, Gardner’s theory sounds eight distinct intelligences that are relatively like a factorial one. It speciﬁ es several abilities independent of each other. that are construed to reﬂ ect intelligence of some The multiple intelligences are linguistic intel- sort. However, Gardner views each ability as a ligence, used to read, write, speak, and listen to separate intelligence, not just as a part of a single speech; logical-mathematical intelligence, used whole. Moreover, a crucial difference between to solve mathematical and logical problems; spa- Gardner’s theory and factorial ones is in the tial intelligence, used to imagine how objects sources of evidence Gardner used for identifying would look if they were rotated or otherwise dis- the eight intelligences. Gardner used converging placed in space; musical intelligence, used to operations, gathering evidence from multiple compose, play, and appreciate music; bodily- sources and types of data. kinesthetic intelligence, used to coordinate one- In particular, the theory uses eight “signs” as self and to participate successfully in athletics; criteria for detecting the existence of a discrete interpersonal intelligence, used to understand kind of intelligence (Gardner 1983 , pp. 63–67): 1. Potential isolation by brain damage. The destruction or sparing of a discrete area of the * R. J. Sternberg ( ) brain (e.g., areas linked to verbal aphasia) Department of Human Development, B44 MVR , Cornell University , Ithaca , NY 14850 , USA may destroy or spare a particular kind of intel- e-mail: [email protected] ligent behavior.

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 229 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_16, © Springer Science+Business Media New York 2015 230 R.J. Sternberg

2. The existence of exceptional individuals (e.g., intelligences. Gardner has speculated as to at least musical or mathematical prodigies). They some of these locales. But hard evidence for the demonstrate extraordinary ability (or defi cit) existence of these separate intelligences has yet to in a particular kind of intelligent behavior. be produced. Furthermore, there is no real evidence 3. An identifi able core operation or set of opera- for the strict modularity of Gardner’s theory. tions (e.g., detection of relationships among Consider the phenomenon of preserved specifi c musical tones). It is essential to performance cognitive functioning in autistic savants. Savants are of a particular kind of intelligent behavior. people with severe social and cognitive defi cits but 4. A distinctive developmental history leading with corresponding high ability in a narrow domain. from novice to master. It is accompanied by dis- They suggest that such preservation fails as evi- parate levels of expert performance (i.e., varying dence for modular intelligences. The narrow long- degrees of expressing this type of intelligence). term memory and specifi c aptitudes of savants may 5. A distinctive evolutionary history. Increases not really be intelligent. Thus, there may be reason in intelligence plausibly may be associated to question the intelligence of infl exible modules. with enhanced adaptation to the environment. I do not detail this theory further because there 6. Supportive evidence from cognitive- has been no empirical evidence collected since experimental research. An example would be MI theory was proposed that validates the theory task-specifi c performance differences across as a whole and the one extensive study that has discrete kinds of intelligence (e.g., visuospatial been done yielded results inconsistent with it tasks versus verbal tasks). They would need to (Visser et al. 2006 ). be accompanied by cross-task performance similarities within discrete kinds of intelligence (e.g., mental rotation of visuospatial imagery The Triarchic Theory of Successful and recall memory of visuospatial images). Intelligence 7. Supportive evidence from psychometric tests indicating discrete intelligences (e.g., differ- The Nature of Intelligence ing performance on tests of visuospatial abilities versus on tests of linguistic abilities). There are many defi nitions of intelligence, 8. Susceptibility to encoding in a symbol system although intelligence is typically defi ned in terms (e.g., language, math, musical notation) or in of a person’s ability to adapt to the environment a culturally devised arena (e.g., dance, athlet- and to learn from experience (Inteligence and its ics, theater, engineering, or surgery as cultur- Measurement: A Symposium, 1921; Sternberg ally devised expressions of bodily-kinesthetic and Detterman 1986 ). The defi nition of intelli- intelligence). gence here is somewhat more elaborate and is Gardner does not dismiss entirely the use of based on my (Sternberg 1984 , 1997 , 1998 , 1999b , psychometric tests. But the base of evidence used 2000 , 2003) theory of successful intelligence. by Gardner does not rely on the factor analysis of According to this defi nition, (successful) intelli- various psychometric tests alone. In thinking gence is (1) the ability to achieve one’s goals in about your own intelligences, how fully inte- life, given one’s sociocultural context; (2) by capi- grated do you believe them to be? How much do talizing on strengths and correcting or compensat- you perceive each type of intelligence as depending for weaknesses; (3) in order to adapt to, shape, ing on any of the others? and select environments; (4) through a combina- Gardner’s view of the mind is modular. tion of analytical, creative, and practical abilities. Modularity theorists believe that different abili- In recent years, I have emphasized that intelligence ties—such as Gardner’s intelligences—can be iso- best serves individuals and societies when it is lated as emanating from distinct portions or modules augmented by wisdom (Sternberg 2003 , 2008 ). of the brain. Thus, a major task of existing and According to the proposed theory of human future research on intelligence is to isolate the intelligence and its development (Sternberg portions of the brain responsible for each of the 1997 , 1999a , 2003 , 2004 , 2009 ), a common set 16 Multiple Intelligences in the New Age of Thinking 231 of processes underlies all aspects of intelligence. ing is invoked when the components are applied These processes are hypothesized to be univer- to experience to adapt to, shape, and select envi- sal. For example, although the solutions to prob- ronments. One needs creative skills and disposi- lems that are considered intelligent in one culture tions to generate ideas, analytical skills and may be different from the solutions considered dispositions to decide if they are good ideas, and to be intelligent in another culture, the need to practical skills and dispositions to implement defi ne problems and translate strategies to solve one’s ideas and to convince others of their worth. these problems exists in any culture. Even within Because the theory of successful intelligence cultures, there may be differences in what differ- comprises three subtheories—a componential ent groups mean by intelligence (Okagaki and subtheory dealing with the components of intelli- Sternberg 1993; Sternberg 1985b ). gence, an experiential subtheory dealing with the Metacomponents, or executive processes, plan importance of coping with relative novelty and of what to do, monitor things as they are being done, automatization of information processing, and a and evaluate things after they are done. Examples contextual subtheory dealing with processes of of metacomponents are recognizing the existence adaptation, shaping, and selection—the theory has of a problem, defi ning the nature of the problem, been referred to from time to time as triarchic. deciding on a strategy for solving the problem, Intelligence is not, as Edwin Boring (1923 ) monitoring the solution of the problem, and eval- once suggested, merely what intelligence tests uating the solution after the problem is solved. test. Intelligence tests and other tests of cognitive Performance components execute the instruc- and academic skills measure part of the range of tions of the metacomponents. For example, infer- intellectual skills. They do not measure the whole ence is used to decide how two stimuli are related range. One should not conclude that a person who and application is used to apply what one has does not test well is not smart. Rather, one should inferred (Sternberg 1977 ). Other examples of perfor- merely look at test scores as one indicator among mance components are comparison of stimuli, justi- many of a person’s intellectual skills. Moreover, fi cation of a given response as adequate although not the kinds of skills posited by hierarchical theories ideal, and actually making the response. (e.g., Carroll 1993 ; Cattell 1971 ; Vernon 1971 ) are Knowledge-acquisition components are used to viewed only as a subset of the skills important in a learn how to solve problems or simply to acquire broader conception of intelligence. declarative knowledge in the fi rst place (Sternberg 1985a ). Selective encoding is used to decide what information is relevant in the context of one’s learn- The Assessment of Successful ing. Selective comparison is used to bring old infor- Intelligence mation to bear on new problems. And selective combination is used to put together the selectively Our assessments of intelligence have been orga- encoded and compared information into a single nized around the analytical, creative, and practi- and sometimes insightful solution to a problem. cal aspects of it. We discuss those assessments Although the same processes are used for all here, singly and collectively. three aspects of intelligence universally, these processes are applied to different kinds of tasks and situations depending on whether a given Analytical Intelligence problem requires analytical thinking, creative thinking, practical thinking, or a combination of Analytical intelligence is involved when the these kinds of thinking. In particular, analytical information-processing components of intelli- thinking is invoked when components are applied gence are applied to analyze, evaluate, judge, or to fairly familiar kinds of problems abstracted compare and contrast. It typically is involved from everyday life. Creative thinking is invoked when components are applied to relatively famil- when the components are applied to relatively iar kinds of problems where the judgments to be novel kinds of tasks or situations. Practical think- made are of a fairly abstract nature. 232 R.J. Sternberg

In some early work, it was shown how analytical although children generally became quicker in kinds of problems, such as analogies or syllo- information processing with age, not all compo- gisms, can be analyzed componentially (Guyote nents were executed more rapidly with age and Sternberg 1981; Sternberg 1977 , 1980b , (Sternberg and Rifkin 1979 ). The encoding com- 1983; Sternberg and Gardner 1983 ), with ponent fi rst showed a decrease in component response times or error rates decomposed to yield time with age and then an increase. Apparently, their underlying information-processing compo- older children realized that their best strategy nents. The goal of this research was to understand was to spend more time in encoding the terms of the information-processing origins of individual a problem so that they later would be able to differences in (the analytical aspect of) human spend less time in operating on these encodings. intelligence. With componential analysis, one A related, third fi nding was that better reasoners could specify sources of individual differences tend to spend relatively more time than do poorer underlying a factor score such as that for “induc- reasoners in global, up-front metacomponential tive reasoning.” For example, response times on planning, when they solve diffi cult reasoning analogies (Sternberg 1977 ) and linear syllogisms problems. Poorer reasoners, on the other hand, (Sternberg 1980a ) were decomposed into their tend to spend relatively more time in local plan- elementary performance components. The gen- ning (Sternberg 1981 ). Presumably, the better eral strategy of such research is to (a) specify an reasoners recognize that it is better to invest information-processing model of task perfor- more time up front so as to be able to process a mance; (b) propose a parameterization of this problem more effi ciently later on. Fourth, it also model, so that each information-processing was found in verbal analogical reasoning that as component is assigned a mathematical parameter children grew older, their strategies shifted so that corresponding to its latency (and another corre- they relied on word association less and abstract sponding to its error rate); and (c) construct cog- relations more (Sternberg and Nigro 1980 ). nitive tasks administered in such a way that it is In the componential analysis work described possible through mathematical modeling to iso- above, correlations were computed between late the parameters of the mathematical model. component scores of individuals and scores on In this way, it is possible to specify, in the solving tests of different kinds of psychometric abilities. of various kinds of problems, several sources of First, in the studies of inductive reasoning important individual or developmental differ- (Sternberg 1977; Sternberg and Gardner 1983 ), ences: (1) What performance components are it was found that although inference, mapping, used? (2) How long does it take to execute each application, comparison, and justifi cation tended component? (3) How susceptible is each compo- to correlate with such tests, the highest correla- nent to error? (4) How are the components com- tion typically was with the preparation-response bined into strategies? (5) What are the mental component. This result was puzzling at fi rst, representations upon which the components act? because this component was estimated as the Research on the components of human regression constant in the predictive regression intelligence yielded some interesting results. equation. This result ended up giving birth to the Consider some examples. First, execution of concept of the metacomponents: higher-order early components (e.g., inference and mapping) processes used to plan, monitor, and evaluate task tends exhaustively to consider the attributes of performance. Second, it was also found that the the stimuli, whereas execution of later compo- correlations obtained for all the components nents (e.g., application) tends to consider the showed convergent-discriminant validation: They attributes of the stimuli in self-terminating fash- tended to be signifi cant with psychometric tests ion, with only those attributes processed that are of reasoning but not with psychometric tests of essential for reaching a solution (Sternberg perceptual speed (Sternberg 1977 ; Sternberg and 1977 ). Second, in a study of the development of Gardner 1983 ). Third, signifi cant correlations fi gural analogical reasoning, it was found that with vocabulary tended to be obtained only for 16 Multiple Intelligences in the New Age of Thinking 233 encoding of verbal stimuli (Sternberg 1977 ; Or they might be told of four kinds of people on Sternberg and Gardner 1983 ). Fourth, it was the planet Kyron, blens, who are born young and found in studies of linear-syllogistic reasoning die young; kwefs, who are born old and die old; (e.g., John is taller than Mary; Mary is taller balts, who are born young and die old; and pros- than Susan; who is tallest? ) that components of ses, who are born old and die young (Sternberg the proposed (mixed linguistic-spatial) model 1982 ; Tetewsky and Sternberg 1986 ). Their task that were supposed to correlate with verbal abil- was to predict future states from past states, given ity did so and did not correlate with spatial incomplete information. In another set of studies, ability; components that were supposed to cor- 60 people were given more conventional kinds of relate with spatial ability did so and did not inductive reasoning problems, such as analogies, correlate with verbal ability. In other words, it series completions, and classifi cations, but were was possible successfully to validate the proposed told to solve them. But the problems had prem- model of linear-syllogistic reasoning not only in ises preceding them that were either conventional terms of the fi t of response-time or error data to (dancers wear shoes) or novel (dancers eat shoes). the predictions of the alternative models but also The participants had to solve the problems as in terms of the correlations of component scores though the counterfactuals were true (Sternberg with psychometric tests of verbal and spatial abil- and Gastel 1989a , b ). ities (Sternberg 1980a ). Fifth and fi nally, it was In these studies, we found that correlations found that there were individual differences in with conventional kinds of tests depended on how strategies in solving linear syllogisms, whereby novel or non-entrenched the conventional tests some people used a largely linguistic model, oth- were. The more novel are the items, the higher are ers a largely spatial model, and most the proposed the correlations of our tests with scores on succes- linguistic- spatial mixed model. Thus, sometimes, sively more novel conventional tests. Thus, the less than perfect fi t of a proposed model to group components isolated for relatively novel items data may refl ect individual differences in strate- would tend to correlate more highly with more gies among participants. unusual tests of fl uid abilities (e.g., that of Cattell and Cattell 1973 ) than with tests of crystallized abilities. We also found that when response times Creative Intelligence on the relatively novel problems were componentially analyzed, some components better mea- Intelligence tests contain a range of problems, sured the creative aspect of intelligence than some of them more novel than others. In some of did others. For example, in the “grue-bleen” task the componential work, we have shown that when mentioned above, the information- processing one goes beyond the range of unconventionality component requiring people to switch from of the conventional tests of intelligence, one starts conventional green-blue thinking to grue-bleen to tap sources of individual differences measured thinking and then back to green-blue thinking little or not at all by the tests. According to the again was a particularly good measure of the theory of successful intelligence, (creative) intel- ability to cope with novelty. ligence is particularly well measured by problems In our original work with divergent reasoning assessing how well an individual can cope with problems having no one best answer, we asked 63 relative novelty (see Sternberg et al. 2002 ). people to create various kinds of products (Lubart We presented 80 individuals with novel kinds and Sternberg 1995 ; Sternberg and Lubart 1991 , of reasoning problems that had a single best 1995 , 1996 ) where an infi nite variety of responses answer. For example, they might be told that were possible. Individuals were asked to create some objects are green and others blue; but still products in the realms of writing, art, advertising, other objects might be grue, meaning green until and science. In writing, they were asked to write the year 2000 and blue thereafter, or bleen, mean- very short stories for which we would give them ing blue until the year 2000 and green thereafter. a choice of titles, such as “Beyond the Edge” or 234 R.J. Sternberg

“The Octopus’s Sneakers.” In art, the participants the components of intelligence to experience so as were asked to produce art compositions with to (a) adapt to, (b) shape, and (c) select environ- titles such as “The Beginning of Time” or “Earth ments. People differ in their balance of adapta- from an Insect’s Point of View.” In advertising, tion, shaping, and selection and in the competence they were asked to produce advertisements for with which they balance among the three possible products such as a brand of bow tie or a brand of courses of action. doorknob. In science, they were asked to solve Much of our work on practical intelligence problems such as one asking them how people has centered on the concept of tacit knowledge. might detect extraterrestrial aliens among us who We have defi ned this construct as what one needs are seeking to escape detection. Participants cre- to know in order to work effectively in an envi- ated two products in each domain. ronment that one is not explicitly taught and that First, we found that creativity comprises the often is not even verbalized (Sternberg et al. components proposed by their investment model 2000; Sternberg and Hedlund 2002 ; Sternberg of creativity: intelligence, knowledge, thinking and Wagner 1993 ; Sternberg et al. 1993 ; styles, personality, and motivation. Second, we Sternberg et al. 1995; Wagner 1987; Wagner and found that creativity is relatively although not Sternberg 1986 ; Will iams et al. 2002 ). We repre- wholly domain-specifi c. Correlations of ratings sent tacit knowledge in the form of production of the creative quality of the products across systems or sequences of “if-then” statements that domains were lower than correlations of ratings describe procedures one follows in various kinds and generally were at about the .4 level. Thus, of everyday situations. there was some degree of relation across domains We typically have measured tacit knowledge at the same time that there was plenty of room for using work-related problems that present prob- someone to be strong in one or more domains but lems one might encounter on the job. We have not in others. Third, we found a range of correla- measured tacit knowledge for both children and tions of measures of creative performance with adults, and among adults, for people in over two- conventional tests of abilities. As was the case for dozen occupations, such as management, sales, the correlations obtained with convergent prob- academia, teaching, school administration, secre- lems, correlations were higher to the extent that tarial work, and the military. In a typical tacit- problems on the conventional tests were non- knowledge problem, people are asked to read a entrenched. For example, correlations were story about a problem someone faces and to rate, higher with fl uid than with crystallized ability for each statement in a set of statements, how tests, and correlations were higher, the more adequate a solution the statement represents. novel the fl uid test was. These results suggest that In the tacit-knowledge studies, fi rst we have tests of creative intelligence have some overlap found that practical intelligence as embodied in with conventional tests (e.g., in requiring verbal tacit knowledge increases with experience, but it skills or the ability to analyze one’s own ideas— is profi ting from experience, rather than experi- Sternberg and Lubart 1995) but also tap skills ence per se, that results in increases in scores. beyond those measured even by relatively novel Some people could have been in a job for years kinds of items on the conventional tests of and still have acquired relatively little tacit knowl- intelligence. edge. Second, we also have found that subscores on tests of tacit knowledge—such as for managing oneself, managing others, and managing Practical Intelligence tasks—correlate signifi cantly with each other. Third, scores on various tests of tacit knowledge, Practical intelligence involves individuals apply- such as for academics and managers, are also cor- ing their abilities to the kinds of problems that related fairly substantially (at about the .5 level) confront them in daily life, such as on the job or in with each other. Thus, fourth, tests of tacit knowl- the home. Practical intelligence involves applying edge may yield a general factor across these tests. 16 Multiple Intelligences in the New Age of Thinking 235

However, fi fth, scores on tacit-knowledge tests do for more examples of cultural work relevant to not correlate with scores on conventional tests of the theory). The test of practical intelligence intelligence, whether the measures used are sin- measured children’s informal tacit knowledge gle-score measures of multiple-ability batteries. for natural herbal medicines that the villagers Thus, any general factor from the tacit-knowledge believe can be used to fi ght various types of tests is not the same as any general factor from infections. tests of academic abilities (suggesting that neither We found no correlation between the test of kind of g factor is truly general but rather general indigenous tacit knowledge and scores on the only across a limited range of measuring instru- fl uid-ability tests. But to our surprise, we found ments). Sixth, despite the lack of correlation of statistically signifi cant correlations of the tacit- practical- intellectual with conventional measures, knowledge tests with the tests of crystallized the scores on tacit-knowledge tests predict perfor- abilities. The correlations, however, were negative. mance on the job as well as or better than do con- In other words, the higher the children scored on ventional psychometric intelligence tests. the test of tacit knowledge, the lower they scored, Seventh, in one study done at the Center for on average, on the tests of crystallized abilities. Creative Leadership, we further found that scores We have considered each of the aspects of on our tests of tacit knowledge for management intelligence separately. How do they fare when were the best single predictor of performance on a they are assessed together? managerial simulation. In a hierarchical regression, scores on conventional tests of intelligence, personality, styles, and interpersonal orientation All Three Aspects of Intelligence were entered fi rst and scores on the test of tacit Together knowledge were entered last. Scores on the test of tacit knowledge were the single best predictor of Internal Validity Studies Several separate managerial simulation score. Moreover, these factor-analytic studies support the internal scores also contributed signifi cantly to the predic- validity of the theory of successful intelligence. tion even after everything else was entered fi rst In one study (Sternberg et al. 1999), we used into the equation. the so-called Sternberg Triarchic Abilities Test Eighth, in work on military leadership (STAT—Sternberg 1993 ) to investigate the (Hedlund et al. 2003 ; Sternberg and Hedlund internal validity of the theory. Three hundred 2002 ; Sternberg et al. 2000 ), it was found that twenty-six high-school students, primarily from scores of 562 participants on tests of tacit knowl- diverse parts of the United States, took the test, edge for military leadership predicted ratings of which comprised 12 subtests in all. There were leadership effectiveness, whereas scores on a four subtests each measuring analytical, cre- conventional test of intelligence and on a tacit- ative, and practical abilities. For each type of knowledge test for managers did not signifi cantly ability, there were three multiple-choice tests predict the ratings of effectiveness. In work with and one essay test. The multiple-choice tests, in Eskimos (Grigorenko et al. 2004 ), it was found turn, involved, respectively, verbal, quantitative, that low achievers in school can have exception- and fi gural content. ally high practical adaptive skills at home. Confi rmatory factor analysis on the data was Even stronger results have been obtained supportive of the triarchic theory of human overseas. In a study in Usenge, Kenya, near the intelligence, yielding separate and uncorrelated town of Kisumu, we were interested in school- analytical, creative, and practical factors. The age children’s ability to adapt to their indige- lack of correlation was due to the inclusion of nous environment. We devised a test of practical essay as well as multiple-choice subtests. intelligence for adaptation to the environment Although multiple- choice tests tended to corre- (see Sternberg and Grigorenko 1997 ; Sternberg, late substantially with multiple-choice tests, Nokes et al. 2001b ; and Sternberg 2004 , 2007 their correlations with essay tests were much 236 R.J. Sternberg weaker. The multiple-choice analytical subtest admit undergraduate students to Tufts University. loaded most highly on the analytical factor, but Each year, all 15,000+ applicants are given a the essay creative and practical subtests loaded selection of essays assessing analytical, creative, most highly on their respective factors. Thus, practical, and also wisdom-based skills. The measurement of creative and practical abilities applicants have the option of completing one of probably ideally should be accomplished with the essays, and then the analytical, creative, other kinds of testing instruments that comple- practical, and wisdom-based skills demonstrated ment multiple-choice instruments. through these essays and other aspects of the application are rated. External Validity Studies We have also looked The exact Kaleidoscope prompts vary from at the external validity of tests assessing year to year (see Sternberg 2010 for a complete successful intelligence. list through 2009). The questions differ in the skills they emphasize. No question is a “pure” The Rainbow Project In a study supported by measure of any single component of successful the College Board (Sternberg and the Rainbow intelligence. Scoring of the exercises is holistic Project Collaborators 2006 ), we used an expanded and is completed by admissions offi cers using set of tests on 1,015 students at 15 different rubrics with which they were provided by the institutions (13 colleges and 2 high schools). Our Center for the Psychology of Abilities, goal was not to replace the SAT but to devise tests Competencies, and Expertise at Tufts (PACE that would supplement the SAT, measuring skills Center). We have found that with training, admis- that this test does not measure. In addition to the sions offi cers can achieve good inter-rater reli- multiple-choice STAT tests described earlier, we ability (consistency) in their evaluations. used 3 additional measures of creative skills and The results at Tufts illustrated that a highly 3 of practical skills: selective college can introduce an “unconven- tional” exercise into its undergraduate admis- Creative Skills The three additional tests were sions process without disrupting the quality of the captioning cartoons, writing creative short stories entering class. It is important to underscore using two of a number of suggested titles, and the point that academic achievement has always orally telling creative stories based on a picture. been and remains the most important dimension of Tufts’ undergraduate admissions process. Practical Skills The three additional tests were Since we introduced the Kaleidoscope pilot in everyday situational judgments based on movie 2006, applications have remained roughly steady scenarios, a common-sense questionnaire based or increased slightly, and the mean SAT scores on problems found in work life, and a common- of accepted and enrolling students increased to sense questionnaire based on problems confronted new highs. In addition, we have not detected sta- in school. tistically meaningful ethnic group differences on We found that our tests signifi cantly and sub- the Kaleidoscope measures. Controlling for the stantially improved upon the validity of the SAT academic rating given to applicants by admis- for predicting fi rst-year college grades (Sternberg sions offi cers (which combines information from and the Rainbow Project Collaborators 2006 ). the transcript and standardized tests), students The test also improved equity: using the test to rated for Kaleidoscope achieved signifi cantly admit a class would result in greater ethnic diver- higher academic averages in their undergraduate sity than would using just the SAT or just the SAT work than students who were not so rated by the and grade-point average. admissions staff. In addition, research found that students with higher Kaleidoscope ratings were The Kaleidoscope Project The Kaleidoscope more involved in, and reported getting more out Project (Sternberg 2005 , 2010; Sternberg et al. of, extracurricular, active-citizenship and leader- 2012 ) has been used over the past 5 years to ship activities in their fi rst year at Tufts. 16 Multiple Intelligences in the New Age of Thinking 237

In sum, as Tufts seeks to identify and develop condition that emphasized either memory, ana- new leaders for a changing world, Kaleidoscope lytical, creative, or practical instruction. For provides a vehicle to help identify the potential example, in the memory condition, they might be leaders who may be best positioned to make a asked to describe the main tenets of a major the- positive and meaningful difference to the world ory of depression. In the analytical condition, in the future. In the fast-paced, data-driven atmo- they might be asked to compare and contrast two sphere of highly competitive college admissions, theories of depression. In the creative condition, Kaleidoscope validates the role of qualitative they might be asked to formulate their own the- measures of student ability and excellence. ory of depression. In the practical condition, they might be asked how they could use what they had learned about depression to help a friend who Instruction for Successful was depressed. Intelligence Students in all four instructional conditions were evaluated in terms of their performance on Instructional studies are a further means of test- homework, a midterm exam, a fi nal exam, and an ing the theory (Sternberg & Grigorenko 2007 ; independent project. Each type of work was eval- Sternberg et al. 2001a , 2008 , 2009 ). uated for memory, analytical, creative, and prac- Several sets of studies investigated instruction tical quality. Thus, all students were evaluated in for academic skills. Four sets are briefl y described exactly the same way. here. Our results suggested the utility of the theory In a fi rst set of studies, researchers explored of successful intelligence. This utility showed the question of whether conventional education itself in several ways. in school systematically discriminates against First, we observed when the students arrived children with creative and practical strengths at Yale that the students in the high creative and (Sternberg and Clinkenbeard 1995 ; Sternberg high practical groups were much more diverse in et al. 1996 , 1999). Motivating this work was the terms of racial, ethnic, socioeconomic, and edu- belief that the systems in most schools strongly cational backgrounds than were the students in tend to favor children with strengths in memory the high-analytical group, suggesting that corre- and analytical abilities. lations of measured intelligence with status vari- The investigators used the Sternberg Triarchic ables such as these may be reduced by using a Abilities Test in some of their instructional work. broader conception of intelligence. Thus, the The test was administered to 326 children around kinds of students identifi ed as strong differed in the United States and in some other countries terms of populations from which they were drawn who were identifi ed by their schools as gifted by in comparison with students identifi ed as strong any standard whatsoever. Children were selected solely by analytical measures. More importantly, for a summer program in (college-level) psychol- just by expanding the range of abilities measured, ogy if they fell into one of fi ve ability groupings: the investigators discovered intellectual strengths high analytical, high creative, high practical, high that might not have been apparent through a con- balanced (high in all three abilities), or low bal- ventional test. anced (low in all three abilities). Students who Second, we found that all three ability tests— came to Yale were then divided into four instruc- analytical, creative, and practical—signifi cantly tional groups. Students in all four instructional predicted course performance. When multiple- groups used the same introductory-psychology regression analysis was used, at least two of these textbook (a preliminary version of Sternberg ability measures contributed signifi cantly to the (1995 )) and listened to the same psychology lec- prediction of each of the measures of achieve- tures. What differed among them was the type of ment. Perhaps as a refl ection of the diffi culty of afternoon discussion section to which they were deemphasizing the analytical way of teaching, assigned. They were assigned to an instructional one of the signifi cant predictors was always the 238 R.J. Sternberg analytical score. (However, in a replication of our could argue that this result merely refl ected the study with low-income African-American stu- way they were taught. Nevertheless, the result dents from New York, Deborah Coates of the suggested that teaching for these kinds of City University of New York found a different thinking succeeded. More important, however, pattern of results. Her data indicated that the was the result that children in the successful- practical tests were better predictors of course intelligence condition outperformed the other performance than were the analytical measures, children even on the multiple- choice memory suggesting that what ability test predicts what tests. In other words, to the extent that one’s criterion depends on population as well as mode goal is just to maximize children’s memory for of teaching.) information, teaching for successful intelli- Third and most importantly, there was an gence is still superior. It enables children to aptitude- treatment interaction whereby students capitalize on their strengths and to correct or to who were placed in instructional conditions that compensate for their weaknesses, and it allows better matched their pattern of abilities outper- children to encode material in a variety of formed students who were mismatched. In other interesting ways. words, when students are taught in a way that fi ts We extended these results to reading curricula how they think, they do better in school. Children at the middle-school and the high-school level. In with creative and practical abilities, who are a study of 871 middle-school students and 432 almost never taught or assessed in a way that high-school students, we taught reading either matches their pattern of abilities, may be at a dis- triarchically or through the regular curriculum. advantage in course after course, year after year. At the middle-school level, reading was taught A follow-up study (Sternberg et al. 1998 ) exam- explicitly. At the high-school level, reading was ined learning of social studies and science by third infused into instruction in mathematics, physical graders and eighth graders. The 225 third graders sciences, social sciences, English, history, for- were students in a very low-income neighborhood eign languages, and the arts. In all settings, stu- in Raleigh, North Carolina. The 142 eighth graders dents who were taught triarchically substantially were students who were largely middle to upper- outperformed students who were taught in stan- middle class studying in Baltimore, Maryland, and dard ways (Grigorenko et al. 2002 ). Fresno, California. In this study, students were The largest-scale study, described in assigned to one of three instructional conditions. In Sternberg et al. (2008 ), was conducted with 196 the fi rst condition, they were taught the course that teachers and 7,702 students. The study spanned basically they would have learned had there been 4 years, 9 states, 14 school districts, and 110 no intervention. The emphasis in the course was on schools. It showed that with many thousands of memory. In the second condition, students were fourth graders, it was possible to obtain gains in taught in a way that emphasized critical (analytical) fourth-grade reading and mathematics that thinking. In the third condition, they were taught in were greater for triarchic instruction for critical a way that emphasized analytical, creative, and thinking or memory. This study suggested that practical thinking. All students’ performance was triarchic instruction can be “scaled up” to reach assessed for memory learning (through multiple- children across a wide variety of geographic choice assessments) as well as for analytical, cre- areas as well as subject matter areas. ative, and practical learning (through performance Thus, the results of these sets of studies sug- assessments). gest that the theory of successful intelligence is As expected, students in the successful- valid as a whole. Moreover, the results suggest intelligence (analytical, creative, practical) that the theory can make a difference not only in condition outperformed the other students in laboratory tests but in school classrooms and terms of the performance assessments. One even the everyday life of adults as well. 16 Multiple Intelligences in the New Age of Thinking 239

might learn and test well, if only they were given Conclusions an opportunity to play to their strengths rather than their weaknesses. This chapter has presented the theory of success- As a society, we can create a closed system ful intelligence. Some psychologists believe the that advantages only certain types of children theory departs too much from the conventional and that disadvantages other types. Children theory of general intelligence (i.e., the theory of who excel in memory and analytical abilities Spearman 1904 , 1927 ). Some disagree with the may end up doing well on ability tests and theory (Gottfredson 2003a , b ; Jensen 1998 ). achievement tests and hence fi nd the doors of Others believe the theory does not depart from opportunity open to them. Children who excel in conventional g theory enough (Gardner 1983 , other abilities may end up doing poorly on the 2006 ). Still others have theories that are more tests and fi nd the doors shut. By treating children compatible, in spirit, with that proposed here, at with alternative patterns of abilities as losers, we least for intelligence (Ceci 1996 ). The theory is may end up creating harmful self-fulfi lling rather newer than that of, say, Spearman (1904 ) prophecies. and has much less work to support it as well as a Institutions should consider pooling their lesser range of empirical support. I doubt the resources and developing a common model and theory is wholly correct—scientifi c theories so common methods of assessment. By working far have not been—but I hope at the same time it separately, they fail to leverage their strengths serves as a broader basis for future theories than, and to share information regarding the best ways say, Spearman’s theory of general intelligence. to make decisions. In essence, each institution No doubt, there will be those who wish to pre- “reinvents the wheel.” A consortium would be far serve this and related older theories, and those more powerful than each institution working on who will continue to do research that replicates its own. Successful intelligence is one model hundreds and thousands of time that so-called such a consortium might use. Doubtless there are general intelligence does indeed matter for suc- many others. The important thing is to work cess in many aspects of life. I agree. At the same together toward a common good—toward devis- time, I suspect it is not suffi cient and also that ing the best ways to select students so as to maxi- those who keep replicating endlessly the fi ndings mize their positive future impact. We all wish our of the past are unlikely to serve as the positive intellectual leaders to show wisdom. We our- intellectual leaders of the future. But only time selves need to do the same. will tell. As noted earlier, there is typically some value to replication in science, although after a point is established, it seems more to continue to References produce papers than to produce new scientifi c breakthroughs. Boring, E. G. (1923, June 6). Intelligence as the tests test The educational system in the United States, it. New Republic , 35 – 37. Carroll, J. B. (1993). Human cognitive abilities: A survey as in many other countries, places great emphasis of factor-analytic studies . New York: Cambridge on instruction and assessments that tap into two University Press. important skills: memory and analysis. Students Cattell, R. B. (1971). Abilities: Their structure, growth who are adept at these two skills tend to profi t and action . Boston: Houghton Miffl in. Cattell, R. B., & Cattell, H. E. P. (1973). Measuring intel- from the educational system, because the ability ligence with the Culture Fair Tests . Champaign: tests, instruction, and achievement tests we use Institute for Personality and Ability Testing. all largely measure products and processes ema- Ceci, S. J. (1996). On intelligence (Rev. and exp. ed.). nating from these two kinds of skills. There is a Cambridge, MA: Harvard University Press. Gardner, H. (1983). Frames of mind: The theory of multi- problem, however, namely, that children whose ple intelligences . New York: Basic. strengths are in other kinds of skills may be Gardner, H. (1993). Multiple intelligences: The theory in shortchanged by this system. These children practice . New York: Basic. 240 R.J. Sternberg

Gardner, H. (2006). Multiple intelligences: New horizons Sternberg, R. J. (1985a). Beyond IQ: A triarchic theory of in theory and practice . New York: Basic. human intelligence . New York: Cambridge University Gottfredson, L. S. (2003a). Discussion: On Sternberg’s Press. ‘Reply to Gottfredson’. Intelligence, 31 (4), 415–424. Sternberg, R. J. (1985b). Implicit theories of intelligence, Gottfredson, L. S. (2003b). Dissecting practical intelli- creativity, and wisdom. Journal of Personality and gence theory: Its claims and evidence. Intelligence, Social Psychology, 49 (3), 607–627. 31 (4), 343–397. Sternberg, R. J. (1993). Sternberg triarchic abilities test . Grigorenko, E. L., Jarvin, L., & Sternberg, R. J. (2002). Unpublished test. School–based tests of the triarchic theory of intelli- Sternberg, R. J. (1995). In search of the human mind . gence: Three settings, three samples, three syllabi. Orlando: Harcourt Brace College Publishers. Contemporary Educational Psychology, 27 , 167–208. Sternberg, R. J. (1997). Successful intelligence. New York: Grigorenko, E. L., Meier, E., Lipka, J., Mohatt, G., Yanez, Plume. E., & Sternberg, R. J. (2004). Academic and practical Sternberg, R. J. (1998). Abilities are forms of developing intelligence: A case study of the Yup’ik in Alaska. expertise. Educational Researcher, 27 , 11–20. Learning and Individual Differences, 14 , 183–207. Sternberg, R. J. (1999a). Intelligence as developing Guyote, M. J., & Sternberg, R. J. (1981). A transitive- expertise. Contemporary Educational Psychology, chain theory of syllogistic reasoning. Cognitive 24 , 359–375. Psychology, 13 , 461–525. Sternberg, R. J. (1999b). The theory of successful intelli- Hedlund, J., Forsythe, G. B., Horvath, J. A., Williams, gence. Review of General Psychology, 3 , 292–316. W. M., Snook, S., & Sternberg, R. J. (2003). Identifying Sternberg, R. J. (Ed.). (2000). Handbook of intelligence . and assessing tacit knowledge: Understanding the New York: Cambridge University Press. practical intelligence of military leaders. Leadership Sternberg, R. J. (2003). WICS: A theory of wisdom, intel- Quarterly, 14 , 117–140. ligence, and creativity, synthesized. New York: (1921). “Intelligence and its measurement”: A symposium. Cambridge University Press. Journal of Educational Psychology, 12 , 123–147, Sternberg, R. J. (2004). Culture and intelligence. American 195–216, 271–275. Psychologist, 59 (5), 325–338. Jensen, A. R. (1998). The g factor: The science of mental Sternberg, R. J. (2005). The theory of successful intelligence. ability . Westport: Praeger/Greenwoood. Interamerican Journal of Psychology, 39(2), 189–202. Lubart, T. I., & Sternberg, R. J. (1995). An investment Sternberg, R. J. (2007). Culture, instruction, and assess- approach to creativity: Theory and data. In S. M. ment. Comparative Education, 43 (1), 5–22. Smith, T. B. Ward, & R. A. Finke (Eds.), The creative Sternberg, R. J. (2008). Schools should nurture wisdom. cognition approach . Cambridge, MA: MIT Press. In B. Z. Presseisen (Ed.), Teaching for intelligence Okagaki, L., & Sternberg, R. J. (1993). Parental beliefs (2nd ed., pp. 61–88). Thousand Oaks: Corwin. and children’s school performance. Child Sternberg, R. J. (2009). The Rainbow and Kaleidoscope Development, 64 (1), 36–56. Projects: A new psychological approach to undergrad- Spearman, C. (1904). ‘General intelligence’, objectively uate admissions. European Psychologist, 14 , 279–287. determined and measured. American Journal of Sternberg, R. J. (2010). College admissions for the 21st Psychology, 15 (2), 201–293. century . Cambridge, MA: Harvard University Press. Spearman, C. (1927). The abilities of man . London: Sternberg, R. J., & Clinkenbeard, P. R. (1995). The triar- Macmillan. chic model applied to identifying, teaching, and assess- Sternberg, R. J. (1977). Intelligence, information pro- ing gifted children. Roeper Review, 17 (4), 255–260. cessing, and analogical reasoning: The componential Sternberg, R. J., & Detterman, D. K. (Eds.). (1986). analysis of human abilities. Hillsdale: Lawrence What is intelligence? Norwood: Ablex Publishing Erlbaum Associates. Corporation. Sternberg, R. J. (1980a). Representation and process in Sternberg, R. J., & Gardner, M. K. (1983). Unities in linear syllogistic reasoning. Journal of Experimental inductive reasoning. Journal of Experimental Psychology: General, 109 , 119–159. Psychology: General, 112 , 80–116. Sternberg, R. J. (1980b). Sketch of a componential sub- Sternberg, R. J., & Gastel, J. (1989a). Coping with novelty theory of human intelligence. Behavioral and Brain in human intelligence: An empirical investigation. Sciences, 3 , 573–584. Intelligence, 13 , 187–197. Sternberg, R. J. (1981). Intelligence and nonentrench- Sternberg, R. J., & Gastel, J. (1989b). If dancers ate their ment. Journal of Educational Psychology, 73 , 1–16. shoes: Inductive reasoning with factual and counter- Sternberg, R. J. (1982). Natural, unnatural, and supernatu- factual premises. Memory and Cognition, 17 , 1–10. ral concepts. Cognitive Psychology, 14 , 451–488. Sternberg, R. J., & Grigorenko, E. L. (2007). Teaching for Sternberg, R. J. (1983). Components of human intelli- successful intelligence (2nd ed.). Thousand Oaks: gence. Cognition, 15 , 1–48. Corwin. Sternberg, R. J. (1984). Toward a triarchic theory of Sternberg, R. J., & Hedlund, J. (2002). Practical intelli- human intelligence. Behavioral and Brain Sciences, 7 , gence, g, and work psychology. Human Performance, 269–287. 15 (1/2), 143–160. 16 Multiple Intelligences in the New Age of Thinking 241

Sternberg, R. J., & Lubart, T. I. (1991). An investment Sternberg, R. J., Forsythe, G. B., Hedlund, J., Horvath, J., theory of creativity and its development. Human Snook, S., Williams, W. M., Wagner, R. K., & Development, 34 (1), 1–31. Grigorenko, E. L. (2000). Practical intelligence in Sternberg, R. J., & Lubart, T. I. (1995). Defying the crowd: everyday life. New York: Cambridge University Press. Cultivating creativity in a culture of conformity . Sternberg, R. J., Grigorenko, E. L., & Jarvin, L. (2001a). New York: Free Press. Improving reading instruction: The triarchic model. Sternberg, R. J., & Lubart, T. I. (1996). Investing in cre- Educational Leadership, 58 (6), 48–52. ativity. American Psychologist, 51 (7), 677–688. Sternberg, R. J., Nokes, K., Geissler, P. W., Prince, R., Sternberg, R. J., & Nigro, G. (1980). Developmental pat- Okatcha, F., Bundy, D. A., & Grigorenko, E. L. terns in the solution of verbal analogies. Child (2001b). The relationship between academic and prac- Development, 51 , 27–38. tical intelligence: A case study in Kenya. Intelligence, Sternberg, R. J., & Rifkin, B. (1979). The development of 29 , 401–418. analogical reasoning processes. Journal of Sternberg, R. J., Kaufman, J. C., & Pretz, J. E. (2002). The Experimental Child Psychology, 27 , 195–232. creativity conundrum: A propulsion model of kinds of Sternberg, R. J., & The Rainbow Project Collaborators. creative contributions . New York: Psychology Press. (2006). The Rainbow Project: Enhancing the SAT Sternberg, R. J., Grigorenko, E. L., & Zhang, L.-F. (2008). through assessments of analytical, practical and cre- Styles of learning and thinking matter in instruction ative skills. Intelligence, 34 (4), 321–350. and assessment. Perspectives on Psychological Sternberg, R. J., & Wagner, R. K. (1993). The g –ocentric Science, 3 (6), 486–506. view of intelligence and job performance is wrong. Sternberg, R. J., Jarvin, L., & Grigorenko, E. L. (2009). Current Directions in Psychological Science, 2 (1), Teaching for wisdom, intelligence, creativity, and suc- 1–4. cess . Thousand Oaks: Corwin. Sternberg, R. J., Wagner, R. K., & Okagaki, L. (1993). Sternberg, R. J., Bonney, C. R., Gabora, L., & Merriﬁ eld, Practical intelligence: The nature and role of tacit M. (2012). WICS: A model for college and university knowledge in work and at school. In H. Reese & admissions. Educational Psychologist, 47 (1), 30–41. J. Puckett (Eds.), Advances in lifespan development Tetewsky, S. J., & Sternberg, R. J. (1986). Conceptual and (pp. 205–227). Hillsdale: Lawrence Erlbaum lexical determinants of nonentrenched thinking. Associates. Journal of Memory and Language, 25 , 202–225. Sternberg, R. J., Wagner, R. K., Williams, W. M., & Thurstone, L. (1938). Primary mental abilities . Chicago: Horvath, J. A. (1995). Testing common sense. University of Chicago Press. American Psychologist, 50 (11), 912–927. Vernon, P. E. (1971). The structure of human abilities . Sternberg, R. J., Ferrari, M., Clinkenbeard, P. R., & London: Methuen. Grigorenko, E. L. (1996). Identiﬁ cation, instruction, Visser, B. A., Ashton, M. C., & Vernon, P. A. (2006). and assessment of gifted children: A construct valida- Beyond g: Putting multiple intelligences theory to the tion of a triarchic model. Gifted Child Quarterly, 40 , test. Intelligence, 34 , 487–502. 129–137. Wagner, R. K. (1987). Tacit knowledge in everyday intel- Sternberg, R. J., & Grigorenko, E. L. (1997). The cogni- ligent behavior. Journal of Personality and Social tive costs of physical and mental ill health: Applying Psychology, 52 (6), 1236–1247. the psychology of the developed world to the prob- Wagner, R. K., & Sternberg, R. J. (1986). Tacit knowledge lems of the developing world. Eye on Psi Chi, 2 (1), and intelligence in the everyday world. In R. J. Sternberg 20–27. & R. K. Wagner (Eds.), Practical intelligence: Nature Sternberg, R. J., Torff, B., & Grigorenko, E. L. (1998). and origins of competence in the everyday world Teaching triarchically improves school achievement. (pp. 51–83). New York: Cambridge University Press. Journal of Educational Psychology, 90 , 374–384. Williams, W. M., Blythe, T., White, N., Li, J., Gardner, Sternberg, R. J., Grigorenko, E. L., Ferrari, M., & H., & Sternberg, R. J. (2002). Practical intelligence Clinkenbeard, P. (1999). A triarchic analysis of an for school: Developing metacognitive sources of aptitude-treatment interaction. European Journal of achievement in adolescence. Developmental Review, Psychological Assessment, 15 (1), 1–11. 22(2), 162–210. Emotional and Social Intelligence and Behavior 17

Richard E. Boyatzis , James Gaskin , and Hongguo Wei

Of the various levels at which emotional intelligence (EI) and social intelligence (SI) exist within a A Review of Various Theoretical person, the behavioral level has received the least and Methodological Approaches amount of attention in academic research but to EI holds the most promise for a concept and measurement approach that relates to job and life To offer a brief review, expanding on Fernandez- outcomes, as well as allowing others to “see” EI Berrocal and Extremera (2006 ), Boyatzis (2009 ), and SI in action (Goleman 1998 ; Boyatzis 2009 ; and Cherniss (2010 ), the three major theoretical Cherniss 2010 ). Emotional intelligence (EI) approaches to EI are described below. Within manifests itself at many levels within a person each theory, we further briefl y describe the (Cherniss and Boyatzis 2013 ). In the past, discus- measures currently appearing in publications. sion of EI was often focused on the different A framework, or grand theory of EI, is offered theoretical models and different ways of assess- in Fig. 17.1 to suggest where the ability, self- ing EI (Matthews et al. 2004 ). In this chapter, we perception, and behavioral levels coexist will review the major models or theories which (Cherniss and Boyatzis 2013 ). constitute levels of EI and SI and the tests appearing in research publications. The chapter will then focus on the behavioral level and the mea- Ability-Based Level of EI surement at this level currently most in use. Most credit Salovey and Mayer ( 1990) with the fi rst use of the phrase “emotional intelligence” in a published article. They defi ned EI as “the sub- * R. E. Boyatzis ( ) set of social intelligence that involves the ability Case Western Reserve University , 2077 Kent Road , Cleveland Heights , OH 44106 , USA to monitor one’s own and others’ feelings and e-mail: [email protected] emotions, to discriminate among them and to use J. Gaskin this information to guide one’s thinking and Marriott School of Management , Brigham Young actions” (Salovey and Mayer 1990). It is based University , Provo , UT , USA on the view of emotional processes as relying on e-mail: [email protected] complex neural activities and as motives to be H . W e i intelligent (Leeper 1948 ). It is a type of “emo- Case Western Reserve University , tional information processing” (Salovey and 1947 E. 120th Street, Apt 3 , Cleveland , OH 44106 , USA Mayer 1990 ). They further clarifi ed EI as “the e-mail: [email protected] ability to perceive and express emotion, assimilate

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 243 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_17, © Springer Science+Business Media New York 2015 244 R.E. Boyatzis et al.

Performance

Observed cluster of EI and/or SI competencies

Behavioral Level, like the ESCI Behavioral Behavioral Behavioral Behavioral expression of EI expression of EI expression of SI expression of SI

competency 1 competency 2 competency 1 competency 2

Self- Self-perception and self-conceptualization of EI and SI Per-

ceived

Level, like Value and philosophical foundations of EI and SI The EQ-i

Motivation, trait, and unconscious dispositions or abilitiesrelated to EI Basic Ability Level, or SI like the MSCEIT

Neural circuits and hormonal patterns related to EI or SI

Fig. 17.1 Emotional intelligence (EI), social intelligence Cherniss and Boyatzis (2013 ) adapted from Boyatzis et al. (SI), and emotional and social competencies as multiple (2000) and Boyatzis ( 2009 ) ) levels within the personality structure (appearing in

emotion in thought, understand and reason.” 1999 ; Salovey and Mayer 1997). It is scored in In their theory, Mayer and Salovey ( 1997 ) claim two ways which has caused some confusion: con- that EI is comprised of four dimensions: emotion sensus and expert scoring. Because they deﬁ ne perception, emotion understanding, emotional EI as a type of cognitive intelligence, it should facilitation, and emotion regulation. correlate with cognitive ability (Joseph and The measure used most often to examine this Newman 2010 ). The MSCEIT has content valid- approach (EI) is the Mayer-Salovey-Caruso ity (Cherniss 2010 ). Reliability of the MSCEIT is Emotional Intelligence Test (MSCEIT). It is a appropriate, with split-half estimates for the direct performance assessment of emotional pro- whole scale of .91 and .93 (Mayer et al. 2003 ). cessing with some scenarios testing (Mayer et al. Test-retest reliability has been estimated as r = .86 17 Emotional and Social Intelligence and Behavior 245

(N = 60) (Brackett and Mayer 2003 ). The overall the concept. That said, the model does seem too internal consistency reliabilities are usually restricted to some. Several emotional-related above .75, although the reliabilities of the sub- qualities commonly ascribed to EI are excluded, scales have not been as good (Conte 2005 ; such as “emotional expressiveness, empathy, Matthews et al. 2002 ). perspective-taking, self-control, and implicit A CFA does not support their 4-factor model emotional skills” (Matthews et al. 2006 , p. 106). (Gignac 2005 ; Rossen et al. 2008 ). The MSCEIT For example, the 4-factor model does not seem seems to have discriminant validity with tests of plausible. A CFA analysis shows that “the non- personality. For instance, MSCEIT factors have constrained 4-factor model yield a non-positive the strongest correlations with agreeableness defi nite matrix, which was interpreted to be due from the Big Five (r = .21–.28), while their cor- to the fact that two of the branch-level factors relations with the other four personality factors (perceiving and facilitating) are collinear” are less than .20 (Mayer et al. 2008 ). (Gignac 2005 , p. 233). Two other studies repli- The low correlations between MSCEIT fac- cated Gignac’s (2005 ) results (Palmer et al. 2005 ; tors and relevant constructs impair the conver- Rossen et al. 2008 ). In addition, the scoring sys- gent validity for the MSCEIT. “Convergent tem is complicated (MacCann and Roberts 2008 ). validity for the MSCEIT seems more problem- It is diffi cult to tell whether a test item is appro- atic…. On the other hand, the MSCEIT corre- priately answered (Matthews et al. 2006 ). In lates with measures of verbal intelligence (r = .36) order to solve this problem, it takes two different and with other kinds of intelligence (r = .10 to approaches in the scoring system: consensus .20) at the levels one would want from a form of scoring and expert scoring. Although these two intelligence that is supposed to be related to but scoring approaches are highly correlated (r = .96– distinct from other types of intelligence” .98), “it is unclear whether a person who thinks (Cherniss 2010 ). For instance, as a form of emo- about the emotional domain differently from tional intelligence measure, MSCEIT is supposed experts or from the average of several peers is to be moderately correlated but distinct with low on that ability or whether that person simply other emotional perception constructs. However, has a new (and perhaps better) way of thinking” Roberts et al. (2006 ) found that the Japanese (Murphy 2006 , p. 348). and Caucasian Brief Affect Recognition Test The format of the MSCEIT has been com- (JACBART) has low correlations with the overall pared to knowledge tests of EI, which may not MSCEIT scale and its subscales (r = .20–.26) and provide an appropriate assessment of a person’s does not correlate with the emotional perception actual ability (Cherniss 2010 ). “The assessment scale of the MSCEIT at all (r = .00). Similarly, the of knowledge in the abstract does not refl ect the MSCEIT subscales have correlations with the live performance of EI in the rich social situation levels of emotional awareness scale (LEAS) from of real life … one might understand that smiling .15 to .20 (Ciarrochi et al. 2003 ). These results at someone can be an effective means of produc- suggest that the MSCEIT has weak or little con- ing a positive emotional reaction, but recognizing vergent validity with other relevant emotional in a live encounter the moment to smile and doing constructs. so in a way that does not seem false or insincere The incremental validity of the MSCEIT is may well be a different ability” (Spector and quantitatively supported. With a student sample, Johnson 2006 , p. 335). Rossen and Kranzler (2009 ) found that after con- The Schutte Self-Report Emotional Intelli- trolling for general mental ability and personal- gence Test (SREIT) is a self-report scale to mea- ity, EI measured with the MSCEIT explained sure “a homogeneous construct of emotional incremental variance to positive relationships intelligence” based on Mayer-Salovey-Caruso with others and alcohol use. model (Schutte et al. 1998 ). It has 33 items. This Criticism from those other than skeptics of EI test will be reviewed here, but all of the charac- in general has focused more on the measure than teristics and criticisms of the self-perception 246 R.E. Boyatzis et al. methods described in the next segment apply to factors should have been rotated obliquely, not this test because the method of assessment is self- orthogonally (Petrides and Furnham 2000 ). report (Schutte et al. 1998 ). Another self-assessment test that was based on The SREIT appears to be a face-valid measure the MSCEIT model was by Wong and Law which of EI (Petrides and Furnham 2000 ). It has inter- offered “a set of interrelated abilities possessed nal consistency (i.e., Cronbach’s alpha) of .90 for by individuals to deal with emotions” (Wong and the 33 items (Schutte et al. 1998 ). A 2-week test- Law 2002 ; Law et al. 2004 ). Their construct has retest reliability is adequate (r = .78) (Conte and four dimensions: self-emotional appraisal, others’ Dean 2006 ). Research results on discriminant emotional appraisal, regulation of emotion, and validity have been mixed. In an initial small study use of emotion. Their test, the WLEIS, is com- with 23 college students, the SREIT’s correlation posed of 16 items. With two samples, the authors with openness to experience was high (r = .54), showed convergent, discriminant, and incremental but the absolute correlations with other Big Five validity of this 16-item EI scale (Wong and Law personality factors were lower (r = .21–.28) 2002 ). Their empirical analysis shows that EI has (Schutte et al. 1998). However, in one larger “incremental predictive power over life satisfac- study, the SREIT’s correlations with the Big Five tion.” After controlling for Big Five personality, were from .18 (agreeableness) to .51 (extraver- both the student sample and the work samples sion) (Saklofske et al. 2003 ). Results from these showed that others’ ratings of EI explained addi- two studies show that the correlations between tional variance (Law et al. 2004 ). All of the criti- SREIT and the Big Five personality factors were cisms of self-assessment explained in the next not consistent across different samples. In a dif- section also apply to their test, with the additional ferent study, the correlation between SREIT and major issue that it is only 16 items which dramati- psychological well-being was pretty high (r = .70) cally limits the scope of the assessment. (Brackett and Mayer 2003), which implies that the SREIT is possibly a different measure of psychological well-being. Self-perception Level of EI Since the SREIT is believed to be measuring a type of intelligence, it should have a moderate Reuven Bar-On developed a model on “an array correlation with the general intelligence (Cherniss of non-cognitive capabilities, competencies, and 2010). However, a study shows that the SREIT skills that infl uence one’s ability to succeed in was not signifi cantly correlated with cognitive coping with environmental demands and pres- ability (Saklofske et al. 2003 ), which suggests sures” (1997 , p. 14). The components included that the SREIT did not have the acceptable confi ve subtypes of EI: intrapersonal intelligence, vergent validity. interpersonal intelligence, adaptability, stress The criticisms again as with the MSCEIT management, and general mood (Bar-On 1997 , focus more on the measure than the concept. For 2006 ). The test, the Emotional Quotient Inventory, example, it was said that the SREIT cannot mea- is composed of 15 scales in four composites: self- sure a general EI factor because the test is not perception includes emotional self- awareness, unifactorial (Petrides and Furnham 2000). The self-regard, and self-actualization; self-expression 33 items of Schutte’s test came from all three includes assertiveness, emotional expression, subcategories of Salovey and Mayer’s original and independence; interpersonal includes empa- EI model. However, in the analysis, Schutte thy, interpersonal relationships, and social failed to demonstrate the three subdimensions responsibility; decision making includes problem of EI as separate (Petrides and Furnham 2000 ). solving, reality testing, and impulse control; “It would have been more appropriate to per- stress management includes optimism, fl exibility, form a factor analysis on the 62 items, rather and stress tolerance. The EQ-i was originally a than a component analysis” (Gignac et al. 2005 , self-report; in 1997, the 360 version was intro- p. 1030). Furthermore, it was claimed that the duced (Bar-On 1997 ). 17 Emotional and Social Intelligence and Behavior 247

Since the model on which it was based and because both convergent and discriminant validity most, if not all, of the research reported in jour- are based on the correlation coeffi cients, an nals and book chapters to date is from the self- exceptional good convergent validity is a threat assessment version, it is included in this section. to the discriminant validity. Actually, evidence To the extent the published research appears has shown that the EQ-i does not have good dis- using the 360 version or Bar-On changes his criminant validity especially considering its rela- model, it could be included in the subsequent tionship with personality variables. Bar-On behavioral level section as well. (2006 , p. 16) mentioned that the EQ-i overlaps The EQ-i has good internal consistency. With with personality tests “probably no more than 243 university students, a study shows that the 15 % based on eight studies in which more than internal consistency reliability ranged from .69 to 1,700 individuals participated.” However, one .96, with an overall estimate of .96 (Dawda and study shows that when using the Big Five pre- Hart 2000 ). Another recent study in North dicts EQ-i scores, the multiple correlation is .79, America showed an internal consistency of .97, which means the Big Five personality accounted and a 6-month test-retest reliability of .72 for for the majority of variance in the EQ-i (Grubb men ( n = 73) and .80 for women (n = 279) (Bar-On and McDaniel 2007 ). 2004 ). However, the internal structure of EQ-i is Another self-assessment test but one based on not consistent. An EFA with a varimax rotation a composite of the various models of all of the generated a 13-factor model rather than the major EI authors is the TEIQue by Petrides and 15-factor model in Bar-On theory. After remov- Furnham (2000 , 2001 , 2003 ). They claim to ing the problematic factors, a CFA generates a assess “trait EI” which they say is “a constella- 10-factor model (Bar-On 2006 ). tion of emotion-related self-perceived abilities Although Bar-On theory was “designed to and dispositions located at the lower levels of examine … a conceptual model of emotional and personality hierarchies” (Petrides and Furnham social functioning” (Bar-On 2006 , p. 15), the crit- 2000 ). It is meant to include all “personality fac- icism of this approach has been focusing on its ets that are specifi cally related to affect” (Petrides overly broad conceptualization (Murphy 2006 ). et al. 2007 , p. 274). The TEIQue has four subdi- Others have claimed that EQ-i has a great deal of mensions: emotionality, self-control, sociability, overlap with personality, “predictive validity may and well-being. simply be a consequence of the EQ-i functioning After controlling for the Big Five, the TEIQue as a proxy measure of personality” (Matthews has a positive relationship with happiness et al. 2004 , p. 16). Thus, the content validity of (Furnham and Petrides 2003 ). It is linked with EQ-i is doubtful given that it includes nonability distinctive reactivity to affect-laden information personality traits and ignores some essential EI and has incremental validity over the Big Five factors such as emotional understanding and (Petrides and Furnham 2003 ; Petrides et al. emotional perception (Cherniss 2010). 2004 ). The TEIQue also has incremental validity The construct validity of Bar-On model and over alexithymia and optimism (Mikolajczak the test confi rm that he/she is testing ESC, which et al. 2006 , 2007 ). is similar to Boyatzis and Goleman model to be The criticisms about the TEIQue are similar to explained in the next section as described by those raised about the and self-assessment in Cherniss (2010 ). An empirical study confi rms general. that EQ-i has convergent validities “with respect to measures of normal personality, depression, somatic symptomatology , intensity of affective The Behavioral Level of EI and SI experience and alexithymia” (Dawda and Hart 2000, p. 797). Also, the EQ-i has been reported The behavioral level has been the most discussed correlating well with some other self-report EI and documented as a “competency” approach measures (r = .58–.69) (Bar-On 2004 ). However, (Boyatzis 2009 ; Cherniss 2010 ). A competency 248 R.E. Boyatzis et al. was originally defi ned as “an underlying ability person describe that person’s typical behavior that leads to or causes effective performance” through a series of questions. Items for the 360 (McClelland 1973 ; Boyatzis 1982; Spencer and were created based on the behavior validated in Spencer 1993 ). The various competencies were the earlier studies with a phrase added that pro- described as a set of related behavior or actions vided the respondent with the intent. For exam- organized around an underlying intent or con- ple, the behavior of “listens attentively” was text for their use. In this sense, the appearance validated as an indicator of empathy in many of of the behavior was described as “alternate the competency studies. To include it in the 360, manifestations” of the underlying characteristics we added the intent. The item now reads, (McClelland 1985 ). “Understands others by listening attentively.” For example, asking someone questions is an The most widely used and cited of the behav- action. But someone can do this for multiple ioral level approaches to EI comes from reasons or intent. Someone might ask questions McClelland, Boyatzis, Goleman, Cherniss, and to better understand someone. Another person, their colleagues at The Hay Group. Several defi - or the same person at another time, might ask nitions have appeared over the decades. One was questions as a way to convince someone else of that “Emotional intelligence [includes] abilities a position on an issue. The former would result such as being able to motivate oneself and persist in the combination of intent and action to be in the face of frustrations; to control impulse and labeled or coded as empathy, but the latter would delay gratifi cation; to regulate one’s moods and be infl uence. keep distress from swamping the ability to think; One explanation as to why these EI and SI to empathize and to hope” (Goleman 1995 , competencies have such a strong relationship to p. 34). In describing their defi nitions of these life outcomes and job performance is that they clusters in detail, Boyatzis ( 2009 ) said, “(a) an were originally derived inductively by comparing emotional, intelligence competency is an ability effective and ineffective people in various occu- to recognize, understand, and use emotional pations in a wide variety of organizations in many information about oneself that leads to or causes countries (Boyatzis 1982 ; Spencer and Spencer effective or superior performance; (b) a social 1993 ; McClelland 1998 ; Goleman 1998 ). intelligence competency is the ability to recog- Outcome criteria or nominations were used to nize, understand and use emotional information identify samples of people who were effective in about others that leads to or causes effective or a job and those who were not. Detailed collection superior performance; and (c) a cognitive intelli- of their behavior and thoughts using variations of gence competency is an ability to think or ana- critical incident interviews (Flanagan 1954 ; lyze information and situations that leads to or Boyatzis 1982; Spencer and Spencer 1993 ) or causes effective or superior performance.” videotaped simulations enabled researchers to In the Boyatzis and Goleman model, EI has develop “codes” to determine whether or not a two dimensions and SI has two dimensions: EI person was demonstrating these EI and SI com- includes self-awareness and self-management petencies in various situations. For more detailed and SI includes social awareness and relation- explanation of the methods, see Boyatzis (1982 , ship management. The Emotional and Social 1998 ) or Spencer and Spencer (1993 ), or Boyatzis Competency Inventory (ESCI), or its univer- ( 2009 ) for a review of the methods. sity version (ESCI-U which includes two cog- Once these “codes” of competencies were nitive competencies), is more outcome oriented applied in many settings and jobs, 360 assess- because of the way in which they were devel- ments were developed to collect the observations oped (Boyatzis and Goleman 1996, 1999; from a broader array of sources and make the Wolff 2007, 2008; Hay Group 2011). Detailed data collection easier (Boyatzis 2009 ). A 360 statistical analysis and results about the psy- measurement is a multisource assessment in chometrics of the tests are summarized in the which people who live and work with a target following section. 17 Emotional and Social Intelligence and Behavior 249

The ESCI assesses 12 competencies and the Table 17.1 The scales and clusters of the Emotional and ESCI-U assesses 14 competencies. Past research Social Competency Inventory (ESCI and ESCI-U—the university version) (Boyatzis and Goleman 1996 , 1999; has shown a set of competencies that differentiate Boyatzis et al. 2001, 2007) outstanding and effective performers from others Emotional intelligence competencies are as follows : in many countries (Bray et al. 1974 ; Boyatzis Self-awareness cluster: Concerns knowing one’s internal 1982 ; Kotter 1982 ; Thornton and Byham 1982 ; states, preferences, resources, and intuitions. The Luthans et al. 1988 ; Howard and Bray 1988 ; self-awareness cluster contains one competency: Campbell et al. 1970 ; Spencer and Spencer 1993 ; Emotional self-awareness : Recognizing one’s emotions Goleman 1998 ; Hopkins and Bilimoria 2008 ; and their effects Koman and Wolff 2008 ; Dreyfus 2008 ; Williams Self-management cluster: Refers to managing one’s internal states, impulses, and resources. The self- 2008 ; Sternberg 1996 ). They can be clustered management cluster contains four competencies: into three sets of competencies, two of which are Emotional self-control : Keeping disruptive emotions related to EI and SI: (1) cognitive competencies, and impulses in check such as systems thinking and pattern recognition; Adaptability : Flexibility in handling change (2) emotional intelligence competencies, includ- Achievement orientation : Striving to improve or ing self-awareness and self-management compe- meeting a standard of excellence tencies, such as emotional self-awareness and Positive outlook : Seeing the positive aspects of things and the future emotional self-control, and what used to be a part Social intelligence competencies are as follows: of EI but is now separated for theoretical as well Social awareness cluster: Refers to how people handle as empirical reasons; and (3) social intelligence relationships and awareness of others’ feelings, needs, competencies, including social awareness and and concerns. The social awareness cluster contains two relationship management competencies, such as competencies: empathy and teamwork. The specifi c competen- Empathy: Sensing others’ feelings and perspectives and taking an active interest in their concerns cies considered to be a behavioral approach to Organizational awareness : Reading a group’s emotional, social, and cognitive intelligence are emotional currents and power relationships shown in Table 17.1 . Relationship management cluster: Concerns the skill or The criticisms of this approach are that the adeptness at inducing desirable responses in others. The concepts are over-inclusive (Matthews et al. cluster contains fi ve competencies: 2004). Other criticisms of Matthews et al. (2004 ) Coach and mentor : Sensing others’ development needs and bolstering their abilities have been addressed in publications in the last 8 Inspirational leadership : Inspiring and guiding years and are reviewed in this section of this individuals and groups chapter. Infl uence : Wielding effective tactics for persuasion There has been one other measure appearing Confl ict managemen t : Negotiating and resolving in selected publications which was developed to disagreements assess the behavioral level of EI. Dulewicz and Teamwork : Working with others toward shared goals. Higgs (1999) developed a measure based on the Creating group synergy in pursuing collective goals work of Boyatzis (1982 ) and Boyatzis and Cognitive intelligence competencies (in the ESCI university version only) are as follows: Goleman (1996 ). They assessed being aware of Systems thinking: Perceiving multiple causal and managing one’s own feelings and emotions, relationships in understanding phenomena or events being sensitive to and infl uencing others, sustain- Pattern recognition: Perceiving themes or patterns in ing one’s motivation, and balancing one’s moti- seemingly random items, events, or phenomena vation and drive with intuitive, conscientious, and ethical behavior (Dulewicz and Higgs 1999). Their measure, the EIQ, has seven dimensions: The alpha for overall EIQ was .77. The alpha self-awareness, emotional resilience, motivation, coeffi cients for each of the element scales ranged interpersonal sensitivity, infl uence, intuitiveness, from .6 to .8 (Dulewicz and Higgs 2000 ). The and conscientiousness (Higgs and Dulewicz EIQ was signifi cantly related with performance 2002 ; Dulewicz et al. 2003). It is also a 360. measures (except for interpersonal sensitivity) 250 R.E. Boyatzis et al.

(Dulewicz and Higgs 2000 ). Dulewicz et al. using a sample of 468 respondents, reported (2003 ) tested EIQ’s content, construct, and crite- the following rWG(J) for each competency: rion-related validity and showed support. emotional self-awareness, 0.86; achievement orientation, 0.94; adaptability, 0.95; emotional self-control, 0.92; positive outlook, 0.95; empathy, The ESCI and ESCI-U 0.92; organizational awareness, 0.92; coach and mentor, 0.91; infl uence, 0.93; confl ict manage- The ESCI is a 68-item test assessing the 12 EI ment, 0.88; inspirational leadership, 0.94; and and SI competencies listed in Table 17.1 . The teamwork, 0.97. ESCI-U (university version) is a 70-item test assessing the 12 EI and SI competencies in Table 17.1 , as well as the two CI (cognitive) com- Model Fit Using CFA petencies listed. The tests were designed to be used as 360, multisource rater instruments. In An exploratory factor analyses on each of the most research, only the “other” assessments from four data sets (ESCI self, ESCI others, ESCI-U informants are used. The self-assessment is used self, and ESCI-U others) showed almost all of the primarily in applications in coaching, training, items loaded on the appropriate, predicted scale and college courses, along with the “other” (Boyatzis and Gaskin 2010 ; Hay Group 2011 ). assessments. The following statistical tests were Nonconforming items were dropped from the computed on the ESCI with a sample of 5,761 analysis. Several items which were close to the self-assessments and 62,297 other assessments. typical cutoffs were slightly reworded for future The effective sample sizes for CFAs with no use of the test but kept in the subsequent analyses missing data were 4,468 and 25,057, respectively. reported here. Confi rmatory factor analyses were The sample was generated from research studies run on each of the four samples to assess the and training programs with managers and profes- model fi t to the 12 scales in the ESCI and 14 sionals from many countries of the world. The scales in the ESCI-U. following statistical tests were computed on the Except for the chi-square to df ratio for the ESCI-U with a sample of 1,629 self-assessments ESCI self and other and ESCI-U other, all esti- and 21,288 other assessments. The effective sam- mates of model fi t were within acceptable stan- ple sizes for CFAs with no missing data were dards to support the scales as an acceptable 1,398 and 8,981, respectively. The ESCI-U sam- model. In the chi-square to df ratio case, it is ple was generated from research and graduate believed that the large sample size contributed to courses with MBAs from the Weatherhead an infl ation of the chi-square, thus rendering that School of Management at Case Western Reserve estimate less useful. Specifi cally, for the self and University. the other, independently calculated, for the ESCI and ESCI-U, the RMSEA, PCLOSE, CFI, PCFI, GFI, and SRMR are within desired levels for a Scale Reliabilities Using Alpha satisfactory model fi t. Prior to running the CFA, all subjects with missing data were eliminated, Cronbach’s alpha was computed for each scale resulting in the smaller sample sizes reported in shown in Table 17.2 . As a comparison, the earlier Table 17.3 . version of the test, called the ECI-2 “others,” showed an average alpha of 0.78 and since 2007 the ESCI “others” shows 0.87. Thus, the ESCI Convergent and Discriminant Validity shows improved scale reliability as well as better of the Scales factor structure. To support aggregation across various informant sources, inter-rater agreement PLS-GRAPH version 3.0 was used to determine scores (rWG(J)) were calculated. Badri ( 2013 ), the reliability and validity of the scales. Partial 17 Emotional and Social Intelligence and Behavior 251

Table 17.2 Cronbach’s alpha for each scale in the ESCI and ESCI-U (sample sizes are shown in parentheses following the alpha) ESCI ESCI-U Self Other Self Other Emotional self-awareness .754 (5,534) .827 (42,215) .771 (1,605) .804 (16,223) Emotional self-control .854 (5,664) .910 (56,713) .843 (1,611) .882 (17,993) Achievement orientation .800 (5,668) .861 (47,975) .705 (1,621) .779 (18,381) Adaptability .720 (5,573) .845 (53,875) .752 (1,605) .820 (16,914) Positive outlook .829 (5,641) .883 (54,598) .825 (1,606) .846 (17,164) Empathy .708 (5,638) .856 (52,138) .725 (1,622) .836 (17,520) Organizational awareness .786 (5,579) .861 (54,475) .764 (1,603) .830 (17,159) Inﬂ uence .721 (5,606) .835 (50,702) .746 (1,569) .822 (14,693) Teamwork .771 (5,668) .886 (58,911) .775 (1,616) .857 (18,354) Coach and mentor .856 (5,546) .920 (48,744) .821 (1,590) .868 (15,369) Conﬂ ict management .682 (5,607) .785 (51,948) .636 (1,592) .733 (16,252) Inspirational leadership .897 (5,221) .887 (51,199) .817 (1,536) .866 (15,749) Systems thinking NA NA .794 (1,582) .821 (15,176) Pattern recognition NA NA .792 (1,577) .831 (13,502)

Table 17.3 Model fi t using CFA using AMOS for the root of the average variance extracted by the ESCI and ESCI-U both self and other responses for each measures of that construct (Lim et al. 2006 ).1 ESCI ESCI-U As shown in Tables 17.4 through 17.7 , for Self Other Self Other each of the scales in each of the analyses, all con- Effective n 4,468 25,057 1,398 8,981 struct composite reliabilities, average variance Chi-square 18,793 100,823 6,606 31,746 extracted, and their relationship have met the df 2,013 2,013 1,921 1,921 respective thresholds to be considered suffi - RMSEA .043 .044 .042 .042 ciently convergent. In addition, the square root of PCLOSE 1.000 1.000 1.000 1.000 the average variance extracted was greater than CFI .849 .910 .875 .919 the inter-scale correlations, showing discriminant PCFI .797 .855 .808 .849 validity for each of the scales for each of the anal- GFI .861 .860 .862 .888 yses. These indicators of convergent and discrim- SRMR .0536 .0424 .0469 .0319 inant validity are shown for the ESCI self in Table 17.4 , the ESCI other in Table 17.5 , the least squares was used rather than the traditional ESCI-U self in Table 17.6 , and for the ESCI-U covariance-based SEM tools such as LISREL or AMOS. PLS is especially well suited for working 1 with new measures (Chin et al. 2003 ). Almost no one reports maximum shared variance any longer as an indication of discriminant validity, but for To test for convergent validity, items in each those afi cionados who still use it, we have a different posi- construct must have loadings over 0.5 (Fornell tion. In many analyses of this type, another measure is 1982 ; Hair et al. 1995 ) and composite reliabilities also recommended for showing discriminant validity, (CR) should be over 0.7 (Nunnally et al. 1994 ) called the maximum shared variance (MSV). But in this case, in which the theoretical model on which the items and greater than their respective average variance and scales were built is a circumplex model, it is assumed extracted (AVE). Lastly, the average variance that some items, as well as scales, will have a high shared extracted should be maximized, with a minimum variance with others. For example, an expression or use of of 50 % (Barclay et al. 1995 ). Discriminant valid- empathy is required to show inspirational leadership. As a result, the maximum shared variance of an item or scale ity is established by showing that the correlation will be deceptive and not an appropriate indicator of dis- between any two constructs is less than the square criminant or discriminant validity. 252 R.E. Boyatzis et al.

TEAM .737

.835

.792

.807

] .742

.729

.735

.737 square root of the AVE on the diagonal AVE of the root square

.794 = 4,468) [ n

.744

.722 SD CR AVE ESA AO ESC A PO EMP OA IL CFM CM μ ESCI self: Convergent and discriminant validity using PLS ( and discriminant validity ESCI self: Convergent Table Table 17.4 Scale ESA AO ESC 3.744 A 4.278 .817 PO 4.023 EMP .695 .848 .698 OA 4.127 4.182 IL .861 3.983 .687 .521 .895 CFM .683 4.243 .663 CM .554 .826 .631 3.955 TEAM 3.867 .670 .875 .819 .314 INF 4.130 .743 .543 4.338 .175 .745 .860 .540 .532 .747 .642 .863 3.848 .301 .221 .882 .551 .279 .448 .902 .773 .856 .652 .456 .627 .415 .409 .296 .697 .801 .544 .391 .368 .365 .319 .419 .441 .341 .575 .355 .483 .369 .321 .427 .406 .447 .290 .347 .288 .196 .474 .380 .222 .280 .407 .497 .344 .367 .350 .212 .347 .533 .327 .482 .383 .378 .420 .428 .357 .379 .325 .545 .452 .342 .362 .320 .455 .445 .598 .435 .519 .372 .413 .323 .252 .458 .363 .358 17 Emotional and Social Intelligence and Behavior 253

TEAM .811

.848

.786 square root of the AVE on the AVE of the root square

.832 = 12,419) [ n

.803

.817

.808

.760

.835

.780

.741 SD CR AVE ESA AO ESC A PO EMP OA IL CFM CM μ ESCI Other: Convergent and discriminant validity using PLS convergent and discriminant validity using PLS ( and discriminant validity using PLS convergent and discriminant validity ESCI Other: Convergent ] ESA AO ESC 4.136 A 4.315 .795 PO 3.892 EMP .768 .879 .844 OA 4.058 3.880 IL .903 3.806 .877 .549 .932 CFM .852 4.124 .841 CM .608 .891 .697 4.036 TEAM 4.024 .815 .919 .889 .570 INF 4.298 .850 .578 4.192 .496 .889 .901 .654 .667 .758 .755 .918 .586 4.262 .498 .865 .645 .563 .711 .939 .802 .920 .692 .697 .617 .608 .626 .573 .720 .871 .658 .639 .590 .617 .579 .604 .637 .611 .575 .650 .714 .601 .546 .685 .655 .678 .604 .645 .547 .546 .705 .623 .499 .582 .659 .734 .648 .597 .635 .520 .670 .686 .593 .700 .588 .678 .674 .710 .668 .674 .626 .775 .689 .601 .622 .656 .692 .689 .796 .693 .766 .638 .688 .667 .598 .727 .629 .687 Table Table 17.5 diagonal Scale 254 R.E. Boyatzis et al.

ST .793

.747

.768

.736

.769

.718 = 1,398) n ] (

.823

.724

.726

.779 square root of the AVE on the diagonal AVE of the root square

.769

.731

.729 SD CR AVE A AO CFM CM EMP ESA ESC INF IL OA PO PR μ ESCI-U self: Convergent and discriminant validity using PLS [ and discriminant validity ESCI-U self: Convergent A AO CFM 4.010 CM 4.159 3.678 EMP .751 .770 ESA .874 3.717 .850 3.945 ESC .821 .890 .880 3.889 INF .771 .531 .534 IL .884 3.892 .591 .839 .816 OA 3.797 .827 .592 .606 .493 .845 PO .527 .810 3.675 .893 PR .407 4.118 .449 .525 .449 .864 .842 ST .677 4.023 .735 .429 .443 TEAM .353 .878 3.785 .516 .821 .565 .855 4.118 .482 3.841 .453 .876 .446 .591 .584 .878 .742 .345 .541 .787 .554 .863 .384 .527 .546 .591 .850 .423 .547 .871 .445 .559 .534 .563 .517 .533 .309 .439 .629 .482 .485 .540 .525 .510 .471 .454 .455 .544 .431 .657 .458 .495 .440 .326 .429 .490 .421 .428 .526 .463 .468 .489 .418 .339 .484 .378 .376 .525 .491 .391 .375 .369 .445 .514 .383 .456 .455 .649 .414 .460 .631 .468 .343 .448 .433 .546 .367 .536 .564 .510 .534 .431 .410 .589 .455 .393 .550 .446 .333 .457 .359 .339 .674 .424 Scale Table Table 17.6 17 Emotional and Social Intelligence and Behavior 255

ST .811

.786

.800

.780

.811

= 8,981) .776 n ] (

.841

.765

.811

.812 square root of the AVE on the diagonal AVE of the root square

.814

.792

.777 SD CR AVE A AO CFM CM EMP ESA ESC INF IL OA PO PR μ ESCI-U other: Convergent and discriminant validity using PLS [ and discriminant validity ESCI-U other: Convergent Scale A AO CFM 4.192 CM 4.330 3.969 EMP .776 .753 ESA .861 4.079 .884 4.034 ESC .871 .875 .911 4.031 INF .838 .604 .628 IL 4.134 .906 .841 .663 .885 OA 4.033 .820 .747 .659 .875 .640 PO .657 .816 4.032 .906 PR .655 4.290 .585 .580 .677 .871 .884 ST .707 4.241 .751 .658 .643 TEAM .602 3.973 .906 .603 .769 .682 .886 4.278 .658 4.046 .607 .835 .671 .658 .696 .899 .776 .575 .608 .792 .694 .890 .671 .695 .657 .639 .907 .597 .742 .885 .634 .618 .673 .725 .704 .661 .555 .691 .658 .708 .679 .691 .700 .694 .701 .674 .623 .714 .623 .793 .697 .645 .687 .598 .659 .647 .647 .704 .693 .657 .630 .692 .641 .621 .668 .572 .658 .730 .652 .647 .652 .595 .660 .736 .625 .667 .656 .758 .630 .642 .724 .663 .553 .675 .650 .690 .591 .729 .711 .678 .696 .676 .669 .758 .672 .648 .731 .672 .613 .704 .616 .575 .773 .614 Table Table 17.7 256 R.E. Boyatzis et al. other in Table 17.7 . For the ESCI other analysis, predicted effectiveness of branch chiefs in the sample size exceeded available memory, so research (middle-level managers) at NASA. half of the sample was randomly chosen for this Williams (2008 ) showed they predicted effective- analysis. In the ESCI self, some items were ness of elementary, middle, and high school dropped from to conform to desirable loadings. principals in a large urban city. Boyatzis and Ratti The specifi c items dropped for the ESCI self (2009 ) made the same links in samples from a analysis included seven reverse-scored items large company in Italy and top executives from (reverse-scored items for emotional self-control, cooperatives also in Italy. Ryan et al. ( 2009 ) coaching and mentoring, achievement orienta- reported a signifi cant prediction of effectiveness tion, inspirational leadership, teamwork, adapt- of the EI and SI competencies in executives in a ability, and empathy). For this same analysis, six variety of European companies. Gutierrez et al. other items were dropped: one from emotional ( 2012) showed the links to effectiveness of coded self-awareness, one from coaching and mentor- EI and SI competencies in top executives in India ing, one from adaptability, one from confl ict and China and compared them to comparable management, and one from infl uence. Each of samples from Western Europe. these items has been reworded to possibly correct Another benefi t of assessing EI and SI at the for the loadings in the latest revision to the behavioral level is that it should be more amena- ESCI. For the ESCI other analysis, only one ble to change and improvement than other levels. empathy item was dropped. For research pur- To document change, EI and SI competencies poses, a list of these specifi c items is available were shown to signifi cantly improve for four from the fi rst author. No items had to be trimmed cohorts of full-time MBA and two cohorts of for the ESCI-U self or other. part-time MBA students as compared to two cohorts of each type of MBA in a baseline study in earlier years (Boyatzis et al. 2002 ) in the USA Validity of the Behavioral Level and in Italy (Camuffo et al. 2009 ). with Coded Interviews

One form of assessment of the behavioral level Validity of the Behavioral Level of EI and SI is the critical incident interview, with Early Versions of the ECI or ECI-U also known as the behavioral event interview (Flanagan 1954 ; Boyatzis 1982 ; Spencer and Using a 360 or multisource assessment as a way Spencer 1993 ). Work samples are collected and to measure EI and SI in behavior—as seen by then coded by blind coders with high inter-rater others—provides an easier assessment tool. It is reliability (Boyatzis 1998). Boyatzis ( 1982 ) pro- less costly in human effort to collect and score vided the fi rst published validation of these com- and provides more of a consensual validation that petencies against effectiveness measures in a improves face validity of the results. It is a test of sample of 253 managers and executives from six the behavioral level of EI and SI. large private sector organizations and six large Using a company-customized variation of the government organizations which included sam- earlier version of the ESCI, called the SAQ/EAQ ples from the US Navy and Marine Corps. (Boyatzis 2009 ), Boyatzis ( 2006) conducted a Although many studies followed in the 1980s and tipping point analysis and showed the operating 1990s, these studies were often done by consul- profi t contribution of senior partners per year in tants and were not published. a longitudinal study of executive performance in a A series of studies done in the 1990s were pub- major international consulting company. When lished later, showing the link between the use of senior partners used EI competencies above the the EI and SI competencies and work effective- tipping point (as seen by others), as compared to ness using the critical incident interviews. Dreyfus those senior partners showing the competencies (2008 ) showed that the EI and SI competencies below the tipping point, they showed 78 % high 17 Emotional and Social Intelligence and Behavior 257 gross operating profi t from their accounts for the African call centers, the EI/Si as seen by others of EI cluster of self-management and 390 % increase 135 call agents was signifi cantly related to job for the EI cluster which they called self- performance (Nel 2001 ). A study of 33 business regulation. In the same study, the increase in development managers at Bass Brewers in the UK operating profi t from senior partners above the showed that EI/SI, as seen by others, was signifi - tipping point (again, as seen by others) in SI cantly predictive of a multidimensional measure competencies was 110 %. Even the operating of job performance (Lloyd 2001 ). All of these profi t contribution of senior partners was 50 % studies used the ECI-2. for those above the tipping point in demonstrated Van Oosten (2013) showed that demonstrating cognitive intelligence competencies. more EI and SI competencies, as measured with Using the earlier version of the ESCI, called the ECI-2, predicted performance of bank execu- the ECI-2, Hopkins and Bilimoria (2008 ) showed tives. Her fi ndings suggested that certain of these that female bank executives showed that the use competencies, the ones associated with leading of the EI and SI competencies (as seen by others) change were the most effective. In a study of fi rst was associated with higher measures of success responders, fi re fi ghters and offi cers in London, in terms of the company’s performance appraisal Stagg and Gunter (2002 ) showed that EI/SI system. Koman and Wolff (2008 ) showed a simi- (again, as seen by others) was statistically related lar pattern for US Navy Commanding Offi cers of to a battery of performance measures. fl ight crews when assessed against actual perfor- In comparing performance against a tradi- mance of the fl ight in combat simulations and air tional measure of academic cognitive intelli- maneuvers. EI/SI as seen by others was signifi - gence, Victoroff and Boyatzis (2013 ) showed cantly related to job performance at Johnson & that assessment of the EI and SI competencies, as Johnson in a study of 358 managers (Cavallo and seen by other dental graduate students 6 weeks Brienza 2002 ). into their program, predicted their grades in the In another domain, Boyatzis et al. (2011 ) third and fourth year courses, which all take place showed that Catholic Pastor priests who demon- in the dental clinic working with patients under strated more EI and SI competencies as seen by faculty supervision. Meanwhile, the Dental others working with them showed higher parish- Admissions Test (DAT) predicted grades in the ioner satisfaction on a complex measure of eight fi rst and second years of the graduate dental pro- dimensions of parishioner involvement in the gram which are all didactic courses, but showed church and community. It did not predict church no prediction of their grades in their third and attendance nor donations per family. fourth clinical years. In a study of Spanish nonprofi t executives, This result supports the fi nding of discriminant Ramo et al. (2009 ) showed that using more EI and validity of EI/SI competencies with traditional mea- SI competencies predicted effectiveness. Camuffo sures of cognitive intelligence. Murensky ( 2000 ) et al. ( 2012) showed a similar set of fi ndings for studied 90 oil company executives and found no Italian executives from a variety of companies in relationship between Watson-Glaser Test of Critical Northern Italy. (Nel 2001 ; Aliaga and Taylor Thinking and the EI and SI competencies. She 2012 ) showed that using more EI and SI resulted found three of the SI competencies to have a nega- in more effectiveness in Peruvian managers in a tive correlation to the Watson-Glaser. In Philippine copper refi nery. Sharma ( 20121) showed that plants of two multinationals, Sergio (2001 ) showed Indian middle-level managers from large compa- that others’ assessments of 134 plant supervisors’ nies and public sector organizations who demon- EI/SI was signifi cantly linked to job performance. strated more EI and SI than others were more A mental ability test was also linked to perfor- effective. Sevinc (2001 ) reported EI/SI scores (as mance, but the mental ability test and EI/SI were not seen by others) were signifi cantly correlated with correlated. These studies support the idea that salary, job, and life satisfaction of 71 Turkish pro- behavioral EI/SI is a different human capability or fessionals in the fi nancial services sector. In South talent than cognitive intelligence. 258 R.E. Boyatzis et al.

Again, assessment at this level of EI and SI created a positive organizational climate (as seen allows for careful studies of change. Boyatzis by their subordinates), while 78 % of the leaders et al. (2002 ) reported on one cohort of graduating with low emotional self-awareness created a full-time MBAs with the SAQ/EAQ; Boyatzis negative climate. and Saatcioglu (2008 ) were able to add another Also in the sales function, Lisicki (2011 ) cohort of full-time MBAs using the SAQ/EAQ showed that EI and SI competencies predicted and two more cohorts using the ECI-U (the ear- sales performance of sales people in a pharma- lier version of the ESCI-U); and Boyatzis et al. ceutical company. He showed that coaching ( 2010 ) added yet another cohort of full-time and mentoring competency was the most potent MBAs with the ECI-U showing dramatic improve- factor in predicting sales performance. ment on the EI and SI competencies, as compared A leader’s EI and SI, as assessed with the to baseline years and the MBAs themselves at ESCI, is associated with their subordinates’ job entry into the program. performance and satisfaction in a study of 20 leaders form a multinational in Egypt (Shams 2008). She found that subordinates’ job satisfac- Validity of the Behavioral Level tion was signifi cantly correlated with all of the EI with the ESCI or ESCI-U competencies and four of the seven SI competencies. Meanwhile, subordinates’ job performance The ESCI and ESCI-U refl ected several major was correlated with emotional self-awareness improvements on the earlier versions of the test and emotional self-control from the EI cluster of EI and SI at the behavioral level. First launched and infl uence from the SI cluster. in 2007, the validation research is just emerging. In a sample of Indian managers, Badri (2013 ) Studying the effectiveness of sales executives in a showed that EI and SI competencies as seen by fi nancial services company, Boyatzis et al. (2012) others assessed with the ESCI predicted leader found that demonstrated EI and SI predicted effectiveness and use of the transformational effectiveness. They were able to assess cognitive leadership style. He found a relationship to trans- intelligence using Ravens Progressive Matrices actional leadership style as well, but only for one and personality using the NEO-PR and found subscale, and others were negatively related to that EI/SI competencies as seen by others pre- use of EI and SI. This supported the fi ndings of dicted effectiveness, but cognitive g and person- Piel (2008 ) in a study of 82 project managers ality did not. In addition, they were able to focus, from various companies that EI and SI, as through multiple regressions, on the fact that assessed with the ESCI, was related to using the adaptability was the most powerful predictor of transformational leadership style. leader effectiveness among the EI competencies Using the self-assessment from the ESCI, and infl uence was among the SI competencies. Quinn (2013 ) showed that it predicted physician In a study of leaders from a variety of compa- leader effectiveness in terms of participation as a nies, Havers (2010 ) showed “leaders who dem- leader in hospitals. Pittenger ( 2012 ) showed that onstrated fewer than three ESCI strengths drew for IT managers and advanced professionals from upon a limited range of leadership styles, tending a variety of companies, EI and SI competencies to rely primarily on the coercive style … In con- predicted effectiveness in terms of engagement. trast, leaders with 10 or more ESCI strengths Using a 360 from their own competency model used a much wider range of leadership styles, of EI and SI, Young and Dulewicz ( 2009 ) showed including those likely to engage their team mem- the same pattern found earlier with the US Navy in bers, providing long term direction and vision, that EI and SI as seen by others predicted leader creating harmony, encouraging new ideas and effectiveness for British Naval offi cers. Ryan et al. investing in others development” (p. 2). Further- (2012 ) showed that EI and SI with a 360 from a more, Havers (2010 ) reported that 92 % of the customized competency assessment in a large leaders showing high emotional self-awareness Swiss company predicted executive effectiveness. 17 Emotional and Social Intelligence and Behavior 259

Bar-On, R. (1997). Bar-On Emotional Quotient Inventory: Conclusion Technical manual . Toronto: Multi-Health Systems. Bar-On, R. (2004). The Bar-On Emotional Quotient Inventory (EQ-i): Rationale, description and summary The behavioral level does provide theoretical and of psychometric properties. In G. Geher (Ed.), Meas- empirical support for the relevance of EI and SI uring emotional intelligence: Common ground and in predicting effectiveness and work and life out- controversy (pp. 111–142). Hauppauge: Nova. Bar-On, R. (2006). The Bar-On model of emotional-social comes. It also shows that at least one measure of intelligence (ESI) 1. Psicothema, 18 (Suplemento), the behavioral level, the ESCI or ESCI-U, satis- 13–25. fi es appropriate psychometric standards of a reli- Boyatzis, R. E. (1982). The competent manager: A model able and valid test, with appropriate convergent for effective performance . New York: Wiley. Boyatzis, R. E. (1998). Transforming qualitative informa- and discriminant validity, and is a good model fi t tion: Thematic analysis and code development . for each of the separate scales. Thousand Oaks: Sage. This explanation of the levels of EI may serve Boyatzis, R. E. (2006). Using tipping points of emotio- to refocus the intellectual debate on the condi- nal intelligence and cognitive competencies to predict fi nancial performance of leaders. Psicothema, 18 , tions when the relationship of EI (and SI) helps 124–131. us to understand performance, effectiveness, and Boyatzis, R. E. (2009). A behavioral approach to emotional life and job outcomes. It offers a theoretical ratio- intelligence. Journal of Management Development, nale as to how the three approaches to EI do func- 28 (9), 749–770. Boyatzis, R. E., & Gaskin, J. (2010). A technical note on tion in unison within the person. But the different ESCI and ESCI-U: Factor structure, reliability, con- levels require different types of measurement. vergent and discriminant validity using EFA and CFA . This is where the ESCI can complement the Boston: The Hay Group. MSCEIT and EQ-i in research. Boyatzis, R. E., & Goleman, D. (1996/1999). Emotional competency inventory. Boston: The Hay Group. Practical implications of the behavioral level Boyatzis, R. E., & Ratti, F. (2009). Emotional, social and are considerable. It is far easier to document cognitive intelligence competencies distinguishing improvement on the behavioral level than other effective Italian managers and leaders in a private levels as has been shown in published research. company and cooperatives. Journal of Management Development, 28 (9), 821–838. The behavioral level guides coaches and trainers Boyatzis, R. E., & Saatcioglu, A. (2008). A twenty year seeking to improve performance of executives, view of trying to develop emotional, social and cogni- managers, and leaders, along with professionals tive intelligence competencies in graduate manage- in organizations. It can be a guide and tool for ment education. Journal of Management Development, 27 (3), 92–108. graduate and undergraduate programs seeking to Boyatzis, R. E., Stubbs, E. C., & Taylor, S. N. (2002). develop the whole person, not just their knowl- Learning cognitive and emotional intelligence compe- edge, and help such academic programs with the tencies through graduate management education. needed outcome assessment for accreditation and Academy of Management Journal on Learning and Education, 1 (2), 150–162. documentation of program impact. Boyatzis, R. E., Lingham, A., & Passarelli, A. (2010). Inspiring the development of emotional, social, and cognitive intelligence competencies in managers. In M. Rothstein & R. Burke (Eds.), Self-management and References leadership development (pp. 62–90). Cheltenham, UK: Edward Elgar Publishers. Aliaga Araujo, S. V., & Taylor, S. N. (2012). The infl uence Boyatzis, R. E., Brizz, T., & Godwin, L. (2011). The of emotional and social competencies on the perfor- effect of religious leaders’ emotional and social com- mance of Peruvian refi nery staff. Cross Cultural petencies on improving parish vibrancy. Journal Management, 19 (1), 19–29. of Leadership and Organizational Studies, 18 (2), Badri, N. (2013). An examination of the relationship 192–206. between emotional intelligence leadership styles and Brackett, M. A., & Mayer, J. D. (2003). Convergent, dis- leadership effectiveness. Unpublished doctoral disser- criminant, and incremental validity of competing mea- tation, Jaypee Institute of Information Technology. sures of emotional intelligence. Personality and Social Barclay, D., Higgins, C., & Thompson, R. (1995). The par- Psychology Bulletin, 29 (9), 1147–1158. tial least squares (PLS) approach to causal modeling: Bray, D. W., Campbell, R. J., & Grant, D. L. (1974). Personal computer adoption and use as an illustration. Formative years in business: A long term AT&T study Technology Studies, 2 (2), 285–309. of managerial lives . New York: Wiley. 260 R.E. Boyatzis et al.

Campbell, J. P., Dunnette, M. D., Lawler, E. E. I. I. I., & Flanagan, J. C. (1954). The critical incident technique. Weick, K. E., Jr. (1970). Managerial behavior, perfor- Psychological Bulletin, 51 , 327–335. mance, and effectiveness . New York: McGraw-Hill. Fornell, C. A. (1982). Second generation of multivariate Camuffo, A., Gerli, F., Borgo, S., & Somia, T. (2009). The analyses: Measurement and evaluation methods . effects of management education on careers and com- New York: Praeger. pensation: A competency-based study of an Italian MBA Furnham, A., & Petrides, K. V. (2003). Trait emotional intel- programme. Journal of Managemetn Development, ligence and happiness. Social Behavior and Personality: 28 (9), 839–858. An International Journal, 31 (8), 815–823. Camuffo, A., Gerli, F., & Gubitta, P. (2012). Competencies Gignac, G. E. (2005). Evaluating the MSCEIT V2. 0 via matter: Modeling effective entrepreneurship in north CFA: Comment on Mayer et al. (2003). Emotion, 5 (2), east of Italy small fi rms. Cross Cultural Management, 233–235. 19 (1), 48–66. Gignac, G. E., Palmer, B. R., Manocha, R., & Stough, C. Cavallo, K., & Brienza, D. (2002). Emotional competence (2005). An examination of the factor structure of the and leadership excellence at Johnson & Johnson: Schutte Self-Report Emotional Intelligence (SSREI) The emotional intelligence and leadership study . scale via confi rmatory factor analysis. Personality and Paper downloaded on 2 May 2002, from http://www. Individual Differences, 39 (6), 1029–1042. eiconsortium.org/ Goleman, D. (1995). Emotional intelligence . New York: Cherniss, C. (2010). Emotional intelligence: Toward clari- Bantam Books. fi cation of a concept. Industrial and Organizational Goleman, D. (1998). Working with emotional intelligence . Psychology, 3 (2), 110–126. New York: Bantam. Cherniss, C., & Boyatzis, R. E. (2013). Using a multi- Grubb, W. L., III, & McDaniel, M. A. (2007). The fak- level theory of performance based on emotional intel- ability of Bar-On’s emotional quotient inventory short ligence to conceptualize and develop “soft” leader form: Catch me if you can. Human Performance, skills. In R. Riggio & S. Tan (ed.). Building 20 (1), 43–59. Interpersonal Skills in Management Programs . 53–72. Gutierrez, B., Spencer, S. M., & Zhu, G. (2012). Thinking Chin, W. W., Marcolin, B. L., & Newsted, P. R. (2003). globally, leading locally: Chinese, Indian, and Western A partial least squares latent variable modeling leadership. Cross Cultural Management, 19 (1), 67–89. approach for measuring interaction effects: Results Hair, J., Anderson, R., Tatham, R., & Black, W. (1995). from a Monte Carlo simulation study and an elec- Multivariate data analysis: With readings . Upper tronic-mail emotion/adoption study. Information Saddle River: Prentice-Hall, Inc. Systems Research, 14 (2), 189–217. Havers, G. (2010). EI at the heart of performance: The Ciarrochi, J. V., Caputi, P., & Mayer, J. D. (2003). The dis- implications of our 2010 ESCI research . London: The tinctiveness and utility of a measure of trait emotional Hay Group. awareness. Personality and Individual Difference, 34 , Hay Group. (2011). Emotional and social competency 1477–1490. inventory: A user guide for accredited practitioners . Conte, J. M. (2005). A review and critique of emotional Boston: The Hay group. intelligence measures. Journal of Organizational Higgs, M. J., & Dulewicz, V. (2002). Making sense of Behavior, 26 (4), 433–440. emotional intelligence (2nd ed.). Windsor: NFER- Conte, J. M., & Dean, M. A. (2006). Can emotional intel- Nelson. ligence be measured? In K. R. Murphy (Ed.), A cri- Hopkins, M., & Bilimoria, D. (2008). Social and emo- tique of emotional intelligence: What are the problems tional competencies predicting success for female and and how can they be fi xed? (pp. 59–77). Mahwah: male executives. Journal of Management Development, Lawrence Erlbaum. 12 (1), 13–35. Dawda, D., & Hart, S. D. (2000). Assessing emotional Howard, A., & Bray, D. (1988). Managerial lives in tran- intelligence: Reliability and validity of the Bar-On sition: Advancing age and changing times . New York: Emotional Quotient Inventory (EQ-i) in university Guilford Press. students. Personality and Individual Differences, 28 , Joseph, D. L., & Newman, D. A. (2010). Emotional intel- 797–812. ligence: An integrative meta-analysis and cascading Dreyfus, C. (2008). Identifying competencies that predict model. Journal of Applied Psychology, 95 (1), 54–78. effectiveness of R and D managers. Journal of Koman, L., & Wolff, S. (2008). Emotional intelligence Management Development, 27 (1), 76–91. competencies in the team and team leader. Journal of Dulewicz, V., & Higgs, M. J. (1999). Emotional intelli- Management Development, 12 (1), 56–75. gence: A review and evaluation study. Journal of Kotter, J. P. (1982). The general managers . New York: Managerial Psychology, 15 (4), 341–368. Free Press. Dulewicz, V., Higgs, M., & Slaski, M. (2003). Measuring Law, K. S., Wong, C., & Song, L. J. (2004). The construct emotional intelligence: Content, construct and and criterion validity of emotional intelligence and its criterion-related validity. Journal of Managerial potential utility in management research. Journal of Psychology, 18 (5), 405–420. Applied Psychology, 87 (3), 483–496. Fernández-Berrocal, P., & Extremera, N. (2006). Emotional Leeper, R. W. (1948). A motivational theory of emotion to intelligence: A theoretical and empirical review of its replace emotion as disorganized response. Psychological fi rst 15 years of history. Psychotema, 18 , 7–12. Review, 55 (1), 5–21. 17 Emotional and Social Intelligence and Behavior 261

Lim, K., Sia, C., Lee, M., & Benbasat, I. (2006). Do I trust Murphy, K. R. (2006). Four conclusions about emotional you online, and if so, will I buy? An empirical study of intelligence. In K. R. Murphy (Ed.), A critique of emo- two trust-building strategies. Journal of Management tional intelligence: What are the problems and how Information Systems, 23 (2), 233–266. can they be ﬁ xed? (pp. 345–354). Mahwah: Lawrence Lisicki, J. M., Jr. (2011). An examination of the relation- Erlbaum. ship between emotional intelligence and work related Nel, H. (2001). An industrial psychological investigation outcomes . An unpublished doctoral dissertation, into the relationship between emotional intelligence Capella University. and performance in the call centre environment . Lloyd, M. (2001). Emotional intelligence and Bass Unpublished master’s thesis, University of Stellenbosch, Brewers Ltd . Dissertation, Nottingham Business School. Department of Industrial Psychology. Luthans, F., Hodgetts, R. M., & Rosenkrantz, S. A. (1988). Nunnally, J., Bernstein, I., & Berge, J. (1994). Real managers . Cambridge, MA: Ballinger Press. Psychometric theory . New York: McGraw-Hill. MacCann, C., & Roberts, R. D. (2008). New paradigms Palmer, B. R., Gignac, G., Manocha, R., & Stough, C. for assessing emotional intelligence: Theory and data. (2005). A psychometric evaluation of the Mayer- Emotion, 8 (4), 540–551. Salovey-Caruso emotional intelligence test version Matthews, G., Zeidner, M., & Roberts, R. D. (2002). 2.0. Intelligence, 33 (3), 285–305. Emotional intelligence. Science and myth . Cambridge, Petrides, K. V., & Furnham, A. (2000). On the dimen- MA: The MIT Press. sional structure of emotional intelligence. Personality Matthews, G., Zeidner, M., & Roberts, R. D. (2004). and Individual Differences, 29 , 313–320. Emotional intelligence: Science and myth . Cambridge, Petrides, K. V., & Furnham, A. (2001). Trait emotional MA: MIT Press. intelligence: Psychometric investigation with refer- Matthews, G., Emo, A. K., Funke, G., Zeidner, M., ence to established trait taxonomies. European Journal Roberts, R. D., Costa, P. T., Jr., & Schulze, R. (2006). of Personality, 17 , 425–448. Emotional intelligence, personality, and task-induced Petrides, K. V., & Furnham, A. (2003). Trait emotional stress. Journal of Experimental Psychology: Applied, intelligence: Behavioral validation in two studies of 12 (2), 96–107. emotion recognition and reactivity to mood induction. Mayer, J. D., & Salovey, P. (1997). What is emotional European Journal of Personality, 17 , 39–75. intelligence? In P. Salovey & D. J. Sluyter (Eds.), Petrides, K. V., Frederickson, N., & Furnham, A. (2004). Emotional development and emotional intelligence The role of trait emotional intelligence in academic (pp. 3–34). New York: Basic Books. performance and deviant behavior at school. Mayer, J. D., Salovey, P., & Caruso, D. R. (1999). Personality and Individual Differences, 36 , 277–293. Emotional intelligence meets traditional standards for Petrides, K. V., Pita, R., & Kokkinaki, F. (2007). The loca- an intelligence. Intelligence, 2 , 267–298. tion of trait emotional intelligence in personality factor Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G. space. British Journal of Psychology, 98 (2), 273–289. (2003). Measuring emotional intelligence with the Piel, M. A. (2008). Emotional intelligence and critical MSCEIT V2.0. Emotion, 3 , 97–105. thinking relationships to transformational leadership . Mayer, J. D., Roberts, R. D., & Barsade, S. G. (2008). Unpublished doctoral dissertation, University of Human abilities: Emotional intelligence. Annual Phoenix. Review of Psychology, 59 , 507–536. Pittenger, L. (2012). IT professionals: Understanding McClelland, D. C. (1973). Testing for competence rather what impacts their desire to engage in work. PhD, than intelligence. American Psychologist, 28 (1), 1–40. Case Western Reserve University, Cleveland. McClelland, D. C. (1985). Human motivation . Glenview: Quinn, J. (2013). The importance of compassion and Scott Foresman. vision in physician leadership. PhD, Case Western McClelland, D. C. (1998). Identifying competencies with Reserve University, Cleveland. behavioral event interviews. Psychological Science, 9 , Ramo, L., Saris, W., & Boyatzis, R. E. (2009). The impact 331–339. of emotional and social competencies on effectiveness Mikolajczak, M., Luminet, O., & Menil, C. (2006). of Spanish executives. Journal of Management Predicting resistance to stress: Incremental validity of Development, 28 (9), 771–793. trait emotional intelligence over alexithymia and opti- Roberts, R. D., Schulze, R., O’Brien, K., MacCann, C., mism. Psicothema, 18 (Suplemento), 79–88. Reid, J., & Maul, A. (2006). Exploring the validity of Mikolajczak, M., Luminet, O., Leroy, C., & Roy, E. the Mayer-Salovey-Caruso Emotional Intelligence (2007). Psychometric properties of the Trait Emotional Test (MSCEIT) with established emotions measures. Intelligence Questionnaire: Factor structure, reliability, Emotion, 6 , 663–669. construct, and incremental validity in a French- Rossen, E., & Kranzler, J. H. (2009). Incremental validity speaking population. Journal of Personality Assess- of the Mayer–Salovey–Caruso Emotional Intelligence ment, 88 (3), 338–353. Test Version 2.0 (MSCEIT) after controlling for Murensky, C. L. (2000). The relationship between emo- personality and intelligence. Journal of Research in tional intelligence, personality, critical thinking Personality, 43 (1), 60–65. ability, and organizational leadership performance at Rossen, E., Kranzler, J. H., & Algina, J. (2008). upper levels of management. Dissertation, George Conﬁ rmatory factor analysis of the Mayer–Salovey– Mason University. Caruso Emotional Intelligence Test V 2.0 (MSCEIT). 262 R.E. Boyatzis et al.

Personality and Individual Differences, 44 (5), job satisfaction . Unpublished bachelor dissertation, 1258–1269. German University, Cairo. Ryan, G., Emmerling, R. J., & Spencer, L. M. (2009). Sharma, R. (2012). Measuring social and emotional intel- Distinguishing high performing European executives: ligence competencies in the Indian context. Cross The role of emotional, social and cognitive compe- Cultural Management, 19 (1), 30–47. tencies. Journal of Management Development, 28 (9), Spector, P. E., & Johnson, H. M. (2006). Improving the 859–875. defi nition, measurement, and application of emotional Ryan, G., Spencer, L. M., & Bernhard, U. (2012). Develo- intelligence. In K. R. Murphy (Ed.), A critique of emo- pment and validation of a customized competency- tional intelligence: What are the problems and how based questionnaire: Linking social, emotional, and can they be fi xed? (pp. 325–344). Mahwah: Lawrence cognitive competencies to business unit profi tability. Erlbaum. Cross Cultural Management, 19 (1), 90–103. Spencer, L. M., Jr., & Spencer, S. M. (1993). Competence Saklofske, D. H., Austin, E. J., & Minski, P. S. (2003). at work: Models for superior performance . New York: Factor structure and validity of a trait emotional intelli- Wiley. gence measure. Personality and Individual Differences, Stagg, G., & Gunter, D. (2002). Emotional intelligence in the 34 , 707–721. fi re service . Unpublished Paper, London Fire Brigade. Salovey, P., & Mayer, J. D. (1990). Emotional intel- Sternberg, R. J. (1996). Successful intelligence: How prac- ligence. Imagination, Cognition and Personality, 9 , tical and creative intelligence determine success in life . 185–211. New York: Simon and Shuster. Salovey, P., & Mayer, J. D. (1997). What is emotional Thornton, G. C., III, & Byham, W. C. (1982). Assessment intelligence? In P. Salovey & D. J. Sluyter (Eds.), centers and managerial performance . New York: Emotional development and emotional intelligence: Academic. Educational implications . New York: Basic Books. Victoroff, K., & Boyatzis, R. E. (2013). An examination Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D., of the relationship between emotional intelligence and Cooper, J. T., Golden, C. J., & Dornheim, L. (1998). dental student clinical performance. Journal of Dental Development and validation of a measure of emotional Education, 77 (4), 416–426. intelligence. Personality and Individual Differences, 25 , Williams, H. (2008). Characteristics that distinguish out- 167–177. standing urban principals. Journal of Management Sergio, R. P. (2001). Emotional intelligence and mental Development, 27 (1), 36–54. ability as determinants of job performance among Wolff, S. B. (2007). Emotional and social competency plant supervisors in selected manufacturing fi rms . inventory: Technical manual up-dated ESCI research Unpublished master’s thesis, De La Salle University- titles and abstracts . Boston: The Hay Group. Dasmariñas, Behavioral Sciences Department, Wong, C. S., & Law, K. S. (2002). The effect of leader and Philippines. follower emotional intelligence on performance and Sevinc, L. (2001). The effect of emotional intelligence on attitude: An exploratory study. Leadership Quarterly, career success: Research on the 1990 graduates of 13 , 243–274. business administration faculty of Istanbul University . Young, M., & Dulewicz, V. (2009). A study into leader- Unpublished master thesis, Istanbul University. ship and management competencies predictive supe- Shams, M. (2008). The impact of the leaders’ emotional rior performance in the British Royal Navy. Journal of intelligence on the employee’s job performance and Management Development, 28 (9), 794–820. Intelligence as a Malleable Construct 18

Lisa S. Blackwell , Sylvia Rodriguez , and Belén Guerra-Carrillo

tions of this research for educational policy and Introduction practice. Why do people care so much about the nature The academic debate about the nature of of intelligence? Traditionally, particularly in intelligence resonates broadly not only with edu- western cultures, intelligence has been seen as cational practitioners and policy makers but also the golden ticket to success. If you had a good with the general public. Most people think they amount of it, you would be rewarded with educa- know what intelligence is and that they know it tional, professional, and fi nancial success, and when they see it. But what do people mean when those with a great deal—the geniuses among they talk about being smart, brilliant, or clever? us—would attain eminence and make a mark on And why does it matter so much? posterity. Implicit in this view is the idea that In this chapter, we focus on research showing intelligence is a “gift”—an innate attribute that how the way that people think about intellectual one possesses in a relatively fi xed quantity, for ability drives the choices they make and the out- better or worse. Historically, the relatively high comes they achieve, sometimes in counterintui- stability in individual performance on intellectual tive ways. We show how a person’s concept of assessments over time and across tasks has led intelligence can impact both their performance on many to assume that this view of intelligence as cognitive tasks in the short run and their achieve- fi xed is correct (see, e.g., Bartels et al. 2002 ; ment over the longer term, and why this is so. Canivez and Watkins 1998; Herrnstein and We review the evidence from cognitive neuroscience Murray 1994 ; Hertzog and Schaie 1986 ), despite for these different conceptions of intelligence. the strong dissent of original developers of the Finally, we discuss how such concepts can be fi rst IQ test, Alfred Binet and Theodore Simon infl uenced and changed and the practical implica- ( Binet 1975 ; Wolf 1973 ). Unfortunately, in this case, believing may make it so. Because our society has presumed that intellectual gifts are innate and could be measured L. S. Blackwell , Ph.D. (*) • S. Rodriguez, Ph.D. Mindset Works, Inc. , 340 S. Lemon Ave. #6463 , accurately, our education system has traditionally Walnut , CA 91789 , USA been structured to identify those students with e-mail: [email protected] apparent above-average intelligence, enrich their B. Guerra-Carrillo instruction, and track them into ever- greater 132 Barker Hall , University of California at Berkeley , opportunities, while those with presumed below- Berkeley , CA 94702 , USA average ability were channeled into programs that e-mail: [email protected]

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 263 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_18, © Springer Science+Business Media New York 2015 264 L.S. Blackwell et al. would prepare them for lower-skilled jobs intellectual ability and skill, their beliefs about (Borland 2003 , 2005 ; Borland and Wright 2001 ; intelligence shape their responses to intellectual Darling-Hammond 1994, 1995; Kaufman 2013 ; challenge. For those who hold a fi xed mindset, Nisbett 2009 ). The result is that those who score the conception of intelligence as a fi xed, uncon- well on performance measures early on are in fact trollable quantity (of which they may have a lot generally offered more opportunities to cultivate or a little) orients them toward measuring and their intellectual ability than those who do not obtaining a positive evaluation of their ability. and, at least partly as a result, often do become Thus, their primary goal is usually to perform more skilled and successful, reinforcing our com- well in order to appear smart—or at least to avoid mon paradigm of innate ability. performing poorly and looking dumb (Blackwell It turns out that something similar happens in et al. 2007; Dweck and Leggett 1988 ). They tend the psychology of the individual as well. Our to think that things come easily if one is smart research, and that of many colleagues, shows that and that effort is both a sign of low ability and people’s “theory” of intelligence—whether they relatively ineffective in overcoming it (e.g., believe it to be fi xed or a malleable quality— Blackwell et al. 2007 ; Hong et al. 1999 ). When infl uences the learning opportunities they will they experience a setback or failure, they are pursue, the effort they will invest, and their result- likely to attribute it to low ability rather than ing growth. It can even impact how their brains effort (e.g., Henderson and Dweck 1990 ), doubt function. their ability to recover, and manifest a “helpless” response, withdrawing effort and giving up rather than risking further exposure as unintelligent or Mindsets About Intelligence untalented (e.g., Robins and Pals 2002 ). On the other hand, those who hold a growth In a fi xed mindset (often referred to as an entity mindset, in which intelligence is a malleable qual- theory in the research literature), people believe ity that can be cultivated, are more focused on that their intelligence is relatively fi xed, and there learning (thus increasing their ability) as a goal, is not much that they can do to develop it. They even if it requires effort, struggle, and errors along agree with statements such as “you can learn new the way (Dweck 1999 ; Dweck and Leggett 1988 ). things, but you cannot really change your basic They consider effort to be a pathway to develop- amount of intelligence.” In contrast, in a growth ment (e.g., Hong et al. 1999 ), and when they expe- mindset (also known as an incremental theory ), rience setbacks, they attribute them to lack of people believe that their intelligence is something suffi cient effort and in turn adopt a mastery- oriented they can change and develop incrementally over approach, increasing their effort and taking on new time. They agree with statements such as “you study strategies (e.g., Robins and Pals 2002 ). can always greatly change how intelligent you Thus, the different mindsets about intelligence are” (Dweck 1999 ). As we will show, these dif- set up different frameworks or “meaning systems” ferent mindsets about intelligence drive the goals (Hong et al. 1999 ) for interpreting situations that that people hold, the challenges they will tackle, involve learning, effort, challenge, and evaluation. the effort they will expend, their persistence in Furthermore, it is when making a transition to a the face of diffi culty, and, as a result, their perfor- situation that poses ongoing, increasing challenge mance and achievement over time (Blackwell (where success is more diffi cult and less certain) et al. 2007; Dweck 1999 ; Dweck and Leggett that these mindsets have the greatest impact on 1988 ; Henderson and Dweck 1990 ). behavior and achievement. In a comprehensive longitudinal study with urban, largely minority students, we examined Mindsets and Motivation how students’ mindsets set up contrasting motivational frameworks and academic outcomes as Over the past few decades, a wealth of research they made their way through a challenging transi- shows that, even when people demonstrate equal tion to junior high school (Blackwell et al. 2007 ). 18 Intelligence as a Malleable Construct 265

We studied three waves of students over three performing better, and these diverging trajecto- successive years, assessing their mindsets at the ries continued over the next two years, widening beginning of their seventh grade year and then the gap between the two groups each term following each wave as they made their way (Fig. 18.1 ). through the following two years of school. First, We examined the pathway from mindset to we examined how their mindsets were related to achievement outcomes using hierarchical linear their goals in school, their attitudes toward effort, modeling and found that the beliefs, goals, and and their responses to failure. Analyses showed attitudes that led to different patterns of behavior that, as found in prior studies, students with a were responsible for the diverging trajectories of growth mindset had stronger learning goals than grades. The increasing challenge level, particu- the fi xed mindset students—for example, they larly in the math curriculum of a health science- said that “It's much more important for me to focused school, spurred the students with a learn things in my classes than it is to get the best growth mindset to focus on learning, work harder, grades”—and had much more positive attitudes and use positive strategies when they encoun- toward effort, agreeing that “when something is tered diffi culty, with the result that they mastered hard, it just makes me want to work more on it, the curriculum better than those who entered not less.” Students with a fi xed mindset, on the with a fi xed mindset, despite the fact that both other hand, were more likely to say that “If you’re groups began with similar skills (Fig. 18.2 ). not good at a subject, working hard won’t make What do these mindsets sound like in the words you good at it,” and “When I work hard at some- of real students? A rising eighth grader with a thing, it makes me feel like I’m not very smart.” growth mindset explained how he thought about How did these two groups of students feel intelligence as a product of one’s choices and about failure? These mindsets, goals, and beliefs behaviors, inextricably tied to learning and effort: about effort in turn predicted how students said Well, you can change it [your intelligence] because they would respond to a poor grade on a quiz: the people are different. One year they can be lazy in growth mindset students showed a clear mastery- school and the other year they’re like, “All right, oriented response, saying that they would “work I’ve got to step it up because I want to get into college.” … What makes me feel smart is participat- harder in this class from now on” and “would ing and doing my homework and everything, spend more time studying for the next test.” because then I know that I’m doing my best. In contrast, many of the students with a fi xed mindset had a helpless response—for example, Asked whether he liked schoolwork that made saying they would “spend less time on this sub- him think hard, he emphasized the value of chal- ject from now on,” with some even admitting that lenge to his growth: they “would try to cheat on the next test” rather Yes, I do, because it gives me a challenge and also than risk another failure! it’ll help me a lot and I can do better with it and everything. Contemplating the prospect of failing a test, Mindsets and Achievement he immediately began seeking solutions based on effort: How did these different mindset frameworks I would feel really bad, but at the same time I impact achievement over this challenging wouldn’t be surprised because maybe the year transition? Based on their prior sixth grade test before that you did really good, and then you know scores, when they were in the less-challenging like you just put that same amount of effort. But elementary school environment, the fi xed and like that year, the new year, it gets harder and everything… Maybe there was some notes that you growth mindset students had similar levels of could write down but you didn’t bother because math skills upon entry into junior high school. you already knew them. Maybe you didn’t have it But by the end of the fi rst term, they began to all memorized, so you forget some of the stuff. I pull apart, with the growth mindset students guess what I would do was maybe work harder, 266 L.S. Blackwell et al.

78 77

76 Incremental (Growth)

Math Grades 71 Entity (Fixed)

68 Fall 7th Grade Spring 7th Grade Fall 8th Grade Spring 8th Grade End of Term Grades

Fig. 18.1 Trajectory of middle school achievement as a grade, the achievement gap was 10 % of the total grade function of student mindset in Study 1 (Notes: Growth points that differentiate a failing grade (60 %) from a per- and ﬁ xed mindset groups entered seventh grade with fect score (100 %). Reprinted from Blackwell et al. 2007 , equal math achievement based on sixth grade test scores p. 251. Reprinted with permission. Copyright 2007 from (not shown). They began to diverge by the end of the fall Society for Research in Child Development, Inc.) term of seventh grade (mid-year). By the end of eighth

and start thinking, Oh, wow. Okay, so I didn’t do done is I would just give up and my friends would so good but maybe this time I can do good. sometimes give the answers to me.

In contrast, his classmate laboring under a When contemplating failure, she shared a ﬁ xed mindset talked about her uncertainty about recent incident and her collapse in the face of her ability to learn and how it made her feel challenge: helpless: I was doing my test and what happened is I was reading this question that I really didn't know … Well I’m going to have to probably agree [that you from there on I just circled randomly and I just com- can’t change your intelligence] because some- pletely gave up on them, even like trying on the test. times – well for me there's limits on what I can learn and what I can’t… I tend to space out a lot. The motivational implications of these two And when I space out it's like the teacher will ask me a question and I have no idea what she’s say- different frameworks, and their resulting impact ing. And so I just have to sit in silence until she on performance and achievement, have been gives up and picks somebody else. demonstrated in many studies spanning kinder- When asked what made her feel smart, she garten through graduate school (Aronson et al. looked to external validation through getting the 2002; Blackwell et al. 2007 ; Dweck and Leggett “right answer” and admitted that she preferred 1998; Good et al. 2003; Heyman et al. 2003 ; things she could do easily versus challenging work: Kray and Haselhuhn 2007 ; Smiley and Dweck 1994; Yeager et al. 2013). Over and over again, Like say I got a question right in front of the whole researchers have shown that the way people class, then that makes me feel like kind of smart and special … I think it’s so much easier and think about their intelligence can become a self- quicker if you know it by heart and you just do it fulﬁ lling prophecy, expanding or limiting their right away and get it over with … Over-thinking motivation, growth, achievement, and, ultimately, sometimes can just really frustrate me. What I’ve their ability. 18 Intelligence as a Malleable Construct 267

Fig. 18.2 Process model depicting the relations between learning strategies and resilient responses to challenge, student mindset, other beliefs and behaviors, and achieve- which in turn predicted greater math achievement gains. ment in Study 1 (Notes: The more ﬁ rmly students held a Reprinted from Blackwell et al. 2007 , p. 253. Reprinted growth mindset (incremental theory), the more they with permission. Copyright 2007 from Society for endorsed learning goals and positive beliefs about effort. Research in Child Development, Inc.) These goals and beliefs were associated with positive

How Malleable Is Intelligence measure a wide variety of knowledge and cogni- Really? tive processes that are highly intercorrelated, such that if you score well on one, chances are Without doubt, people can gain knowledge and that you will also score well on another. James skills through learning, but can they really Flynn (2007 ) explained this calculation with a develop their intelligence as we understand it? clever analogy comparing it to measuring perfor- A robust debate about the true nature of intelligence mance in a decathlon, where performance is continues (see, e.g., Kaufman 2013 ; Nisbett computed from 10 events that each assess a dif- 2009 ), but most people think of intelligence as a ferent ability. For example, strength can be calcu- generalized capacity for learning and reasoning lated from performance on throwing events, that can be assessed by instruments such as IQ tests. while speed can be assessed through sprinting Without weighing in on that complex question, events. Similarly, different subtests of intelli- we can agree that the version of intelligence gence assessments measure cognitive factors measured by standard IQ tests is the result of such as our ability to maintain and manipulate combining scores from various subtests that information in mind (working memory), inte- 268 L.S. Blackwell et al. grate features of and consider relationships change in IQ, with low-SES children showing a between stimuli (reasoning), and process infor- decrease in IQ over time (Breslau et al. 2001 ). mation fl uidly (processing speed), among others. To what extent are scores on assessments of A portion of these subtests may also measure the these sorts malleable, and how does experience accumulation of knowledge about the meaning and learning impact performance on them? Over of words or arithmetic rules (Naglieri and the last few years, research aimed at answering Goldstein 2009 ). Intuitively, one can suspect that this question has provided strong evidence sug- what we do or are exposed to in our daily life gesting that cognitive skills such as those tested could infl uence how well we score on one or by intelligence tests can in fact improve with many of these different subtests and subse- practice. quently affect how intelligent we are deemed to In one study, elementary school children at a be. However, for a long time, it was believed that low SES school played with one of two sets of intelligence was something we inherited and board games and video games for 8 weeks could not do much to change (see, e.g., Herrnstein (Mackey et al. 2011 ). In the fi rst group, children and Murray 1994 ). played with games that engaged their reasoning The evidence for the primacy of innate ability ability, such as games that asked them to consider has not been well supported by accounts measur- and integrate multiple rules or complete patterns ing population changes in IQ performance since of shapes. A second group of children played the inception of the Weschsler Intelligence Scale, with games that involved processing speed, one of the main measures of intelligence. The which required them to make motor responses to well-documented Flynn effect (Flynn 2007 ) visual cues as fast as possible following simple describes how IQ scores on multiple well- game rules. At the beginning and again at the end established assessments of intelligence have been of the 8 weeks, tests in relevant areas pertaining on the rise—in some instances, dramatically— to either reasoning ability or processing speed from generation to generation, even on assessments were administered. The results were striking. that are deemed to be largely “culture-free.” Children who had played with the reasoning A compelling interpretation is that the perfor- games increased their nonverbal reasoning mance capacities measured by these tests func- by 32 %, which translates roughly to a 10-point tion as skills that can be improved and shaped by increase in IQ. On the other hand, those who had experience and schooling and that these experi- played with the processing speed-focused games ences have shifted over time in a way that has increased their processing speed by 27 %, but not changed how and what is learned by the majority their reasoning skills. These results demonstrate of the population (Flynn 2007 ; Nisbett 2009 ). not only that IQ scores can change signifi cantly In fact, over the past century, various studies in a short period of time but, more importantly, conducted all over the world have documented the that targeted interventions can improve perfor- role that schooling plays in cultivating students’ mance on the cognitive processes assessed in intelligence. If intelligence is a fi xed ability, intelligence measures, contradicting the belief environmental experiences, such as educational that our intelligence is fi xed. enrichment, should not alter it. Yet, countless Studies examining other cognitive functions examples confi rm the fi nding that, relative to chil- measured by intelligence tests have shown con- dren who remain in school, those who are denied verging evidence that speaks to the malleability educational experiences often display a gradual of these constructs. For example, Brehmer et al. but persistent decline in performance on intelli- ( 2012) asked a group of adults to use an adaptive gence measures—as much as 6 IQ point decre- compu terized program to train working memory ments for every year of schooling lost (see Ceci and compared them to a control group that used 1991 ; Nisbett 2009 for a review). Similarly, a nonadaptive, low diffi culty working memory related environmental factors such as socioeco- training. Before and after the training period, nomic status have been found to predict individual participants completed assessments of working 18 Intelligence as a Malleable Construct 269

memory, which were the same as those adminis- Neuroplasticity and Brain Function tered as part of the Weschsler Intelligence Scale. As part of the training, participants practiced How do learning-based gains in cognitive perfor- maintaining in mind over short periods of time mance relate to the modifi ability of neural struc- multiple bits of information (words or locations tures in the brain? Research in the neuroscience of objects in space). They also sequenced these of learning demonstrates signifi cant plasticity in items in a particular order according to the exer- brain structure and function and shows that these cise’s instructions. For those in the adaptive changes are highly dependent on both behavior training program, the quantity of information and environment. that needed to be held or manipulated in working The brain is composed of billions of special- memory changed depending on their perfor- ized cells known as neurons (see Fig. 18.3 ). mance, increasing (and therefore becoming more Neurons form part of the body’s communication diffi cult) as participants became more profi cient. hub, processing, sending, and receiving vital infor- After fi ve weeks of training, participants in this mation through an interconnected network. adaptive training group showed signifi cant Surrounding a neuron’s center are a series of improvements in their working memory perfor- extending branches known as dendrites that over- mance. These improvements were greater com- see and transport the collection of chemical mes- pared to the group that did not receive an adaptive sages received from other neurons. Collected training, even though both groups began their information is eventually converted into an electri- training with similar scores. Interestingly, in a cal impulse that travels down a long channel follow-up assessment, researchers found that the known as the neuron’s axon. This axon branches performance gains made by the adaptive training out into smaller axon channels that each culminate group were maintained three months after the in a small bulb that forms part of a structure called training period. These fi ndings are especially a synapse. It is at a synapse where the impulse trig- impressive given the brief nature of the gers the release of chemical substances, or neu- intervention. rotransmitters, from the neuron. These In sum, these examples support the idea that neurotransmitters then enter the receiving neuron the cognitive skills measured in intelligence tests through receptors of the receiving neuron. can be improved and maintained by targeted Although brain plasticity may not be specifi c training using increasing demands (challenge). to neural changes, much of the research has The signifi cance of the fl exibility of these cogni- focused on how the connectivity between neurons tive functions extends beyond improvements in a can change with experience. Studies on animals test score—they speak to our capacity to continue have informed us about the changes that occur at to adapt to new cognitive demands imposed by a cellular level that may modify the connection of our environment. neurons within and between brain regions. Based 270 L.S. Blackwell et al. on those fi ndings, Zatorre et al. (2012 ) discuss devoted to practicing problems that relied heavily some of the changes that may happen at the on reasoning, the fMRI scans showed changes in cellular level and underlie plasticity in the human the intrinsic connectivity of the student’s brains. brain. For example, axons can become insulated The measure of intrinsic connectivity, known as with myelin, a fatty tissue that makes up the resting-state fMRI, is thought to refl ect repeated “white matter” of the brain and speeds the trans- history of synchronized activity between regions, mission of electrical impulses, or their existing since the scan is captured at resting state, when a myelin covering can become thicker. Also, groups person is asked not to engage in any task. The of axons projecting between brain regions can group of individuals who underwent the reasoning become more organized, thus improving their training showed greater intrinsic connectivity connectivity. In humans, one or many of these between areas involved in reasoning skills com- transformations may refl ect changes in the integ- pared to their well-matched controls, and a specifi c rity of the microstructure of the brain’s white pattern of these connections was related to greater matter, which contains groups of myelinated improvements in their LSAT scores. axons whose cell bodies form the brain’s gray In addition, these researchers also examined matter. In addition to speeding up their transmis- the changes in white matter microstructure result- sion, neurons can improve their connectivity by ing from this training program (Mackey et al. increasing their surface area of connections 2012 ). They found an increase in the coherence through the creation of new neurons and synapses of white matter tracts connecting regions of the or branching of dendrites. These events may be brain that support reasoning skills, refl ecting refl ected in the structural and functional changes the integrity of the structure of white matter dis- of the gray matter after repeated practice of a cussed earlier. Although the specifi c mechanisms physical or mental skill. behind these changes are unknown, they are, An example of our brain’s ability to adapt to nonetheless, thought to refl ect strengthening of slight modifi cations to our daily activities can be the connectivity between brain regions brought seen in a seminal study where a group of indi- on by experience or development. viduals with no juggling experience were taught The brain’s malleability also makes it suscep- how to juggle over a period of three months. tible to negative factors, such as stress or unstable In comparison to their own brain scans taken at home environments (see, e.g., Erikson et al. 2003 ; the beginning of the study and also to those of a Hackman and Farah 2009 ; Lupien et al. 2009 ). To group of individuals who were not taught how to counteract this, Neville and colleagues (2013 ) juggle, the post-training brain scans of the jug- developed an 8-week intervention targeting selec- glers showed an increase in the gray matter thick- tive attention, the ability to control where our ness of brain regions that support the ability to focus is directed, in part aimed at increasing perceive motion and anticipate where objects will school readiness for preschoolers of low socio- be in space (Draganski et al. 2004 ). economic status. They reasoned that because a Changes in the brain are not limited to develop- stressful environment and more inconsistent ing a new visual-motor skill. Exciting fi ndings parenting practices are often more prevalent in from studies where participants practiced a cogni- low-SES compared to higher-SES households, tive process have revealed the brain’s ability to training preschoolers’ primary caregivers might adapt to different cognitive demands. Using func- also be benefi cial. Thus, they compared three tional magnetic resonance imaging (fMRI), which groups of children. One received their preschool measures changes in blood oxygenation and fl ow education as usual. A second group received in the brain associated with neural activity, Mackey attention exercises only. A third group of children et al. (2013 ) measured the neural effects of inten- also received attention exercises, and their parents sive reasoning training in young adults. After a received training in a curriculum targeted to three- month law school admission exam (LSAT) develop family stress regulation and other strate- preparation course, in which a little over 60 h were gies aimed at improving the way parents inter- 18 Intelligence as a Malleable Construct 271 acted with, disciplined, and facilitated their in overt behaviors, such as practicing and tack- children’s attention. To explore the impact of the ling more challenging tasks, were the only way training on cognitive functioning, researchers used that mindsets could impact learning—or whether electroencephalography (EEG)—a noninvasive the beliefs and goals that make up the different measure with excellent temporal sensitivity that mindsets might directly infl uence the way the can be used to capture changes in electrical poten- brain processes information. tials occurring within the brain that are elicited from the scalp. The signal embedded in this EEG, known as event-related potentials (ERPs), can The Neuroscience of Mindsets map out attentional and conceptual processes that emerge in response to specifi c task-related events, What is happening in the brain when people such as hearing a particular sound. In a test where are laboring under the different mindsets? children were asked to focus their attention on Researchers have begun to explore some of the only one of two stories played simultaneously, neural mechanisms underlying a growth and early attention ERP components related to probes fi xed mindset. embedded in the stories showed that children who A fascinating series of studies looking at brain received the family- based intervention were more activity in relation to mindset and different per- successful at focusing their attention on the story formance conditions showed that mindsets can they were instructed to attend. Additionally, this lead to different patterns of observed activity in group also improved in measures of nonverbal the brain, with consequences for cognitive func- IQ, while their parents showed lower levels of tioning. Moser et al. (2011 ), for example, tracked stress. how students allocated their attention while com- Together, these studies suggest that experience pleting a task that required continuous monitor- and learning can result in tangible, measurable ing and responding to a target displayed on a impacts on the brain and in turn on a variety of computer screen. How did students with different cognitive functions. This is especially promising mindsets react, especially after making a mis- because these brain changes are seen in response take? To explore this question, researchers looked to small changes in the experiences a person at specifi c ERPs that have been previously engages with, such as practicing a specifi c skill mapped to attention and awareness to errors. over a short period of time. Given these fi ndings, They found that individuals with a growth mind- we can anticipate that, as they continue to work set were more likely to attend to the errors they and study, virtually all students can continue to made than those with a fi xed mindset and were develop their abilities over time through positive also more likely to improve their accuracy on the behaviors like effort and practice, in a way that next trial. would ultimately be evidenced by changes to Interestingly, additional analyses revealed that brain structures and activity. their attentional response mediated their perfor- The evidence of performance improvement mance. In other words, it was because participants and neuroplasticity seen in these studies lends with a growth mindset oriented their attention to support to the concept of malleable intelligence errors that they did better on the task. These fi nd- that underlies a growth mindset. Further, the fact ings showed that people with a growth mindset that such changes are the result of behaviors were more successful at reorienting their attention such as deliberate practice and engaging with to the task at hand and suggest that they were not increasingly diffi cult tasks helps illuminate why discouraged by errors but responded in an adap- the increased effort, challenge seeking, and per- tive way that allowed them to persist and improve. sistence associated with a growth mindset would Individuals with a fi xed mindset may also result in higher achievement. Intelligence can be orient attention to errors when there is salient developed—but only if it is exercised. But our negative feedback, but may do so in a way that colleagues and we wondered whether engaging ultimately undermines learning (Mangels et al. 272 L.S. Blackwell et al.

2006). An ERP study by Jennifer Mangels and (Rodriguez et al. 2014 ). We recruited undergradu- her colleagues found that individuals with a fi xed ates to complete a challenging general knowledge mindset showed an enhanced awareness of and task drawn from Mangels et al. (2006 ). The task, orientation towards errors made on a challenging which contained two blocks of questions, general knowledge question task, in which indi- prompted students to complete all questions in a viduals received accuracy feedback (whether block before being presented with the second their response was correct or incorrect) followed block. Importantly, as students were presented by learning feedback (the correct answer) after with a block, they fi rst read task instructions each question. However, unlike the growth mind- that differentially framed it. In the performance set group in the previous study, this orienting did frame, students read instructions that oriented not aid their performance. Fixed mindset parti- their focus on their accuracy and how their cipants showed a neural response to learning performance would be compared to that of other feedback that was indicative of lower success at university students. However, in the mastery encoding the correct answer or storing and com- frame, these same students were instead asked to mitting the information to memory. In fact, in a focus on those questions that they found most surprise retest of all the items that they had previ- interesting and learned the best from rather than ously answered incorrectly, fi xed mindset partici- on their performance. How would students pants corrected fewer items than their growth respond to these two different but comparably mindset counterparts. challenging situations? These fi ndings illustrate how people’s mind- When task instructions emphasized perfor- sets may differentially impact their attentional mance and proving one’s ability relative to oth- response, particularly following challenge. On one ers, students completing the diffi cult task hand, individuals harboring fi xed mindset think- displayed a neural response to corrective infor- ing may inadvertently set themselves up for fail- mation (i.e., the correct answer) following an ure by directing their attention to their performance error that was consistent with processes related to and discounting an opportunity to learn from their superfi cial encoding of that information. Howe- mistakes, whereas this does not seem to be the ver, these very same students, when completing a case for those with a growth mindset. task that instead emphasized learning and mas- As previously discussed, students with a fi xed tery, experienced a neural response to the correct mindset often hold a performance focus in which answer (after an error) that was consistent with they are particularly concerned with the goal of processes that refl ect deeper encoding of that proving their abilities and achieving highly, espe- information. cially in comparison to others, whereas those This work suggests that although learning may with a growth mindset typically endorse goals occur in both mastery- and performance-based of learning and mastery (Blackwell et al. 2007 ; contexts, the nature of that learning may be very Dweck and Leggett 1988 ; Dweck 1999 ). What different. In a performance environment, students happens when students fi nd themselves in aca- may attend to problem solving only insofar as it demic contexts that promote either a performance allows them to get the right answer, understand- or mastery goal? As students navigate from ing it only at its surface, whereas in a learning one academic context to the other, it is very environment, students may not be solely focused possible that they may be receiving different on their outcomes and instead orient their efforts messages from their environment about what is to understanding the content in a manner that may valued in each domain, potentially impacting ultimately contribute to longer-term retention. how and what they learn. These fi ndings are especially intriguing since In support of this possibility, recent ERP these different neural processes, which are sug- research fi nds that students do indeed have very gestive of qualitatively different kinds of learning, different neural experiences when they encounter emerged within person after just brief exposures to a mastery- versus performance-based context each frame. Thus, they provide continued support 18 Intelligence as a Malleable Construct 273 for students’ sensitivity and differential response which they would learn. While the children to input from their learning environment. As we praised for process overwhelmingly chose the will see, messages conveyed to students through more diffi cult ones, the majority of children their interpersonal experiences with others can praised for intelligence chose to repeat the same also powerfully shape their beliefs and behaviors easy puzzles! Rather than giving children the in both the short and long term. confi dence to tackle a challenge, praise for intelligence had actually made them want to stay in their safe zone, even though it meant that they Mindsets and the Infl uence would learn nothing new. of Others Clearly, over the longer term, sacrifi cing such learning opportunities could have a negative As we have seen, the mindsets that individuals carry impact on skill development. But strikingly, with them affect their goals, cognitive functioning, the praise also had an immediate effect on the motivation-related behavior, and academic out- children’s intellectual performance. To test the comes. However, these mindsets themselves are not impact of praise on the children’s performance and fi xed. The messages that people get from others in resilience following challenge, the researchers their environment can infl uence their mindsets and next had the students work on more diffi cult puz- impact motivation and performance in immediate, zles, on which all the students struggled. They powerful, and often surprising ways. In fact, it turns then gave them another easier set, similar to the out that the very messages that one might think fi rst. How would they perform? The differences would be most encouraging—such as praise for were telling. The students praised for effort intelligence—can actually undermine performance improved signifi cantly on the easy puzzles over on intellectual tasks. their performance on the fi rst trial (perhaps hon- A pioneering series of studies by Claudia ing their skills on the more diffi cult problems). Mueller and Carol Dweck (1998 ) examined the But those who had been praised for intelligence impact of praise on fi fth graders’ challenge seek- performed worse on the second attempt—they ing and performance. Mueller and Dweck had the had lost confi dence that they were smart at puz- children complete a set of puzzles drawn from zles, and so they performed poorly. It is particu- Raven’s Progressive Matrices (a common mea- larly notable that the Raven’s task is one used to sure of nonverbal reasoning). Initially, they gave measure “fl uid” intelligence (often considered them problems matched to their grade level and to be an inherent problem-solving ability) and children solved most of them successfully. Then has often been used to assign children to gifted the researchers praised the students for their programs, yet it turned out that performance on performance. They told one group of randomly this test could be undermined (or enhanced) by chosen children, “Wow, that’s a really good a single sentence. Intelligence praise activated a score. You must be smart at this” (intelligence fi xed mindset framework, along with the goals of praise condition). A second group was told, looking smart and succeeding without effort, and “Wow, that’s a really good score. You must have produced a “helpless” response to challenge and worked hard at this” (effort/process praise condi- an immediate decrease in apparent ability. tion). (Process praise can refer to anything about More recent studies have replicated and extended the process the child engaged in: their strategy, these fi ndings. For example, one study explored focus, effort, choices, or perseverance.) Then how feedback linking success on an upcoming they looked at how these different kinds of challenging activity to a group’s supposed inherent praise would affect the students’ behavior and ability impacted kids’ performance on that task performance. First, they asked them which type (Cimpian et al. 2012 ). The researchers found that of puzzle they would prefer to do next: an easy when young (4–7-year- old) children were told that one, like the ones they had done, on which they either girls or boys were really good at a game, would perform well, or more diffi cult ones, from the children, regardless of their gender, underper- 274 L.S. Blackwell et al. formed on the game, especially on more diffi cult study looking at the impact of teacher mindsets items, relative to children who were provided with on student achievement found that when teachers other kinds of instructions that did not suggest the had a fi xed mindset, students who had entered inherent ability of groups. Cimpian and his col- their class as low achievers remained so. In con- leagues argue that this occurred because children trast, when teachers had a growth mindset, many came to attribute their performance on the task as of the students who had started the year as being innately linked to something out of their con- low achievers showed remarkable progress trol—a “fi xed” aspect of their identity, which in (Rheinberg et al. 2000 ). Thus, for the fi xed mind- turn led to their underperformance. set teachers, their experience confi rmed their Since the Mueller and Dweck ( 1998 ) studies, beliefs, as their students’ relative status remained other researchers have investigated the impact of unchanged, whereas the growth mindset teachers praise in real-world contexts on mindsets and saw their confi dence in students’ ability to grow performance over longer periods of time. They realized. Once again, the mindsets that people have found that parents can play a signifi cant role hold can become self-reinforcing. in formulating their children’s beliefs about their Crucially, a growth mindset is not the same as ability, sometimes with long-term effects. One self-confi dence or drive to achieve. In fact, as we longitudinal study found that children whose saw in the praise studies, successful students mothers gave them more process praise at 14–38 who derive much of their self-esteem from per- months were more likely to display a growth forming well can be vulnerable when they mindset and a greater desire for challenge at 7–8 encounter challenge or diffi culty—particularly if years old (Gunderson et al. 2013 ). In a similar they are in an evaluative context. Here, a young study with older children, 8–12-year-olds whose student about to begin middle school seems to mothers praised them for ability were more likely have a robust sense of self-confi dence as he to exhibit fi xed mindset thinking and a reduced explains his relish for a challenge—but his pri- desire for challenge six months later (Pomerantz mary goal is to demonstrate his ability, rather and Kemper 2013 ). than to learn: As the praise studies suggest, teachers too are Right now my favorite subject is math because I’m in a position to infl uence their students’ mindsets. really good at it – my grades in my report card have For example, just as praise for intelligence can been really high in math so that’s why I like it … backfi re, the way that we console children when A lot of times kids don’t really want to work hard. But if you really want to know something new and they struggle may inadvertently trigger the fi xed really get good at it, you’re going to have to work mindset pattern. Aneeta Rattan and her col- hard. And some kids like working hard. So it’s like leagues (Rattan et al. 2012 ) asked adults (some of a challenge, like a puzzle … I like doing stuff that whom were math teachers) to imagine a student I’m good at and I know I’ll get a good grade. And then I like thinking because I like – I’m the type of in their class who had gotten a poor grade on the guy that likes puzzles and I like challenges ’cause fi rst math test of the year and to report how they it makes you feel like you’re up to it and you’re would respond to the student. The adults in a showing the person what you can do and it makes fi xed mindset were signifi cantly more likely to you feel good. try to console the student by saying that not This student enjoys a challenge, as long as everyone could be good in math—a message that he can be successful. But in a situation where he students reported would lead them to conclude may make mistakes in class and lose his status as that they have low ability and to feel like giving a top performer in the eyes of others, his motiva- up. However, adults in a growth mindset were tion takes a nosedive: more likely to urge students to try harder, and If you’re doing it [making mistakes] in front of they gave them practical recommendations for people? I wouldn't really want to do it because strategies to achieve mastery. when you make a mistake you kind of tend to get In these ways, the mindsets that adults hold embarrassed and people will say, “No, you’re doing can be a factor in students’ success. Indeed, a it wrong. What are you doing?” and that tends to be 18 Intelligence as a Malleable Construct 275

embarrassing … Like if you’re doing a group and of the year. Fully three-fourths of the students there’s tons of kids around you, when you ask for their teachers identifi ed were from the growth help then they know you’re not understanding it. And then some people say that’s really easy and mindset workshop group—a signifi cant difference. they’re like, “How are you not understanding it?” Here is a typical comment a teacher made about It makes you feel stupid, and then you get the observed changes: embarrassed. Your workshop has already had an effect. L., who Thus, as we saw in both the study of students never puts in any extra effort and often doesn’t turn making the transition to junior high school and in in homework on time, actually stayed up late working for hours to fi nish an assignment early so the praise studies, a fi xed mindset framework I could review it and give him a chance to revise it. may not hurt performance in conditions of He earned a B+ on the assignment (he had been relatively low challenge, where students’ skills getting C’s and lower). exceed the demands of the task and success is We examined the students’ performance in readily obtainable. But when the possibility of math over the course of the study. Prior to the failure looms in a situation where ability may be intervention, the grades of students in both groups evaluated, mindset makes all the difference in had very similar trajectories: they were declining whether the student will be resilient and bounce from those obtained in sixth grade (the previous back from diffi culty or become helpless and year) in the same school as the challenge level in founder (Blackwell et al. 2007 ; Mueller and the curriculum increased. And indeed, in the term Dweck 1998 ). following the intervention, the grades of the students in the control group continued to decline, but the grades of students in the growth mindset Changing Mindsets workshop reversed course, erasing the downturn (see Fig. 18.4 ) (Blackwell et al. 2007 ). Thus, the Given that mindsets are infl uenced by messages students who received instruction in the mallea- from others, we wondered whether it would be ble brain and developing ability not only became possible to teach a growth mindset and improve more motivated, their math performance reboun- students’ motivation and performance as a result. ded even as the curriculum continued to become To do this, we developed an eight-session more diffi cult over the course of seventh grade. workshop to teach students a growth mindset by We have since developed and tested a blended- teaching them about the brain and how it develops learning curriculum, Brainology ®, based on this and grows stronger through learning. The work- workshop, Brainology, and found that it pro- shop included an article, “You Can Grow Your moted a similar shift in mindset, motivation, and Intelligence,” images and explanations of how the performance. Here is how middle school students brain works, and discussions about learning and who completed this curriculum described the growth, along with lessons on study skills. We impact of learning about the malleable brain on randomly assigned seventh grade students in an their view of intelligence, effort, and challenge: urban middle school to either this growth mindset You’re not born dumb or born smart … Once you workshop or an alternative version that taught stu- know how your brain works, it’s much easier to dents about the brain and study skills, but without control it—once you develop more neurons and the information and activities focused on the connections, it’s much easier to approach something that’s harder for you. malleable brain and developing intelligence. Both Probability , I was just like, “I don’t get this at workshops were taught in the students’ advisory all. But I was just like, okay, I’m going to do this sections by separate teams of researchers. … I want to do this since it’s so hard. I’m going to After the workshops, we asked the students’ be like, Brain, you cannot just run away from this. I’m going to do this! teachers, who were blind to the workshop condition of the students, to identify and describe those Other studies, with participants from middle who had improved in motivation over the course school to college, have shown similar impacts 276 L.S. Blackwell et al.

3.00

2.90

2.80 Experimental 2.70 (Growth Mindset)

2.60

2.50

Math Grades 2.40 Control 2.30

2.20

2.10

2.00 Spring 6th Grade Fall 7th Grade Spring 7th Grade (Pre-Intervention) (Post-Intervention) End of Term Grades

Fig. 18.4 Student achievement as a function of workshop grade term), students in the intervention group (who group in Study 2 (Notes: All students’ math grades had learned about the malleable brain) rebounded, while the declined from the end of sixth grade to the end of the fall grades of students in the control group continued to term in seventh grade, and there was no statistically sig- decline (p < .05). Reprinted from Blackwell et al. 2007 , nifi cant difference between the two groups at either time p. 257. Reprinted with permission. Copyright 2007 from point. Following the workshop (end of Spring seventh Society for Research in Child Development, Inc.) of teaching a growth mindset on motivation and Latino students in academics or female students performance (e.g., Aronson et al. 2002 ; Good in math and science) encounter a situation that et al. 2003; Yeager et al. 2013). A series of new puts them at risk of being judged in light of that studies by David Yeager, David Paunesku, and stereotype and, potentially, of confi rming it in their colleagues, for example, have found that the eyes of others. This concern with confi rming even brief, one-time mindset interventions deliv- a negative stereotype can interfere with thinking ered online can lead to signifi cant gains in and motivation and, therefore, performance. For achievement, including among students from tra- example, a female student taking a test of math ditionally underrepresented groups (Yeager et al. ability given by a male administrator may worry 2013 ). Thus, mindsets can be changed and moti- that if she doesn’t perform well, it will be seen as vation and achievement improved as a result. confi rmation that females are not good at math, or a students of color taking the SAT may anticipate that their performance will be seen as Mindsets and the Achievement Gap refl ecting on the intelligence of their race if they perform poorly. A particularly striking way that the social con- The anticipation of such group-based negative text can impact mindset and performance is evaluation can lead to a host of detrimental con- embodied in the phenomenon of stereotype sequences, including negative thoughts (Cadinu threat. Originally identifi ed by Claude Steele et al. 2005 ; Keller and Dauenheimer 2003 ), anxi- and Joshua Aronson (Steele and Aronson 1995 ), ety (Marx and Stapel 2006 ), and physiological stereotype threat occurs when students from a arousal that can reduce cognitive functioning negatively stereotyped group (e.g., Black and (Blascovich et al. 2001 ; Krendl et al. 2008 ; 18 Intelligence as a Malleable Construct 277

Osborne 2006 , 2007 ; Vick et al. 2008 ), including intelligence was successful in increasing enjoy- working memory (Beilock et al. 2007 ; Schmader ment and valuing of academics and academic and Johns 2003 ) and attentional and behavioral performance, including GPA, among Black control (Inzlicht et al. 2006 ; Smith and White college students (Aronson et al. 2002 ). Similarly, 2002 ), especially on a challenging task (O’Brien in a study with middle school minority students, and Crandall 2003 ; Stone and McWhinnie 2008 ), teaching about malleable intelligence raised all of which can result in reduced performance achievement test scores in both reading and math and widening achievement gaps between groups and narrowed the achievement gap between male (Beilock et al. 2007; Brown and Day 2006 ; and female students (Good et al. 2003 ). Finally, Cadinu et al. 2005; Keller and Dauenheimer 2003 ; explaining gender differences in math perfor- O’Brien and Crandall 2003; Schmader and Johns mance as the result of genetic factors rather 2003 ; Steele and Aronson 1995 ). Thus, the preoc- than experience (Dar-Nimrod and Heine 2006 ) or cupation with ability and performance induced by effort (Thoman et al. 2008 ) reduced the perfor- stereotype threat can actually reduce both (at least mance of females on a math test consistent with temporarily) for students in contexts where the the notion that malleable, experiential- based stereotype is relevant. However, when the task is explanations of ability buffer students from the defi ned as non- diagnostic of ability, the perfor- negative effects of stereotype threat. mance gap is narrowed (Aronson et al. 1999 ; These sets of studies highlight the powerful Steele and Aronson 1995 ). role that the psychological experience and The negative effects of stereotype threat context plays in shaping students’ motivation, emerge for children as young as seven. In one set learning, and performance. In sum, the evidence of studies (Hartley and Sutton 2013 ), boys under- strongly suggests that cognitive performance is performed on math, reading, and writing assess- the product of a synergistic relationship between ments when they held the belief that boys were individual aptitudes, beliefs, and preferences and “inferior at academics” and also when they infl uences from the environment, and that mind- were explicitly provided this message prior to sets about intelligence are a critical part of this working on the assessments. Interestingly, these relationship. effects disappeared when boys were instead told that there were no real differences in abilities between boys and girls. Implications for Future Research Aronson and colleagues (2002 ) noted that the impact of stereotype threat on students—preoc- Many questions still remain to be answered cupation with evaluation, anxiety during assess- regarding the interplay of mindsets and intel- ments, lower performance, and disidentifi cation lectual achievement. From a developmental per- with academics—looked similar to the pattern spective, we know too little about how early observed in people with a fi xed mindset concept experience forges mindsets and the impact that it of intelligence under conditions of challenge. has on the development of talent and skills over a Perhaps, they reasoned, teaching students to child’s early years. In particular, research show- think of their abilities as malleable could buffer ing the emergence of the characteristic mindset them against the negative effects of stereotype patterns in young children, even before they threat: knowing that they could always develop develop a differentiated concept of intelligence, their ability, they would be less worried about suggests that parental infl uence needs more whether they scored well on a particular test investigation (Giles and Heyman 2003 ; Heyman (Aronson et al. 2002 ). To test this, they taught et al. 2003; Smiley and Dweck 1994 ). The more students that their intelligence was malleable and general nature of these early mindsets raises a then looked at their performance under condi- question of whether a more global growth or tions of stereotype threat. Teaching malleable fi xed mindset lies behind the much-investigated 278 L.S. Blackwell et al. concept of ability: perhaps some children come one’s abilities and talents? Walter Michel’s impact- to believe early on that people are generally fi xed ful work has demonstrated how the tendency to in their attributes, and this paradigm is later delay immediate gratifi cation during the preschool populated by more specifi c concepts such as years is correlated with positive events across intelligence, character, and the like. the lifespan (cf. Mischel et al. 2011 ). How does How susceptible are mindsets to enduring holding a growth mindset impact people over a change, and what is the minimum required inter- lifetime? Research in the workplace, for example, vention to drive meaningful change? Can lasting fi nds that growth mindset in leadership roles change be achieved through single point, targeted (“leaders are made”) can contribute to greater con- interventions teaching a growth mindset to students, fi dence and positive affect (Hoyt et al. 2012 ). or does it require a combination of malleable intel- Future research should continue to explore this ligence instruction with ongoing reinforcement possibility. through implicit messages, such as process praise? How much should we make these frameworks explicit and examined in order to transform them? Implications for Educational Many studies indicate short-term effects of even Practice brief growth mindset interventions (see, e.g., Yeager et al. 2013 ), but we do not yet know what it Closing the Achievement Gap takes to achieve a permanent shift. How complex and context dependent are While the debate over the nature of intelligence mindsets? Note that, in much of the research lit- continues, our educational system, from K-12 to erature, mindset has been treated as a global college, is grappling with the application of these and categorical variable, contrasting fi xed versus concepts and measures to policy and practice, growth mindsets about general intelligence. with often unintended consequences. In many However, it is possible to have different mindsets schools, students are still ranked and tracked by about different kinds of ability. For example, as ability based on prior achievement or their scores shown in the studies using mindset interventions on assessments, exposing them to differing cur- to reduce stereotype threat, some groups are vul- ricula and standards. Standing on achievement nerable in specifi c subject areas, despite superior tests can be misapplied in practice as ability academic performance overall (e.g., females in labels that lead both students and teachers to math and science; Good et al. 2012 ). In addition, adopt a fi xed mindset and lower expectations, as we have seen, messages received from the which can then become a self-fulfi lling prophecy. learning environment, including our interactions This emphasis on normative assessments and with others, can impact mindset-related behavior grading practices can make achievement appear signifi cantly (Blackwell et al. 2007 ; Mueller and to be a zero-sum game, with enormous implica- Dweck 1998 ) and can shape the neural underpin- tions due to competitive access to schools and nings of the learning experience (e.g., Rodriguez higher education opportunities. The persistent et al. 2014 ). These fi ndings suggest that many achievement gap due to unequal educational people may harbor mixed or fl exible mindsets opportunities and the psychological burden and rely on environmental cues to activate the imposed by societal stereotypes for African one deemed most appropriate to the situation. American and Latino students and for females Further research to learn more about the contex- in math, science, and engineering still signal that tual factors that can activate different mindsets a large number of our young people are laboring would be helpful in designing interventions under identity-based fi xed mindset conceptions and educational programs to support struggling of their ability that limit them in fulfi lling their learners. potential. Finally, what are the long-term consequences Indeed, even among individuals with both of holding a growth mindset for the development of high math and verbal ability, women are less 18 Intelligence as a Malleable Construct 279 likely to pursue careers in science, technology, ally ensure failure and undermine the focus on engineering, or math (e.g., Wang et al. 2013 ). process and growth that are critical to promoting These fi ndings are especially striking because positive motivation. women in the Wang et al. (2003) study repre- Recent research suggests that motivation-related sented a greater percentage of those individuals behaviors, such as “grit” and perseverance in pur- with high scores on both math and verbal assess- suing goals, may be more predictive of success ments. Recruiting students into these fi elds may than IQ; for example, Angela Duckworth and col- pose challenges for additional reasons, one of leagues have found that students who exhibited these arising from the way in which these abili- greater self-control earned higher grades, whereas ties are portrayed in popular media. For example, students’ IQ scores were not related to achieve- when students were presented with a (fi ctitious) ment (Duckworth and Seligman 2005 ). We know newspaper articles conveying the biological from a large body of research that teaching con- “nature” of gender differences, readers were tent in the absence of positive academic mindsets more likely to agree with gender stereotypes, and basic learning skills often falls short and that whereas the opposite was true when social expla- programs aimed at changing students’ learning nations were used to describe differences behavior directly are less effective than interven- (Brescoll and LaFrance 2004). Explanations for tions that change their mindsets. (See Farrington group differences that rest upon biology implic- et al. 2012 for a review.) Yet the vast majority of itly convey a fi xed mindset conception of ability our educational efforts are devoted to core subject and reinforce the stereotypes that can undermine curriculum and assessment, sometimes at the motivation and achievement. Thus, both academic expense of teaching academic mindsets and foun- practices and their refl ection in the popular cul- dational learning skills. ture can inadvertently constrain the performance The evidence shows that intellectual develop- of vulnerable students. ment and performance are highly dependent on the interaction of environmental supports and individual effort and active engagement—and that Educational Systems and Structures practices that instill and nurture a growth mindset also promote and sustain effort, engagement, The research shows that messages that highlight and achievement. What could we achieve if, rather a person’s ability, rather than their effort and pro- than measuring and comparing students with one cess, reinforce a fi xed mindset and often precipi- another, we focused on providing them with a tate a helpless pattern when the person encounters solid foundation of self-effi cacy and skills and diffi culty. However, while individual teachers then on creating opportunities for them to grow? can change the way they praise and criticize stu- With the knowledge we have gained from decades dents in their classroom, they also operate within of research in the role of mindsets in achievement, a larger context of assessment and incentives that we have the opportunity to provide the current are not informed by this research. For example, generation of learners—and their teachers—with most schools and districts still adhere to one- a solid foundation for future growth. size- fi ts all curricula and standards with age and grade-level expectations, grade students in comparison to their peers on a rigid timeline, and rely References on a small number of high-stakes tests to measure Aronson, J., Lustina, M., Good, C., Keough, K., Steele, student and school success and to select students C., & Brown, J. (1999). When white men can’t do for access to future learning opportunities. For math: Necessary and suffi cient factors in stereotype students who initially lack foundational skills threat. Journal of Experimental Social Psychology, 35 , and learning strategies, these policies may virtu- 29–46. 280 L.S. Blackwell et al.

Aronson, J., Fried, C. B., & Good, C. (2002). Reducing Canivez, G., & Watkins, M. (1998). Long-term stability of the effects of stereotype threat on African American the Wechsler intelligence scale for children—third college students by shaping mindsets of intelligence. edition. Psychological Assessment, 10 , 285–291. Journal of Experimental Social Psychology, 38 , Ceci, S. J. (1991). How much does schooling infl uence 113–125. general intelligence and its cognitive components? Bartels, M., Rietveld, M., Van Baal, G., & Boomsma, D. A reassessment of the evidence. Developmental (2002). Genetic and environmental infl uences on the Psychology, 27 (5), 703–722. development of intelligence. Behavior Genetics, 32 , Cimpian, A., Mu, Y., & Erickson, L. C. (2012). Who is 237–249. good at this game? Linking an activity to a social Beilock, S. L., Rydell, R. J., & McConnell, A. R. (2007). category undermines children’s achievement. Psycho- Stereotype threat and working memory: Mechanisms, logical Science, 23 (5), 533–541. alleviations, and spillover. Journal of Experimental Darling-Hammond, L. (1994). Performance-based assess- Psychology: General, 136 , 256–276. ment and educational equity. Harvard Educational Binet, A. (1975). Modern ideas about children. (trans: Review, 64 , 5–31. Heisler, S.). Menlo Park: Suzanne Heisler. (originally Darling-Hammond, L. (1995). Cracks in the bell curve: work published in 1909). How education matters. The Journal of Negro Education, Blackwell, L., Trzesniewski, K., & Dweck, C. S. (2007). 64, 340–353. Implicit theories of intelligence predict achievement Dar-Nimrod, I., & Heine, S. J. (2006). Exposure to scien- across an adolescent transition: A longitudinal study tifi c theories affects women’s math performance. and an intervention. Child Development, 78 , 246–263. Science, 314 , 435. Blascovich, J., Spencer, S. J., Quinn, D. M., & Steele, Draganski, B., Gaser, C., Busch, V., Schuierer, G., C. M. (2001). African-Americans and high blood pres- Bogdahn, U., & May, A. (2004). Neuroplasticity: sure: The Role of stereotype threat. Psychological Changes in grey matter induced by training. Nature, Science, 12 , 225–229. 427 , 311–312. Borland, J. H. (2003). The death of giftedness. In J. H. Duckworth, A., & Seligman, M. (2005). Self-discipline Borland (Ed.), Rethinking gifted education (pp. 105– outdoes IQ in predicting academic performance of 124). New York: Teachers College Press. adolescents. Psychological Science, 16 , 939–944. Borland, J. H. (2005). Gifted education without gifted Dweck, C. S. (1999). Self-theories: Their role in moti- children: The case for no conception of giftedness. In vation, personality, and development . Philadelphia: R. J. Sternberg & J. E. Davidson (Eds.), Conceptions Psychology Press. of giftedness (2nd ed.). New York: Cambridge Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive University Press. approach to motivation and personality. Psychological Borland, J. H., & Wright, L. (2001). Identifying and Review, 95 , 256–273. educating poor and underrepresented gifted students. Erikson, K., Drevets, W., & Schulkin, J. (2003). Gluco- In K. A. Heller, F. J. Monks, R. J. Sternberg, & R. F. corticoid regulation of diverse cognitive functions in Subotnik (Eds.), International handbook of research normal and pathological emotional states. Neuroscience and development of giftedness and talent (pp. 587– and Biobehavioral Reviews, 27, 233–246. 594). London: Pergamon Press. Farrington, C. A., Roderick, M., Allensworth, E., Brehmer, Y., Westerberg, H., & Backman, L. (2012). Nagaoka, J., Keyes, T. S., Johnson, D. W., & Beechum, Working-memory training in younger and older adults: N. O. (2012). Teaching adolescents to become learn- Training gains, transfer, and maintenance. Frontiers in ers. The role of noncognitive factors in shaping school Human Neuroscience, 6, 1–7. performance: A critical literature review . Chicago: Brescoll, V. L., & LaFrance, M. (2004). The correlates University of Chicago Consortium on Chicago School and consequences of newspaper reports of research Research. on gender differences. Psychological Science, 15 (8), Flynn, J. (2007). What is intelligence? New York: 515–521. Cambridge University Press. Breslau, N., Chilcoat, H., Susser, E., Matte, T., Liang, K., Giles, J. W., & Heyman, G. D. (2003). Preschoolers’ & Peterson, E. (2001). Stability and change in chil- beliefs about the stability of antisocial behavior: dren’s intelligence quotient scores: A comparison Implications for navigating social challenges. Social of two socioeconomically disparate communities. Development, 12 , 182–197. American Journal of Epidemiology, 154 , 711–717. Good, C., Aronson, J., & Inzlicht, M. (2003). Improving doi: 10.1093/aje/154.8.711 . adolescents’ standardized test performance: An inter- Brown, R. P., & Day, E. A. (2006). The difference isn’t vention to reduce the effects of stereotype threat. black and white: Stereotype threat and the race gap on Applied Developmental Psychology, 24 , 645–662. Raven’s advanced progressive matrices. Journal of Good, C., Rattan, A., & Dweck, C. (2012). Why do Applied Psychology, 91 , 979–985. women opt out? Sense of belonging and women’s Cadinu, M., Maass, A., Rosabianca, A., & Kiesner, representation in mathematics. Journal of Personality J. (2005). Why do women underperform under stereo- and Social Psychology, 102(4), 700–717. doi: 10.1037/ type threat? Psychological Science, 16 , 572–578. a0026659 . 18 Intelligence as a Malleable Construct 281

Gunderson, E., Gripshoever, S., Romero, C., Dweck, C., in children. Developmental Science, 14 (3), 582–590. Goldin-Meadow, S., & Levine, S. (2013). Parent praise doi: 10.1111/j.1467-7687.2010.01005.x . to 1- to 3- year-olds predicts children’s motivational Mackey, A., Whitaker, K., & Bunge, S. (2012). Experience- frameworks 5 years later. Child Development, 84(5), dependent plasticity in white matter microstructure: 1526–1541. doi: 10.1111/cdev.12064 . reasoning training alters structural connectivity. Hackman, D., & Farah, M. (2009). Socioeconomic status Frontiers in Neuroanatomy, 6 , 1–9. doi: 10.3389/ and the developing brain. Trends in Cognitive Science, fnana.2012.00032 . 13 (2), 65–73. Mackey, A., Miller-Singley, A., & Bunge, S. (2013). Hartley, B., & Sutton, R. (2013). A stereotype threat Intensive reasoning training alters patterns of brain account of boys’ academic underachievement. Child connectivity at rest. Journal of Neuroscience, 33 (11), Development, 84 , 1716–1733. doi: 10.1111/cdev.12079 . 4796–4803. Henderson, V. L., & Dweck, C. S. (1990). Motivation and Mangels, J. A., Butterfi eld, B., Lamb, J., Good, C. D., & achievement. In S. S. Feldman & G. R. Elliot (Eds.), Dweck, C. S. (2006). Why do beliefs about intelli- At the threshold: The developing adolescent (pp. 308– gence infl uence learning success? A social-cognitive- 329). Cambridge: Harvard University Press. neuroscience model. Social Cognitive and Affective Herrnstein, R. J., & Murray, C. (1994). The bell curve: Neuroscience, 1 , 75–86. Intelligence and class structure in American life . Marx, D. M., & Stapel, D. A. (2006). Distinguishing ste- New York: Free Press. reotype threat from priming effects: On the role of the Hertzog, C., & Schaie, K. (1986). Stability and change in social self and threat-based concerns. Journal of adult intelligence: 1. Analysis of longitudinal covari- Personality and Social Psychology, 91 , 243–254. ance structures. Psychology and Aging, 1 , 159–171. Mischel, W., Ayduk, O., Berman, M., Casey, B., Gotlib, J., Heyman, G., Gee, C., & Giles, J. (2003). Preschool chil- Kross, E., Teslovich, T., Wilson, N., Zayas, V., & dren’s reasoning about ability. Child Development, 74 , Shoda, Y. (2011). Willpower over the lifespan: 516–534. Decomposing self-regulation. Social Cognitive and Hong, Y., Chiu, C., Dweck, C., Lin, D., & Wan, W. (1999). Affective Neuroscience, 6 (2), 252–256. Implicit theories, attributions, and coping: A meaning Moser, J., Schroder, H., Heeter, C., Moran, T., & Lee, Y. system approach. Journal of Personality and Social (2011). Mind your errors: Evidence for a neural Psychology, 77 , 588–599. mechanism linking growth mindset to adaptive post- Hoyt, C., Burnette, J., & Inella, A. (2012). I can do that: error adjustments. Psychological Science, 22(12), The impact of implicit theories on leadership 1484–1489. role model effectiveness. Personality and Social Mueller, C. M., & Dweck, C. S. (1998). Intelligence praise Psychology Bulletin, 38, 257–268. doi: 10.1177/ can undermine motivation and performance. Journal of 0146167211427922 . Personality and Social Psychology, 75 , 33–52. Inzlicht, M., McKay, L., & Aronson, J. (2006). Stigma as Naglieri, J., & Goldstein, S. (2009). Practicioner’s guide ego depletion: How being the target of prejudice affects to assessing intelligence and achievement . Hoboken: self-control. Psychological Science, 17, 262–269. Wiley. Kaufman, S. (2013). Ungifted: Intelligence redefi ned . Neville, H., Stevens, C., Pakulak, E., Bell, T., Fanning, J., New York: Basic Books. Klein, S., & Isbell, E. (2013). Family-based training Keller, J., & Dauenheimer, D. (2003). Stereotype threat in program improves brain function, cognition, and the classroom: Dejection mediates the disrupting threat behavior in lower socioeconomic status preschoolers. effect on women’s math performance. Personality and Proceedings of the National Academy of Sciences, Social Psychology Bulletin, 29 , 371–381. 110 , 12138–12143. doi: 10.1073/pnas.1304437110 . Kray, L. J., & Haselhuhn, M. P. (2007). Implicit negotia- Nisbett, R. (2009). Intelligence and how to get it: Why tion beliefs and performance: Experimental and longi- schools and cultures count. New York: W.W. Norton & tudinal evidence. Journal of Personality and Social Company. Psychology, 93 , 49–64. O’Brien, L. T., & Crandall, C. S. (2003). Stereotype threat Krendl, A. C., Richeson, J. A., Kelley, W. M., & and arousal: Effects on women’s math performance. Heatherton, T. F. (2008). The negative consequences Personality and Social Psychology Bulletin, 29 , of threat: A functional magnetic resonance imaging 782–789. investigation of the neural mechanisms underlying Osborne, J. W. (2006). Gender, stereotype threat and anxiety: women’s underperformance in math. Psychological Psychophysiological and cognitive evidence. Journal Science, 19 , 168–175. of Research in Educational Psychology, 8 , 109–138. Lupien, S., McEwen, B., Gunnar, M., & Heim, C. (2009). Osborne, J. W. (2007). Linking stereotype threat and anxi- Effects of stress throughout the lifespan on the brain, ety. Educational Psychology, 27 , 135–154. behaviour and cognition. Nature Reviews Neuro- Pomerantz, E. M., & Kempner, S. G. (2013). Mothers’ science, 10 , 434–445. daily person and process praise: Implications for chil- Mackey, A. P., Hill, S., Stone, S., & Bunge, S. A. (2011). dren’s theory of intelligence and motivation. Differential effects of reasoning and speed training Developmental Psychology. 282 L.S. Blackwell et al.

Rattan, A., Good, C., & Dweck, C. S. (2012). “It’s ok – not Journal of Personality and Social Psychology, 69 , everyone can be good at math:” Instructors with an 797–811. entity theory comfort (and demotivate) students. Journal Stone, J., & McWhinnie, C. (2008). Evidence that blatant of Experimental Social Psychology, 48, 731–737. versus subtle stereotype threat cues impact perfor- Rheinberg, F., Vollmeyer, R., & Rollett, W. (2000). mance through dual processes. Journal of Experimental Motivation and action in self-regulated learning. In M. Social Psychology, 44 , 445–452. Boekaerts, P. R. Pintrich, & M. Zeidner (Eds.), Thoman, D. B., White, P. H., Yamawaki, N., & Koishi, H. Handbook of self-regulation: Theory, research and (2008). Variations of gender-math stereotype content application (pp. 503–529). San Diego: Academic. affect women’s vulnerability to stereotype threat. Sex Robins, R. W., & Pals, J. L. (2002). Implicit self-theories Roles, 58 , 702–712. in the academic domain: Implications for goal orienta- Vick, S. B., Seery, M. D., Blascovich, J., & Weisbuch, M. tion, attributions, affect, and self-esteem change. Self (2008). The effect of gender stereotype activation on and Identity, 1 , 313–336. challenge and threat motivational states. Journal of Rodriguez, S., Mangels, J., Guerra-Carrillo, B., & Higgins, Experimental Social Psychology, 44 , 624–630. T. (2014). Frame of mind: Focusing students on perfor- Wang, M., Eccles, J., & Kenny, S. (2013). Not lack of ability mance or mastery yields a double dissociation of the but more choice: Individual and gender differences in neural processes predicting subsequent memory . choice of careers in science, technology, engineering, and Manuscript in preparation. mathematics. Psychological Science, 24 (5), 770–775. Schmader, T., & Johns, M. (2003). Converging evidence Wolf, T. (1973). Alfred Binet. Chicago: University of that stereotype threat reduces working memory capac- Chicago Press. ity. Journal of Personality and Social Psychology, 85 , Yeager, D., Paunesku, D., Walton, G., & Dweck, C. 440–452. (2013). How can we instill productive mindsets at Smiley, P. A., & Dweck, C. S. (1994). Individual differ- scale? A review of the evidence and an initial R&D ences in achievement goals among young children. agenda. [white paper]. Retrieved from: http://homepage. Child Development, 65 , 1723–1743. psy.utexas.edu/homepage/group/YeagerLAB/ADRG/ Smith, J. L., & White, P. H. (2002). An examination of Pdfs/Yeager%20et%20al%20R&D%20agenda% implicitly activated, explicitly activated, and nulliﬁ ed 20-%206-10-13.pdf stereotypes on mathematical performance: It’s not just Zatorre, R., Fields, R., & Johansen-Berg, H. (2012). a woman’s issue. Sex Roles, 47 , 179–191. Plasticity in gray and white: Neuroimaging changes in Steele, C., & Aronson, J. (1995). Stereotype threat and the brain structure during learning. Nature Neuroscience, intellectual test performance of African americans. 15 (4), 528–536. doi: 10.1038/nn.3045 . Creativity and Intelligence 19 Jonathan A. Plucker , Amber Esping , James C. Kaufman , and Maria J. Avitia

and talent development. Several seminal studies Creativity and Intelligence of creativity focus at least in part on intelligence (e.g., Barron 1963; Getzels and Jackson 1962 ; Given the importance of the construct of human Wallach and Kogan 1965 ); indeed, one of the intelligence, it is not surprising that the relation- ﬁ rst leading creativity researchers, J. P. Guilford, ship between intelligence and other constructs began by studying intelligence (Guilford 1967 ). has been frequently studied. For example, psy- This attention is due, in part, to the fact that both chologists maintain active research programs on constructs are integral to understanding talent the relationship between intelligence and work- and giftedness, and scholarship on gifted educa- ing memory (Burgess et al. 2011 ), intelligence tion has traditionally included studies in this area. and bias in logical thinking (Stanovich, West, and As Sternberg and O’Hara ( 1999 ) observed, the Toplak 2013 ), and – increasingly – intelligence degree to which creativity and intelligence are and health (Deary et al. 2010 ). related – and the nature of any such relationship – The potential association of creativity and is “theoretically important, and its answer prob- intelligence has consistently received attention, ably affects the lives of countless children and with the early scientiﬁ c studies of intelligence adults” (p. 269), and Plucker and Renzulli (1999 ) (e.g., Galton’s Hereditary Genius, 1869 ) overlap- concluded that it is now not a matter of discover- ping considerably with the constructs of creativity ing whether intelligence and creativity are related, but rather of how they are related. Although those sentiments are nearly a genera- J. A. Plucker (*) • J. C. Kaufman tion old at this point, the amount of theoretical Department of Educational Psychology, Neag School and empirical work on the topic has continued to of Education , University of Connecticut , grow (e.g., J. C. Kaufman and Plucker 2011 ; Kell 2131 Hillside Road, Unit 3007 , Storrs , CT 06269-3007 , USA et al. 2013; Kim et al. 2010), providing evidence e-mail: [email protected]; that the topic will remain popular into the future. [email protected] Another reason for heightened attention is the A. Esping growing popularity of the “21st century skills” in College of Education , Texas Christian University , education systems around the globe. Many mod- TCU Box 297900 , Fort Worth , TX 76021 , USA els and frameworks of these skills include con- M. J. Avitia structs that are similar, if not identical, to University of Connecticut Storrs , intelligence and creativity (see Partnership for 456 Tolland Tpke Apt 25a , Willington , CT 06279 , USA 21st Century Skills 2013 ). These models implic- e-mail: [email protected] itly endorse the view that problem solving is a

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 283 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_19, © Springer Science+Business Media New York 2015 284 J.A. Plucker et al. key competency for both life and workplace suc- In addition, most current IQ tests use the Cattell- cess and that intelligence and creativity are inte- Horn- Carroll (CHC; see Flanagan et al. 2007 ) gral parts of the ability to solve problems model or other cognitive theories (e.g., Das et al. effectively and effi ciently. 1994) to give separate index scores in addition to The specifi c relationship of intelligence and producing an overall g score (A. S. Kaufman creativity to each other and to general problem 2009 ). Plucker and Esping (2014 ) note that this solving or other positive outcomes has tremendous bemusement is completely understandable, and practical implications. If intelligence and creativ- they recommend focusing on the defi nitions of ity are very closely entwined, for example, then an the constructs being studied in order to wade overdependence on ability and achievement mea- through and understand the wide range of con- sures (such as the college entrance examinations) ceptualizations. Although they were specifi cally would be less worrisome. If the two constructs are talking about defi nitions of intelligence, the rec- only slightly related, then workplaces that value ommendation holds for defi nitions of creativity, both intelligence and creativity would need to and certainly (and perhaps most appropriately!) make sure to specifi cally assess both abilities. At a for comparisons of the two constructs. broader level, if intelligence and creativity are Fortunately, Sternberg and O’Hara (1999 ) pro- essential and distinct components for successful vide a framework for examining the construct defi - problem solving, then we need to make sure that nitions. Although Sternberg’s approach is nearly our schools and workplaces are doing their best to 15 years old, a similarly productive alternative has nurture and support both abilities. not been subsequently proposed, and most, if not all, subsequent research fi ts into his framework. Due to the durability and usefulness of the model, we In What Ways Can Intelligence use it as the framework for this chapter. Sternberg’s and Creativity Be Related? model suggests fi ve possible relationships: creativity as a subset of intelligence, intelligence as Despite the importance of the creativity- a subset of creativity, creativity and intelligence intelligence relationship, a novice reading the lit- as overlapping sets, creativity and intelligence as erature for the fi rst time could be forgiven for coincident sets, and creativity and intelligence as walking away bewildered: The breadth of theo- disjoint sets. In the following sections, we provide retical treatments is vast, and there is surprisingly examples of each type of relationship. The last two little empirical convergence. For example, thresh- categories, coincident and disjoint sets, are quite old theory suggests intelligence is a necessary but rare and are not described here.1 not a suffi cient condition of creativity (Barron 1969 ; Yamamoto 1964 ), certifi cation theory focuses on environmental factors that allow peo- Creativity as a Subset ple to display creativity and intelligence (Hayes of Intelligence 1989), and an interference hypothesis suggests that very high levels of intelligence may interfere A number of psychometric theories include cre- with creativity (Simonton 1994 ; Sternberg 1996 ). ativity, either explicitly or implicitly, as a part of To complicate things further, intelligence and intelligence, such as Guilford’s Structure of the creativity are each time-consuming to assess Intellect (SOI; 1967) model. He specifi cally well. Most studies use divergent thinking tests to included divergent thinking as a cognitive opera- measure creativity and group-based tests of g to tion within the SOI model – many of the diver- measure intelligence (see Barron and Harrington gent thinking assessments used over the past 50 1981; Kim 2005 ). Although group IQ tests serve a strong purpose in research studies, they are not 1 That said, Nusbaum and Silvia ( 2011) offer some evi- considered ideal for psychoeducational assess- dence of a coincident set perspective, which bears watch- ment (A. S. Kaufman and Lichtenberger 2006 ). ing for future developments. 19 Creativity and Intelligence 285 years originated from this model, and Renzulli intelligence. The CHC model is a combination of (1973 ) developed a creativity curriculum based the Cattell-Horn theory of fl uid and crystallized on the divergent thinking operation. intelligence (Horn and Cattell 1966 , 1967 ) and Gardner (1993 ), coming at the constructs from Carroll’s (1993 ) Three-Stratum Theory. Both the his developmental and qualitative perspective, has Cattell-Horn and Carroll models essentially take used Multiple Intelligence Theory to study cre- Spearman’s (1904 ) g, with Horn and Cattell pro- ativity, implicitly suggesting that creativity is a posing two distinct forms of g: fl uid intelligence subset of MI Theory. In Gardner’s seminal book, (Gf ), the ability to apply a variety of mental oper- Creating Minds, he used case studies of eminent ations to solve novel problems, and crystallized creators to argue that people can excel creatively intelligence (Gc ), the breadth and depth of a per- as a function of embodying different intelligences. son’s accumulated knowledge of a culture and For example, he highlighted Picasso (spatial intel- the ability to use that knowledge to solve prob- ligence), Freud (intrapersonal), Stravinsky (musi- lems (Horn and Cattell 1966 ; see also Horn and cal), Einstein (logical-mathematical), T. S. Eliot Blankson 2005 ). (linguistic), Martha Graham (bodily-kinesthetic), Creativity was originally hypothesized to be and Gandhi (interpersonal). strongly associated to Gf in the early stages of the Another theory that includes creativity as a Cattell-Horn Gf - Gc theory (Cattell and Butcher core component is Sternberg’s (1996 , 1997 , 1968 ). However, this connection is no longer 1999 ; Sternberg et al. 2008 ) theory of successful explicitly part of the CHC theory, in part because intelligence. This theory comprises three “sub- empirical support has been mixed, with some theories”: a componential subtheory , which studies supporting a link between Gc and creativ- relates intelligence to the internal world of the ity (Batey et al. 2009; Cho et al. 2010 ; Furnham individual; an experiential subtheory , which and Chamorro-Premuzic 2006 ) and others pro- relates intelligence to both the external and the viding evidence of a Gf-creativity link (Batey internal worlds of the individual; and a contex- et al. 2010a , b ; Nusbaum and Silvia 2011 ). tual subtheory, which relates intelligence to the In current versions of CHC theory, creativity external world of the individual. The componen- is primarily placed under the category of long- tial subtheory specifi es the mental mechanisms term storage and retrieval (Glr ), the ability to responsible for planning, carrying out, and evalu- store information in and fl uently retrieve new or ating intelligent behavior. The experiential sub- previously acquired information (e.g., concepts, theory expands on this defi nition by focusing on ideas, items, names) from long-term memory. those important behaviors that involve either Indeed, Glr explicitly includes originality/cre- adjustment to relative novelty, automatization of ativity as one of its components (Flanagan et al. information processing, or both. The contextual 2007 ). Glr has two parts, learning effi ciency subtheory defi nes intelligent behavior as involv- (learning and retaining new information) and fl u - ing purposeful adaptation to, selection of, and ency (generating many different possible solu- shaping of real-world environments relevant to tions). Carroll (1993 ) proposed that these two one’s life (Sternberg et al. 2008 ). The experien- parts were distinct abilities, labeled memory and tial subtheory is directly related to creativity. learning and idea production, respectively. With Sternberg’s application of creativity assessments the creation of CHC theory (McGrew 2005 ), they to admissions data increased prediction of col- were combined into Glr . Carroll’s idea produc- lege success beyond that obtained with standard tion has long been associated with creativity admissions tests; in addition, ethnic-group differ- (Schneider and McGrew 2012), most notably ences were signifi cantly reduced (Sternberg Guilford’s (1967 ) operation of divergent 2006 , 2010 , 2012 ; Sternberg and The Rainbow production. Project Collaborators 2005 , 2006 ). McGrew (2009 ) recently noted that “Some Glr The Cattell-Horn-Carroll (CHC) model of narrow abilities have been prominent in creativity intelligence also includes creativity as a subset of research (e.g., production, ideational fl uency, or 286 J.A. Plucker et al. associative fl uency)” (p. 6). In an otherwise (Bjork et al. 1998) is linked to better creative detailed description of the model, this sentence is problem solving. the only mention of creativity, originality, or An interesting extension of the CHC perspec- divergent thinking. Fluid intelligence (Gf ) is dis- tive was provided by Martindale ( 1999 ), who cussed in terms of its relationship to problem proposed that people who are creative are selec- solving and coping with novel problems (both tive with their information processing speed. considered to be highly related to creativity). Early in the creative process, they focus on pro- Nonetheless, in current discussions of the rela- cessing larger amounts of information, but as the tionship of creativity to intelligence as presented problem in question becomes better understood, by the CHC model, the emphasis is on Glr . they shorten their attention span, thereby increas- Several of the narrow abilities incorporated into ing their processing speed. Martindale’s work is Glr , such as ideational fl uency, associational fl u- similar to Sternberg’s (1981 ) hypothesis that ency, and originality/creativity (obviously), have brighter people spend more time in initial global been specifi cally hypothesized to relate strongly planning so that they do not have to spend as to creativity (Kaufman et al. 2011 ). much time in local planning later in the process. There is a striking lack of studies on how cre- A relatively new theory, the Dual Process ativity is empirically connected to Glr . Very Theory of Intelligence (S. B. Kaufman 2013 ), recently, Silvia et al. (2013 ) conducted one of the treats creativity as a subset of intelligence. S. B. few studies to examine this relationship. They Kaufman (2013 ), in an attempt to combine cogni- examined how divergent thinking performance tive models with intelligence research, posits a related to multiple verbal fl uency tests (represent- two-factor model of intelligence, with one factor ing the lower-order Glr factors related to fl u- representing controlled cognition (type 1) – goal- ency). Silvia et al. found that the larger Glr factor directed thoughts and actions that include, but are had a signifi cant effect on both fl uency and origi- entirely explained by, g – and one representing nality in divergent thinking. spontaneous cognition (type 2) – unconsciousness- There is much more work to be done, and this related thoughts and actions that include implicit gap is notable not only because of the theoretical learning ability and daydreaming, among many link between Glr and creativity but also because other constructs. of the existing work on the positive link between In S. B. Kaufman’s view, creativity results memory and creativity. Much of the work on cre- from the combination of type 1 and type 2 pro- ative cognition, from Wallas (1926 ) and later cesses, with the role and importance of each type scholars (Finke et al. 1992 ; Mumford et al. 1991 ) of process varying during any specifi c individu- to neuropsychological models (Bristol and al’s creative process. S. B. Kaufman notes the Viskontas 2006 ; Gabora 2010 ), discusses cogni- pertinent research of Vartanian ( 2009 ) and col- tive processes that apply preexisting knowledge to leagues (Vartanian et al. 2009 ), in which they new concepts. Mednick’s (1962 ) theory of remote provide evidence that creative individuals may associations posited that the ability to organize have the ability to focus attention to varying and access ideas aided creativity by allowing for degrees depending on the context in which their more remote associations between ideas. creative cognition is occurring and the type of Mednick’s (1968 ) measure, the Remote creative activity undertaken. Associates Test, is a common measure of creative problem solving that requires people to fi nd a word that is associated with three other words Intelligence as a Subset (e.g., sleeping, bean, and trash are all connected of Creativity to the word bag). Storm et al. (2011 ) found that when people take the Remote Associates Test, In contrast, other researchers have hypothesized they are more likely to forget other common that intelligence is a part of creativity. Although associated words. This goal-directed forgetting this approach has received scant attention among 19 Creativity and Intelligence 287 intelligence researchers, systems models of Kaufman and Baer 2004 ). In an amusement creativity tend to emphasize the contribution of park, there are initial requirements (e.g., a intelligence and related cognitive processes as one ticket, a ride to the location) that apply to all factor among many that infl uence the development areas of the park. Similarly, there are initial of creativity. As systems theories have grown in requirements that, to varying degrees, are neces- prevalence in the social sciences, they have grown sary for creative performance in all domains. in importance within creativity (Kozbelt et al. One such essential initial requirement is intelli- 2010 ), making this conceptual category rather gence. There are then a series of subcompo- more popular than in the earlier years of the fi eld. nents, general thematic areas , domains , and One major creativity theory that includes microdomains , that get more and more specifi c intelligence is Sternberg and Lubart’s ( 1996 ) (e.g., social sciences to psychology to educa- “investment” theory, in which they use the meta- tional psychology). Different aspects of intelli- phor of the stock market. The key to being cre- gence that are more or less important across ative, they argue, is to buy low and sell high with these specifi c areas, for example, Gc, might be your ideas. In this model, a successful creator particularly important to a historian, whereas Gf will have ideas that may be at fi rst be unpopular might be essential for an engineer. or underappreciated yet will preserve and eventually persuade other people that his or her ideas are valuable. The creator will then know at what Overlapping Sets point to move on to pursue other ideas. According to this theory, six main elements Sternberg’s third grouping conceptualizes intelli- contribute to successful creativity: intelligence, gence and creativity as overlapping yet distinct knowledge, thinking styles, personality, motiva- constructs. Renzulli’s (1978 ) three-ring conception, and the environment. Intelligence contrib- tion of giftedness theorizes that giftedness – utes using three elements drawn from Sternberg’s implicitly cast as high-level creative production – is triarchic theory (1988 , 1996 ; later expanded into caused by the overlap of high intellectual ability, the theory of successful intelligence as described creativity, and task commitment. From this per- earlier). spective, creativity and intelligence are distinct Another theory that casts creativity as being a constructs but overlap considerably under the blend of different abilities is Amabile’s (1982 , right conditions. In a similar vein, the concept of 1996 ) componential model of creativity. She planning abilities in the planning, attention- argued that three variables were needed for cre- arousal, simultaneous and successive (PASS) ativity to occur: domain-relevant skills, theory appears to overlap with creativity (Naglieri creativity- relevant skills, and task motivation. and Kaufman 2001 ), and Plucker et al. (2004 ) Domain-relevant skills include knowledge, tech- view creativity and intelligence as related but dis- nical skills, and specialized talent (i.e., a creative tinct in their defi nition of creativity as “the inter- writer should know basic grammar and styles). action among aptitude, process, and environment Creativity-relevant skills are more personal fac- by which an individual or group produces a per- tors that have been associated with creativity. ceptible product that is both novel and useful as These can include tolerance for ambiguity, self- defi ned within a social context” (p. 90). discipline, and risk-taking. Finally, Amabile highlights motivation (intrinsic or extrinsic) Threshold Theory. Traditional research has toward the task at hand. Intelligence would pri- argued for a threshold theory, in which creativity marily occur at the domain-relevant skill level. and intelligence are positively, if moderately, A third theory that accounts for multiple correlated up until an IQ of approximately 120; variables but takes a more domain-specifi c in people with higher IQs, the two constructs approach is the Amusement Park Theoretical show little relationship (e.g., Fuchs-Beauchamp Model (APT Model; Baer and Kaufman 2005 ; et al. 1993 ; Getzels and Jackson 1962 ). Sternberg 288 J.A. Plucker et al. places this perspective in the overlapping sets et al. (2007 , 2008 ) found that, within this very category. This view is so common as to be intelligent group, intellectual talent was highly considered part of the conventional wisdom correlated with educational attainment. This by about creativity, intelligence, and giftedness. itself is not surprising, but they also found that a Yet some recent work has called into question range of indicators of adult creative accomplish- the presence of a threshold (whether lower or ments (e.g., patents, publications, awards) was higher than a 120 IQ). For example, Preckel et al. also correlated with intelligence. Wai et al. (2005 ), (2006 ) studied Gf and creativity (as measured looking at the same population, found that differ- through divergent thinking tests) and found mod- ences in SAT scores – even within such an elite est correlations across all levels of intellectual group – predicted creative accomplishments 20 abilities. In a meta-analysis of 21 studies, Kim years later. These results, emerging as they do ( 2005) found virtually no support for the thresh- from studies that address the limitations of previ- old theory; there were small positive correlations ous research, raise serious doubts about the between measures of ability and measures of cre- threshold effect, and they fi rmly refl ect a belief in ativity and divergent thinking. the overlapping sets perspective. Many of the early studies that formed the basis Recent research by Beaty and Silvia (2012 ) of threshold theory would now be considered provides potential insight into the mechanisms at quite dated. Any study conducted before the play here. They had college students complete 1970s would have (obviously) used measures of divergent thinking tasks over a 10-min period, intelligence that do not refl ect current theory. and they unsurprisingly found that participants Other studies have defi ned creativity rather nar- reported more creative ideas as they progressed rowly as being only divergent thinking or the throughout the session, as time marched on. ability to generate multiple ideas in response to a However, they also administered a Gf measure, single prompt. and – surprisingly – the higher the Gf score, the Fortunately, researchers have begun to address fl atter the slope of the creativity-time curve. This these limitations, with interesting results. For means that the most intelligent people in their example, Sligh et al. (2005 ) used a contemporary, sample did not come up with appreciably more individually administered IQ test (Kaufman creative ideas over time, rather providing fairly Adolescent and Adult Intelligence Test; A. S. creative ideas from the beginning to the end of Kaufman and Kaufman 1993 ) and a creative the session. Less intelligent participants had invention task (in which people use shapes to cre- increasingly steep slopes, meaning that they def- ate product objects and then name and describe initely were more creative as time progressed. their invention; see Finke 1990 ). By assessing This study raises the possibility that there are both Gc and Gf , they were able to show moder- underlying cognitive mechanisms behind recent ate, positive correlations between Gc and creativ- observations that intelligence and creativity are ity (i.e., similar to previous studies); however, correlated even at high levels of intelligence and intelligence and creativity were signifi cantly cor- that those mechanisms may be a combination of related for the high-IQ group, which was not the executive processes related to information case for people with average intelligence scores – retrieval and manipulation and associative pro- an opposite pattern than predicted by threshold cesses that involve activation of various parts of theory. one’s cognitive schema. In a similar line of research (but with much dif- One recent study, by Jauk et al. (2013 ), uses ferent participants), the Study of Mathematically sophisticated statistical techniques to support the Precocious Youth has been following a cohort of threshold effect, if not the traditional cutoff level. students from late childhood/early adolescence They used segmented regression analysis and into adulthood. These students all scored in the found that the threshold level varied based on the top 1 % on college entrance examinations before creativity score used. Ideational fl uency’s thresh- the age of 13, so they are a very bright group. Park old was at an IQ of 86 (which is astoundingly 19 Creativity and Intelligence 289 lower than the more traditional 120); if the top Barron, F. (1963). Creativity and psychological health . two originality scores were used, the threshold Princeton: D. Van Nostrand Company. Barron, F. (1969). Creative person and creative process . was 104. New York: Holt, Rinehart & Winston. Barron, F., & Harrington, D. M. (1981). Creativity, intelligence, and personality. Annual Review of Psychology, Conclusion 32 , 439–476. Batey, M., Chamorro-Premuzic, T., & Furnham, A. (2009). Intelligence and personality as predictors of Each of the fi ve possible relationships in divergent thinking: The role of general, fl uid, and crys- Sternberg’s framework enjoys at least some tallized intelligence. Thinking Skills and Creativity, 4 , empirical support (Sternberg and O’Hara 1999 ; 60–69. Batey, M., Chamorro-Premuzic, T., & Furnham, A. Kaufman and Plucker 2011 ), but the diffi culty in (2010a). Individual differences in ideational behavior: interpreting empirical results illustrates the prob- Can the big fi ve and psychometric intelligence predict lems associated with reaching a consensus on the creativity scores? Creativity Research Journal, 22 , validity of any of these fi ve relations. For exam- 90–97. Batey, M., Furnham, A., & Safi ullina, X. (2010b). ple, Haensly and Reynolds (1989 ) believe that Intelligence, general knowledge and personality as Mednick’s (1962 ) Association Theory supports predictors of creativity. Learning and Individual the creativity as a subset of intelligence position, Differences, 20 , 532–535. yet Sternberg and O’Hara (1999 ) feel that this Beaty, R. E., & Silvia, P. J. (2012). Why do ideas get more creative across time? An executive interpretation of body of work supports the overlapping sets posi- the serial order effect in divergent thinking tasks. tion. If Gardner’s work with creativity had come Psychology of Aesthetics, Creativity, and the Arts, 6 , before his work with MI Theory, we would prob- 309–319. ably be arguing that his efforts fall within the Bjork, E. L., Bjork, R. A., & Anderson, M. C. (1998). Varieties of goal-directed forgetting. In J. M. Golding & C. MacLeod intelligence as a subset of creativity category. (Eds.), Intentional forgetting: Interdisciplinary approaches As Plucker and Esping (2014 ) note, defi nitions (pp. 103–137). Hillsdale: Erlbaum. are critically important when dealing with psycho- Bristol, A. S., & Viskontas, I. V. (2006). Dynamic pro- logical constructs, as the way in which each con- cesses within association memory stores: Piecing together the neural basis of creative cognition. In J. C. struct is conceptualized and assessed will have a Kaufman & J. Baer (Eds.), Creativity and reason in signifi cant impact on any empirical results when cognitive development (pp. 60–80). New York: comparing two or more constructs. As this chapter Cambridge University Press. has showed, the range of creativity and intelligence Burgess, G. C., Gray, J. R., Conway, A. R., & Braver, T. S. (2011). Neural mechanisms of interference control defi nitions makes the complexity of possible intel- underlie the relationship between fl uid intelligence ligence-creativity relationships unsurprising. Few and working memory span. Journal of Experimental people believe creativity and intelligence are com- Psychology: General, 140 , 674–692. pletely unrelated, but the nature of any relation- Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies . New York: Cambridge ship – even with so much research already University Press. conducted – is an open question. Cattell, R. B., & Butcher, H. (1968). The prediction of achievement and creativity . Indianapolis: Bobbs-Merrill. Cho, S. H., Nijenhuis, J. T., van Vianen, N. E. M., Kim, H.-B., & Lee, K. H. (2010). The relationship between References diverse components of intelligence and creativity. Journal of Creative Behavior, 44 , 125–137. Amabile, T. M. (1982). Social psychology of creativity: A Das, J. P., Naglieri, J. A., & Kirby, J. R. (1994). Assessment consensual assessment technique. Journal of of cognitive processes: The PASS theory of intelli- Personality and Social Psychology, 43 , 997–1013. gence . Boston: Allyn & Bacon. Amabile, T. M. (1996). Creativity in context: Update to Deary, I. J., Weiss, A., & Batty, G. D. (2010). Intelligence “The Social Psychology of Creativity” . Boulder: and personality as predictors of illness and death: How Westview Press. researchers in differential psychology and chronic dis- Baer, J., & Kaufman, J. C. (2005). Bridging generality and ease epidemiology are collaborating to understand and specifi city: The Amusement Park Theoretical (APT) address health inequalities. Psychological Science in Model of creativity. Roeper Review, 27 , 158–163. the Public Interest, 11 (2), 53–79. 290 J.A. Plucker et al.

Finke, R. (1990). Creative imagery: Discoveries and Kaufman, A. S., & Lichtenberger, E. O. (2006). Assessing inventions in visualization . Hillsdale: Erlbaum. adolescent and adult intelligence (3rd ed.). New York: Finke, R. A., Ward, T. B., & Smith, S. M. (1992). Creative Wiley. cognition: Theory, research, and applications . Kaufman, J. C., & Plucker, J. (2011). Creativity and intel- Cambridge, MA: MIT Press. ligence. In R. J. Sternberg & S. B. Kaufman (Eds.), Flanagan, D. P., Ortiz, S., & Alfonso, V. C. (2007). The Cambridge handbook of intelligence (pp. 771– Essentials of cross-battery assessment (2nd ed.). 783). New York: Cambridge. New York: Wiley. Kaufman, J. C., Kaufman, S. B., & Lichtenberger, E. O. Fuchs-Beauchamp, K. D., Karnes, M. B., & Johnson, L. J. (2011). Finding creativity on intelligence tests via (1993). Creativity and intelligence in preschoolers. divergent production. Canadian Journal of School Gifted Child Quarterly, 37 , 113–117. Psychology, 26 , 83–106. Furnham, A., & Chamorro-Premuzic, T. (2006). Kell, H. J., Lubinski, D., Benbow, C. P., & Steiger, J. H. Personality, intelligence and general knowledge. (2013). Creativity and technical innovation: Spatial Learning and Individual Differences, 16 , 79–90. ability’s unique role. Psychological Science, 24 , Gabora, L. (2010). Revenge of the ‘neurds’: Characterizing 1831–1836. creative thought in terms of the structure and dynamics of Kim, K. H. (2005). Can only intelligent people be cre- human memory. Creativity Research Journal, 22, 1–13. ative? Journal of Secondary Gifted Education, 16 , Galton, F. (1869). Hereditary genius . London: Macmillan 57–66. and Company. Kim, K. H., Cramond, B., & VanTassel-Baska, J. (2010). Gardner, H. (1993). Creating minds . New York: Basic The relationship between creativity and intelligence. Books. In J. C. Kaufman & R. J. Sternberg (Eds.), Cambridge Getzels, J. W., & Jackson, P. W. (1962). Creativity and handbook of creativity (pp. 395–412). New York: intelligence: Explorations with gifted students . Cambridge University Press. New York: Wiley. Kozbelt, A., Beghetto, R. A., & Runco, M. A. (2010). Guilford, J. P. (1967). The nature of human intelligence . Theories of creativity. In R. J. Sternberg & J. C. New York: McGraw-Hill. Kaufman (Eds.), The Cambridge handbook of creativ- Haensly, P. A., & Reynolds, C. R. (1989). Creativity and ity (pp. 20–47). New York: Cambridge. intelligence. In J. A. Glover, R. R. Ronning, & C. R. Martindale, C. (1999). Biological bases of creativity. In Reynolds (Eds.), Handbook of creativity (pp. 111– R. J. Sternberg (Ed.), Handbook of creativity (pp. 137– 132). New York: Plenum. 152). New York: Cambridge University Press. Hayes, J. R. (1989). Cognitive processes in creativity. In McGrew, K. S. (2005). The Cattell-Horn-Carroll theory of J. A. Glover, R. R. Ronning, & C. R. Reynolds (Eds.), cognitive abilities: Past, present, and future. In D. P. Handbook of creativity (pp. 135–145). New York: Flanagan & P. L. Harrison (Eds.), Contemporary intel- Plenum. lectual assessment: Theories, tests, and issues (2nd Horn, J. L., & Blankson, N. (2005). Foundations for better ed., pp. 136–181). New York: Guilford Press. understanding of cognitive abilities. In D. P. Flanagan McGrew, K. S. (2009). CHC theory and the human cogni- & P. L. Harrison (Eds.), Contemporary intellectual tive abilities project: Standing on the shoulders of the assessment: Theories, tests, and issues (2nd ed., giants of psychometric intelligence research. pp. 41–68). New York: Guilford. Intelligence, 37 , 1–10. Horn, J. L., & Cattell, R. B. (1966). Refi nement and test of Mednick, S. A. (1962). The associative basis of the cre- theory of fl uid and crystallized intelligence. Journal of ative process. Psychological Review, 69 , 220–232. Educational Psychology, 57 , 253–270. Mednick, S. A. (1968). The remote associates test. Journal Horn, J. L., & Cattell, R. B. (1967). Age differences in of Creative Behavior, 2 , 213–214. fl uid and crystallized intelligence. Acta Psychologica, Mumford, M. D., Mobley, M. I., Uhlman, C. E., Reiter- 26 , 107–129. Palmon, R., & Doares, L. M. (1991). Process analytic Jauk, E., Benedek, M., Dunst, B., & Neubauer, A. C. models of creative capacities. Creativity Research (2013). The relationship between intelligence and cre- Journal, 4 , 91–122. ativity: New support for the threshold hypothesis by Naglieri, J. A., & Kaufman, J. C. (2001). Understanding means of empirical breakpoint detection. Intelligence, intelligence, giftedness, and creativity using PASS 41 , 212–221. theory. Roeper Review, 23 , 151–156. Kaufman, A. S. (2009). IQ Testing 101 . New York: Nusbaum, E. C., & Silvia, P. J. (2011). Are intelligence Springer. and creativity really so different? Fluid intelligence, Kaufman, S. B. (2013). Ungifted: Intelligence redefi ned . executive processes, and strategy use in divergent New York: Basic Books. thinking. Intelligence, 39 , 36–45. Kaufman, J. C., & Baer, J. (2004). The Amusement Park Park, G., Lubinski, D., & Benbow, C. P. (2007). Theoretical (APT) Model of creativity. Korean Journal Contrasting intellectual patterns predict creativity in of Thinking and Problem Solving, 14 , 15–25. the arts and sciences. Psychological Science, 18 , Kaufman, A. S., & Kaufman, N. L. (1993). Kaufman 948–952. Adolescent and Adult Intelligence Test (KAIT) . Circle Park, G., Lubinski, D., & Benbow, C. P. (2008). Ability Pines: American Guidance Service. differences among people who have commensurate 19 Creativity and Intelligence 291

degrees matter for scientifi c creativity. Psychological Sternberg, R. J. (1999). The theory of successful intelli- Science, 19 , 957–961. gence. Review of General Psychology, 3 , 292–316. Partnership for 21st Century Skills. (2013). Framework Sternberg, R. J. (2006). Creating a vision of creativity: for 21st century learning . Accessed at http://www. The fi rst 25 years. Psychology of Aesthetics, Creativity, p21.org/our-work/p21-framework and the Arts, S , 2–12. Plucker, J., & Esping, A. (2014). Intelligence 101 . Sternberg, R. J. (2010). Teaching for creativity. In R. A. New York: Springer. Beghetto & J. C. Kaufman (Eds.), Nurturing creativity Plucker, J., & Renzulli, J. S. (1999). Psychometric in the classroom (pp. 394–414). New York: Cambridge approaches to the study of human creativity. In R. J. University Press. Sternberg (Ed.), Handbook of creativity (pp. 35–60). Sternberg, R. J. (2012). The assessment of creativity: An New York: Cambridge University Press. investment-based approach. Creativity Research Plucker, J., Beghetto, R. A., & Dow, G. (2004). Why isn’t Journal, 24 , 3–12. creativity more important to educational psycholo- Sternberg, R. J., & The Rainbow Project Collaborators. gists? Potential, pitfalls, and future directions in cre- (2005). Augmenting the SAT through assessments of ativity research. Educational Psychologist, 39 , 83–96. analytical, practical, and creative skills. In W. Camara Preckel, F., Holling, H., & Wiese, M. (2006). Relationship & E. Kimmel (Eds.), Choosing students: Higher edu- of intelligence and creativity in gifted and non-gifted cation admission tools for the 21st century (pp. 159– students: An investigation of threshold theory. 176). Mahwah: Lawrence Erlbaum Associates. Personality and Individual Differences, 40 , 159–170. Sternberg, R. J., & Lubart, T. I. (1996). Defying the crowd . Renzulli, J. S. (1973). New directions in creativity . New York: Free Press. New York: Harper & Row. Sternberg, R. J., & O’Hara, L. A. (1999). Creativity and Renzulli, J. S. (1978). What makes giftedness? Reexamining intelligence. In R. J. Sternberg (Ed.), Handbook of cre- a defi nition. Phi Delta Kappan, 60 , 180–261. ativity (pp. 251–272). Cambridge, MA: Cambridge Schneider, W. J., & McGrew, K. (2012). The Cattell- University Press. Horn- Carroll model of intelligence. In D. Flanagan & Sternberg, R. J., & The Rainbow Project Collaborators. P. Harrison (Eds.), Contemporary intellectual assess- (2006). The Rainbow Project: Enhancing the SAT ment: Theories, tests, and issues (3rd ed., pp. 99–144). through assessments of analytical, practical and cre- New York: Guilford. ative skills. Intelligence, 34 , 321–350. Silvia, P. J., Beaty, R. E., & Nusbaum, E. C. (2013). Verbal Sternberg, R. J., Kaufman, J. C., & Grigorenko, E. L. fl uency and creativity: General and specifi c contribu- (2008). Applied intelligence . Cambridge: Cambridge tions of broad retrieval ability (Gr) factors to divergent University Press. thinking. Intelligence, 41 , 328–340. Storm, B. C., Angello, G., & Bjork, E. L. (2011). Thinking Simonton, D. K. (1994). Greatness: Who makes history can cause forgetting: Memory dynamics in creative and why . New York: Guilford Press. problem solving. Journal of Experimental Psychology Sligh, A. C., Conners, F. A., & Roskos-Ewoldsen, B. Learning Memory and Cognition, 37 , 1287–1293. (2005). Relation of creativity to fl uid and crystallized Vartanian, O. (2009). Variable attention facilitates creative intelligence. Journal of Creative Behavior, 39 , problem solving. Psychology of Aesthetics, Creativity, 123–136. and the Arts, 3 , 57–59. Spearman, C. (1904). “General intelligence”, objectively Vartanian, O., Martindale, C., & Matthews, J. (2009). determined and measured. American Journal of Divergent thinking ability is related to faster related- Psychology, 15 , 201–293. ness judgments. Psychology of Aesthetics, Creativity, Stanovich, K. E., West, R. F., & Toplak, M. E. (2013). and the Arts, 3 , 99–103. Myside bias, rational thinking, and intelligence. Wai, J., Lubinski, D., & Benbow, C. P. (2005). Creativity Current Directions in Psychological Science, 22 , and occupational accomplishments among intellectu- 259–264. ally precocious youths: An age 13 to age 33 longitudi- Sternberg, R. J. (1981). Intelligence and nonentrench- nal study. Journal of Educational Psychology, 97 , ment. Journal of Educational Psychology, 73 , 1–16. 484–492. Sternberg, R. J. (1988). A three-facet model of creativity. Wallach, M. A., & Kogan, N. (1965). Modes of thinking in In R. J. Sternberg (Ed.), The nature of creativity young children: A study of the creativity-intelligence (pp. 125–147). New York: Cambridge University distinction . New York: Holt, Rinehart and Winston. Press. Wallas, G. (1926). The art of thought. New York: Harcourt Sternberg, R. J. (1996). Successful intelligence . New York: Brace. Simon & Schuster. Yamamoto, K. (1964). Creativity and sociometric choice Sternberg, R. J. (1997). Successful intelligence . New York: among adolescents. Journal of Social Psychology, 64 , Plume. 249–261. Part IV

Assessment of Intelligence Hundred Years of Intelligence Testing: Moving from Traditional 20 IQ to Second-Generation Intelligence Tests

Jack A. Naglieri

“Do not go where the path may lead, go instead where there is no path and leave a trail.” –Ralph Waldo Emerson

Training School in Vineland, New Jersey, on May C o n t e x t 28. The committee considered many types of group tests and several that Arthur S. Otis devel- April 6, 1917, is remembered as the day the oped when working on his doctorate under Lewis United States entered World War I. On that same Terman at Stanford University. The goal was to day a group of psychologists held a meeting in fi nd tests that could effi ciently evaluate a wide Harvard University’s Emerson Hall to discuss the variety of men, be easy to administer in the group possible role they could play with the war effort format, and be easy to score. By June 9, 1917, the (Yerke s 1921 ). The group agreed that psycho- materials were ready for an initial trial. Men who logical knowledge and methods could be of had some educational background and could importance to the military and utilized to speak English were administered the verbal and increase the effi ciency of the Army and Navy quantitative (Alpha) tests and those that could not personnel. The group included Robert Yerkes, read the newspaper or speak English were given who was also the president of the American the Beta tests (today described as nonverbal). Psychological Association. Yerkes made an The Alpha tests were designed to measure appeal to members of APA who responded by general information (e.g., how many months are providing a group of psychologists to assist with there in a year?), common sense (e.g., why do we the war effort. Members from APA were joined use stoves?), and verbal knowledge (synonyms/ by psychologists of the National Research antonyms, verbal analogies, number series, disar- Council, the National Academy of Sciences, and ranged sentences) (e.g., determine if a group of the American Association for the Advancement words could be sequenced to make a true state- of Science, and a number of committees were ment) and to determine how well the examinee organized to develop effective measures of could follow verbal directions (e.g., draw a line ability. from circle 3 to circle 6). The Beta tests included One group of psychologists whose task was to completion of a maze, construction of a design begin identifying possible tests met at the using blocks, number symbol association, identifying what is missing in a picture, and copying geometric shapes. Why two tests? Because, as J. A. Naglieri , Ph.D. (*) Yoakum and Yerkes (1920 ) clearly stated, the University of Virginia , 6622 White Post Road , Centreville , VA 20121 , USA Alpha test was an appropriate measure of intelli- e-mail: [email protected] gence for men who could read and write English

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 295 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_20, © Springer Science+Business Media New York 2015 296 J.A. Naglieri suffi ciently. The Beta tests were intended for of those data, and R. Cattell was initially on the those who had diffi culty reading or spoke English original National Research Council meeting that poorly as well as those who were illiterate or not led to the development of the measures. able to understand English (p. 51). The testing The transition from Army Alpha and Beta to procedures ensured that men “who fail in alpha Wechsler IQ as we know it today is the result of are sent to beta in order that injustice by reason of the work of those psychologists who fi rst met on relative unfamiliarity with English may be April 6, 1917, and those that developed, vali- avoided (Yoakum & Yerkes, p. 19).” dated, and used the Army Alpha and Beta tests. It By July 7, 1917, the initial development of would likely have been well beyond the expecta- tests to measure intellectual ability and a study tions of Major Lewis M. Terman, Captain Edwin involving 400 cases was completed. The data G. Boring, and related personnel including obtained from testing sites in Indiana, Tennessee, R. S. Woodworth, E. L. Thorndike, A. S. Otis, and Syracuse and Brooklyn, New York, was and L. L Thurstone that the work they did would shipped to the statistical unit in New York City become the most widely used approach to mea- for data analysis. Statistical analysis was directed suring intelligence in history. And in addition, by E. L. Thorndike with assistants A. S. Otis and their work (described in the book Army Mental L. L. Thurstone. The report of their analysis com- Tests by Yoakum and Yerkes 1920 ) would defi ne pleted on July 20, 1917, showed that the tests the structure of individual- (e.g., Wechsler could be appropriate to (a) “aid in segregating Scales) as well as group-administered (Otis- and eliminating the mentally incompetent, (b) Lennon 1979 ) IQ tests for the next 100 years. classify men according to their mental ability, The evolution of traditional IQ tests from their and (c) assist in selecting competent men for birth in 1917 has been defi ned by the many revi- responsible positions” (Yerkes 1921 , p. 19). sions of Wechsler and Otis-Lennon tests, the most Thus, July 20, 1917, could be considered the current version of the latter being number 8 and the birth date of the verbal, quantitative, and nonver- forthcoming fi fth edition of the Wechsler bal IQ test format which will be called traditional Intelligence Scale for Children (WISC-V, Wechsler IQ in the remainder of this chapter. 2014). Ironically, there have been many revisions By early 1918 a group of about 260 men trained of these tests long after the authors died (Otis in in the Medical Corps School for Military 1964 and Wechsler in 1981). What has evolved is a Psychology began using the Army Alpha and Beta body of knowledge about these tests and how to tests. Among them was the 22-year-old David interpret them. For example, the fi rst book on the Wechsler (1896–1981) who arrived at Fort Logan, development and interpretation of the Wechsler Texas, in August that year. Wechsler, like Yerkes, Scales was the 1939 publication of Wechsler’s who wrote in the Introduction to the Psychological Measurement and Appraisal of Adult Intelligence . Examining in the United States Army ( 1921 ), The fi fth edition of that book was published noticed “the educational, industrial, and signifi - (authored by Matarazzo) in 1976. Shortly thereaf- cance of the methods [Alpha and Beta] (p. 5)”. ter, Alan Kaufman’s 1979 book, Intelligence Wechsler’s adaptation some 20 years later resulted Testing with the WISC-R, provided practitioners in the Wechsler-Bellevue Intelligence Scale with a wealth of information about both psycho- ( 1939) which also contained verbal and quantita- metrically sound and clinically relevant interpretative (the Alpha) and performance (the Beta) test tion methods. This was followed by Intelligence questions. These tests would have been familiar to Testing with the WISC-III (Kaufman 1994 ) and the founders of the company that ultimately pub- WISC-IV (Flanagan and Kaufman 2004 ). Other lished Wechsler’s test (Wechsler 2005 ), because authors have also provided books on how to extract many years earlier (in 1917), R. S. Woodworth information about Wechsler’s scales (e.g., Prifi tera was the chief examiner at the Brooklyn testing site and Saklofske 1998 ; Weiss et al. 2006 ). All this collecting initial data on the Army Alpha and effort has been focused on ways to interpret a test Beta, Thorndike was responsible for the analysis based on measures fi rst assembled in 1917. 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 297

Despite the widespread use and acceptance of nized the fact that traditional IQ lacked a theory, the traditional verbal/quantitative (Alpha) and just as Pintner (1923 ) noted when he wrote “we nonverbal (Beta) IQ approach to intelligence, it is did not start with a clear defi nition of general important to ask if this approach meets the intelligence… [but] borrowed from every-day demands of psychology and education today. Are life a vague term implying all-round ability and… the tests appropriate for diverse populations? we [are] still attempting to defi ne it more sharply Does the information traditional IQ tests yield and endow it with a stricter scientifi c connota- assist in planning instruction and academic inter- tion” (p. 53).” The Kaufman’s stressed the point ventions? How well do these tests evaluate the that a test of intelligence should be built upon a intellectual component of a specifi c learning dis- theory of intelligence. ability? Perhaps most importantly, should we Another second-generation ability test which continue to use IQ tests which were devoid of a was published by Naglieri and Das in 1977 fur- theoretical foundation from the time they were ther stimulated the evolution of the fi eld. This fi rst introduced more than 100 years ago (Naglieri test, called the Cognitive Assessment System , and and Kaufman 2008 )? The purpose of this chapter the more recent CAS2 (Naglieri et al. 2014 ), was is to provide answers to these questions and evi- developed on a specifi c theory derived from the dence for an alternative to traditional IQ based on integration of cognitive and neuropsychological brain function. The short answer is yes, our research described by A.R. Luria (1963 , 1966 , understanding of intelligence has evolved and 1973, 1980 ). The theory is called PASS which better ways to measure it are now available. stands for planning, attention, simultaneous, and successive neurocognitive abilities. Planning is the ability to perform complex decision making Starting Over (related to the frontal lobes); attention is the ability to focus thinking and resist distractions The evolution of intelligence tests was stimulated (related to the brain stem); simultaneous pro- by the publication of two second-generation abil- cessing ability is needed for understanding inter- ity tests. First was the publication of the Kaufman relationships (occipital/parietal area); and Assessment Battery for Children (Kaufman and successive processing ability is used whenever Kaufman 1983 ). This test was based on a blend of sequencing is required (temporal lobes). The perspectives about what intelligence may be. What PASS theory was initially presented in the book the Kaufman’s did that was most important was to Assessment of Cognitive Processes: The PASS tie a test of intelligence with a two-dimensional Theory of Intelligence (Das et al. 1994 ) and elab- model of ability conceptualized within a cognitive orated by Naglieri (1999 ). More recently, the processing context. Another very important aspect validity of the PASS theory as measured by the of the KABC and the second edition (K-ABC-II, CAS and CAS2 is summarized in several Kaufman and Kaufman 2004 ) was the shift away resources (e.g., Naglieri and Das 2001; Naglieri from organizing their test based on the verbal, 2012; Naglieri and Goldstein 2011; Naglieri and quantitative, and nonverbal content of the test Otero 2011 ; Naglieri and Conway 2009 ; Naglieri questions. Instead, tests with verbal content were et al. 2012, 2014 ). placed on an achievement scale and ability was When the KABC and CAS were introduced, measured using the sequential and simultaneous these second-generation tests marked a change in processing scales. This shift in emphasis from test the way intelligence was conceptualized and, just content to the process needed to solve the problem as importantly, measured. These two tests are put more emphasis on the cognitive activities of most alike in their emphasis on measuring ability the examinee and resulted in a dramatic change in separately from academic skills. That is, they conceptualization of ability. move away from the verbal/quantitative (Alpha) The Kaufmans’ emphasis on the need for a and nonverbal (Beta) organization of questions. view of ability was also important. They recog- The authors of these tests recognized that even 298 J.A. Naglieri though verbal and nonverbal are often described do. Children are required to defi ne a word like as types of intelligence, the authors of the Alpha “bat” on subtests included in the SB-5 and and Beta never intended to measure two abilities, WISC-IV intelligence tests and the WJ-III achieve- and neither did Wechsler. Wechsler’s view of ment test. The WJ-III Cognitive battery contains a intelligence was not that verbal and nonverbal Verbal Comprehension subtest that has the item were two types of intelligence. Despite the fact like “tell me another word for small” and the that his tests yielded verbal and performance WJ-III Achievement battery contains a Reading (nonverbal) IQ scores, Wechsler (1958) wrote: Vocabulary question: “tell me another word for “the subtests are different measures of intelli- little.” In addition, an item on the WJ-III Reading gence, not measures of different kinds of intelli- Vocabulary achievement test is “Tell me another gence” (1958, p. 64). Boake (2002 ) noted that word for (examiner points to the word big),” and in Wechsler viewed verbal and nonverbal (also a Cognitive battery the examiner asks something labeled as performance) tests as equally valid like: “Tell me another word for tiny.” Additionally, measures of intelligence. Similarly, Naglieri the WJ-III Cognitive battery Verbal Comprehension (2003 ) clarifi ed that the terms verbal and nonver- contains Picture Vocabulary items, and the WJ-III bal “refer to the content of the test, not a type of Achievement battery includes Picture Vocabulary ability” (p. 2). Moreover, Wechsler argued that items, some of which are the same. The WJ-III nonverbal tests help to “minimize the over- Cognitive tests also require the subject to name as diagnosing of feeble-mindness that was, he many examples as possible from a given category believed, caused by intelligence tests that were in a 1-min time period and the same question too verbal in content… and he viewed verbal and appears on the WIAT-II Oral Expression achieve- performance tests as equally valid measures of ment test. These examples do not comprise a com- intelligence and criticized the labeling of perfor- plete list of item overlap but do represent the most mance [nonverbal] tests as measures of special strikingly similar questions. abilities” (Boake 2002 , p. 396). Tests of intelligence and achievement also Elimination of Army Alpha-type questions from include arithmetic test questions. For example, a measure of intelligence was a bold move by these the oldest intelligence test, now in its fi fth edition authors of second-generation ability tests and one is the Stanford-Binet 5 (SB-5; Roid, 200×), con- that raised two important questions: (1) “How sim- tains Quantitative Reasoning items, one of which ilar are verbal and quantitative test questions on an requires the child to calculate the total number of IQ test to an achievement test?” and (2) “Can ver- stars on a page (e.g., two stars in one box plus bal and quantitative tasks be taken out of a measure four in a second box plus one in a third box). of intelligence without losing validity?” Similarly, the Wechsler Intelligence Scale for Children – Fourth Edition (WISC-IV; Wechsler 2003) arithmetic subtest requires the child to Do Verbal and Quantitative Test count the number of butterfl ies pictured on a Questions Differ from Achievement page. Although the scores these test items yield Test Questions? are used to determine the child’s level of intelligence, very similar items appear on the Wechsler It would seem reasonable that an IQ test should Individual Achievement Test (WIAT-II, Wechsler, measure something different than an academic 200×). On that test, for example, a Numerical achievement test, but this is not the case. The ver- Operations subtest item requires the child to bal and quantitative portions of traditional IQ tests determine the total number of marbles shown are remarkably similar to questions found in the (e.g., 3 plus 5). Similarly, a Woodcock-Johnson achievement tests used to measure knowledge and Tests of Achievement (WJ-III; Woodcock et al. skills. For example, verbal questions are found on 2001 ) Applied Problems subtest item asks the both traditional IQ tests and measures of achieve- child to count the number of crayons pictured on ment. All traditional IQ tests include a measure of the stimulus book (e.g., 4). Moreover, a SB-5 word knowledge just as measures of achievement Quantitative Reasoning item requires the child to 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 299 complete a simple math problem (e.g., 3 + 2 = 5) to achievement tests enhances the predictive just as the WJ-III Math Fluency (e.g., 5 + 2 = ?) validity of these measures but at a cost to those and the WIAT-II Numerical Operations (e.g., with limited educational backgrounds (recall that 2 + 2 = ?) achievement tests do. There is lack of the Beta tests were used to measure intelligence distinction between the arithmetic questions on fairly for those with limited familiarity with these tests of achievement and intelligence, yet English). The question remains, however, can sec- the interpretations of the scores each test yields ond-generation intelligence tests correlate with are considerably different. In one instance the achievement as well as traditional IQ? This ques- score is used to determine level of math achieve- tion was examined by Naglieri ( 1999) who fi rst ment, but in the other the scores are used to deter- reported that the correlations between achievement mine level of intelligence. The same overlap in test scores with the CAS and KABC were as high content is found for verbal tests. as or higher than those found for the WISC-III and The use of items with similar content across WJ-R. More recent fi ndings are provided next. achievement and IQ tests is alarming for several Naglieri and Rojahn ( 2004) examined the rela- reasons. First, because the correlation between IQ, tionships between the planning, attention, simulta- especially verbal sections of IQ tests, and achieve- neous, and successive (PASS) theory, as ment test scores has been considered a source of operationalized by the CAS, and achievement, as evidence for the validity of IQ tests, the correla- measured by the Woodcock-Johnson Tests of tions between IQ tests with verbal/quantitative Achievement – Revised (WJ-R; Woodcock and items and achievement tests should be considered Johnson 1989), using a nationally representative overestimates of the relationship between ability sample of 1,559 students. The correlation between and achievement. Moreover, the authors and/or the CAS Full Scale with the WJ-R Tests of publishers of IQ tests should justify how similar Achievement was .71. More recently, Naglieri, questions can be used across supposedly different Goldstein, DeLauder, and Schwebach (2006a ) constructs and how very different interpretations compared the Wechsler Intelligence Scale for (e.g., achievement vs. intelligence) can be made Children – Third Edition (WISC-III; Wechsler given the similarity of item content. Second, the 1991) to the CAS and the WJ-III Test of obvious achievement content must be justifi ed by Achievement. The CAS Full Scale score correlated those that use these IQ tests when assessing cultur- .83 with the WJ-III achievement scores compared ally and/or linguistically diverse children and to a coeffi cient of .63 for the WISC-III Full Scale especially Hispanics who now constitute the larg- IQ. The results suggest that when the same children est minority group in the United States. This group took the two ability tests and those scores were cor- is particularly at risk of being misdiagnosed related with the same achievement scores, both because they often have parents with limited edu- showed a strong relationship between ability and cational background and/or limited English lan- achievement, but the CAS correlated signifi cantly guage skills (Ramirez and de la Cruz 2002 ) which higher (Naglieri et al. 2006a ). Most recently, reduces the opportunity to acquire the knowledge Naglieri et al. (2014 ) reported an average correla- of words (Hart and Risley 1995 ). tion between the CAS2 and achievement of .70. The KABC-II Mental Processing Index (MPI) which excludes measures of knowledge corre- Can Verbal and Quantitative lated, on average, .68 with total achievement on Tasks Be Taken Out of a Measure the Peabody Individual Achievement Test – of Intelligence Without Losing Revised (PIAT-R; Markwardt 1997 ), .70 with total Validity? achievement on the WJ-III Achievement Scale, .74 with total achievement on the Wechsler Individual Explaining current academic successes and failures Achievement Test – Second Edition (WIAT-II; and predicting achievement over time is a critically Wechsler 2005), and .74 with total achievement on important role an intelligence test can play. Having the Kaufman Test of Educational Achievement – IQ test questions that measure very similar content Second Edition Comprehensive Form (KTEA-II; 300 J.A. Naglieri

Kaufman and Kaufman 2004b ) (Kaufman and been advocated by Kaufman (1994 ) and others Kaufman 2004a , Tables 8-23 to 8-30; Kaufman (e.g., Sattler 1988 ) as a way to identify intellec- and Kaufman 2004b , Table 7-25). Taken as a tual strengths and/or weaknesses. Information whole, the average correlation is .69. about strengths and weaknesses is then used to The studies of the CAS, CAS2, KABC, and generate hypotheses that are integrated with KABC-II summarized here illustrate that a cogni- other information so that decisions can be made tive approach to understanding children’s intelli- regarding eligibility, diagnosis, and treatment. gence is strongly correlated with achievement test Despite the widespread use of this method, scores. Interestingly, these studies show that cogni- some have argued that subtest profi le analysis tive processes are as effective for prediction of aca- does not provide useful information beyond that demic performance as traditional IQ tests even which is obtained from the IQ scores (e.g., though the CAS, CAS2, KABC, and KABC-II do McDermott et al. 1990; Dombrowski and not include academically laden measures such as Watkins 2013 ). Naglieri ( 1999 ) proposed that vocabulary and arithmetic. This provides an advan- subtest analysis is problematic because of limi- tage for understanding achievement strengths and tations in subtest reliability and validity and fur- weaknesses for children who come from disadvan- ther suggested that what is needed is profi le taged environments as well as those who have had analysis based upon a sound theory of cognitive a history of academic failure. abilities, rather than individual subtest level analysis. Theoretically derived scales could be helpful if the ability test shows a specifi c pattern Are There Advantages to Second- for a specifi c group of exceptional students Generation Intelligence Tests? which in turn could have implications for understanding the cognitive characteristics of the Having shown that second-generation ability group, allow for guidance during the eligibility tests correlate as well with achievement test process (see Naglieri 2011 ), and guide interven- scores as traditional IQ test which contain aca- tions (Naglieri and Pickering 2010 ). demic content, the next question to consider is do Recently, Naglieri (2011 ) summarized reports these tests have other advantages? For example, found in the Wechsler Intelligence Scale for do these tests yield ability profi les that help Children – Fourth Edition (WISC-IV; Wechsler understand the role a cognitive weakness may 2003 ) technical manual, the Woodcock-Johnson play in academic failure. A second important III Tests of Cognitive Abilities (WJ-III; Woodcock issue is related to fair assessment of diverse pop- et al. 2001 ) from Wendling et al. (2009 ), and CAS ulations. More specifi cally, how do race differ- data from the technical manual and Naglieri, ences on traditional IQ compare to those found Otero, DeLauder, and Matto (2007 ). In the current for second-generation ability tests? And fi nally, chapter fi ndings for students with autism spectrum can second-generation ability tests inform disorders (ASD) from the WISC-IV (Wechsler instruction and academic intervention? Each of 2003 ), the CAS (from Naglieri and Otero 2011 ), these issues will be addressed next. the WJ-III (from Wendling et al. 2009 ), and the KABC-II (technical manual) were added. The fi ndings (see Fig. 20.1 ) must be considered with Do First- and Second-Generation recognition that the samples were not matched on Tests Detect Cognitive Problems That demographic variables across the various studies, Underlie Academic Failure? the accuracy of the diagnosis may not have been verifi ed, and some of the sample sizes were small. All intelligence tests give a full-scale score which Notwithstanding these limitations, the fi ndings is comprised of scales which in turn are com- provide important insights into the extent to which prised of subtests. The analysis of subtest and these various tests are likely to yield scale-level scale variation on tests such as the Wechsler profi les that are distinctive, theoretically logical, Scales is a method called profi le analysis that has and relevant to instruction. 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 301

105

100

75 ASD 70 SLD ADHD 65 Planning Attention Successive Planning/Gf Learning/Glr Simultaneous Knowledge/Gc Fluid Reasoning Sequential/Gsm Simultaneous/Gv Processing Speed Processing Speed Working Memory Auditory Processing Long-Term Retrieval Short-Term Memory Perceptual Reasoning Visual-Spatial Thinking Verbal Comprehension Comprehension-Knowledge WISC-IV WJ-III KABC-II CAS

Fig. 20.1 Scale proﬁ les on several measures of cognitive ability for students with SLD, ADHD, and autism

The results of the summary of scale profi les for evidence of a cognitive problem related to the WISC-IV, WJ-III, KABC-II, and CAS pro- ADHD, except for a low score on the planning vided in Fig. 20.2 suggest that some of these tests scale of the CAS. Diffi culty with planning (e.g., yield profi les that are more distinct than others executive function) for children with ADHD is across the three groups of exceptional children. consistent with Barkley’s view that ADHD is a The scores across all scales on the WJ-III for stu- failure of self-control (Barkley 1997 ) which has dents with specifi c reading decoding diffi culty been described as frontal lobe functioning (SLD) were all within the average range, and all (Goldberg 2009 ). of the KABC-II scores were in the 80s. The The results for individuals with autism spec- WISC-IV profi le was lowest for the Working trum disorders (ASD) show that processing speed Memory Scale. The CAS profi le showed variabil- scores on both the WISC-IV and the WJ-III were ity across the four PASS scales with a very low very low for individuals with ASD. This is simi- score of 82.9 on the successive scale. These fi nd- lar to the fi ndings for these two tests for ings are consistent with the view that students individuals with ADHD. The low processing with specifi c reading decoding failure also have speed scores provide little insight into the cogni- considerable diffi culty with tasks that involve tive characteristics of students with ASD and sequencing of information (Das et al. 2007 ). ADHD. Importantly, the low attention score on The intelligence test profi les for students with the CAS is consistent with the conceptualization ADHD showed that all the scores for the scales that individuals with ASD have been described as on the WISC-IV, WJ-III, and KABC-II were with having “diffi culties in disengaging and shifting the average range. None of these tests provided attention” (p. 214) (Klinger et al. 2009 ). 302 J.A. Naglieri

Fig. 20.2 Discrepancy/ consistency model for speciﬁ c learning disabilities

The fi ndings for those with ASD, like the results chological abilities is more appropriate for for those with SLD and ADHD, show that the diverse populations (Fagan 2000 ; Naglieri and only test that had distinctive profi les for these dif- Otero 2011 ). Fagan (2000 ) and Suzuki and ferent groups was the CAS. Valencia (1997 ) argued that measures of cognitive processes which do not rely on tests with language and quantitative content are more Are Race Differences for Second- appropriate for assessment of culturally and lin- Generation Tests the Same guistically diverse populations. Although there is as for Traditional IQ? considerable evidence for the validity of general intelligence as measured by traditional IQ tests The need for intelligence tests that are appropri- (see Jensen 1980 ), researchers have traditionally ate for diverse populations of children has found a mean difference of about 12–15 points become progressively more important as the between African-Americans and Whites on mea- characteristics of the US population have sures of IQ that include verbal, quantitative, and changed, and recent Federal law (e.g., IDEA nonverbal tests (Kaufman and Lichtenberger 2004) stipulates that assessments must be selected 1999 ). Results for second-generation intelligence and administered so as to be nondiscriminatory tests have been different. on a racial or cultural basis. It is, therefore, criti- The fi rst evidence of smaller race differences cal that any measures used for evaluation be eval- for second-generation ability test was reported in uated for test bias. This should include internal the original KABC Manual. For children aged evidence such as reliability, item diffi culty, and 2.5–12.5, without controlling for background factor structure (see Jensen 1980 ) as well as mean variables, Whites ( N =1,569) scored 7 points score differences. higher than African-Americans (N =807) and 3 Some researchers have suggested that intelli- points higher than Hispanics ( N = 160) on the gence conceptualized on the basis of neuropsy- global measure of mental processing (i.e., the 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 303 total test score). These differences are consider- Table 20.1 Mean score differences in standard scores by ably smaller than the differences of 16 points and race on traditional IQ and second-generation intelligence tests 11 points, respectively, reported for the WISC-R Full Scale IQ (Kaufman and Kaufman 1983 , Test Difference Tables 4.36 and 4.37; Kaufman et al. 2005 , Traditional Table 6.7). Similar fi ndings were reported by SB-IV (matched) 12.6 WISC-IV (normative sample) 11.5 Naglieri ( 1986) in a study of 172 fi fth-grade stu- WJ-III (normative sample) 10.9 dents (86 whites and 86 blacks matched on basic WISC-IV (matched) 10.0 demographic variables) who were administered Second generation the KABC and the WISC-R. The difference KABC (normative sample) 7.0 between the groups on WISC-R Full Scale was KABC (matched) 6.1 9.1 but the difference for the KABC was 6.0. KABC-2 (matched) 5.0 Results for the KABC-II (Kaufman and Kaufman CAS2 (normative sample) 6.3 2004) showed a similar reduction in race/ethnic CAS (demographic controls) 4.8 differences. When controlling for gender and CAS2 (demographic controls) 4.3 mother’s education, African-American children Notes: Stanford-Binet IV (SB-IV) from Wasserman at ages 3–18 years earned mean MPIs that were (2000); (Woodcock-Johnson III) WJ-III from Edwards only 5 points lower than the means for White and Oakland ( 2006); Kaufman Assessment Battery for Children (KABC) matched from Naglieri ( 1986 ); children (A. S. Kaufman and Kaufman 2004a , Kaufman Assessment Battery for Children – 2 from Tables 8.7 and 8.8; A. S. Kaufman et al. 2005 , (Lichtenberger et al. 2009); CAS from Naglieri, Rojahn, Table 6.7). Similar fi ndings have been reported Matto, and Aquilino ( 2005 ); Wechsler Intelligence Scale for the CAS. for Children – IV (WISC-IV) from O’Donnell (2009 ) Naglieri, Rojahn, Matto, and Aquilino ( 2005 ) compared PASS scores on the CAS for 298 African-American children and 1,691 White high correlation between the scores (.94). Results children. Controlling for key demographic vari- for the CAS2 Full Scale scores were reported in ables, regression analyses showed a CAS Full the test manual (Naglieri et al. 2014 ). For chil- Scale mean score difference of 4.8 points in favor dren and adolescents aged 5–18 years without of White children. Similarly, Naglieri, Rojahn, controlling for demographic variables, African- and Matto (2007 ) examined the utility of the Americans and non-African-Americans differed PASS theory with Hispanic children by compar- by 6.3 standard scores, and with controls for ing performance on the CAS of Hispanic and demographic characteristics, the difference was White children. The study showed that the two 4.5. Similarly, without controlling for demo- groups differed by 6.1 points using unmatched graphic differences, Hispanics and non- Hispanics samples, 5.1 with samples matched on basic differed on the CAS Full Scale scores by 4.5 demographic variables, and 4.8 points when points, and with controls for demographic char- demographics differences were statistically con- acteristics, the difference was 1.8. trolled. Naglieri, Otero, DeLauder, and Matto The importance of the fi ndings presented (2007 ) compared scores obtained on the CAS above for the CAS and KABC is best understood when administered in English and Spanish to within the context of differences found on tradi- bilingual children (N = 40) referred for reading tional intelligence tests. Table 20.1 provides a diffi culties. They found a 3.0-point difference summary of standard score differences by race for between the CAS Full Scale scores and these the Stanford-Binet IV (SB-IV; Roid 2003 ), scores were highly correlated (.96). Otero, Woodcock-Johnson Tests of Cognitive Abilities – Gonzales, and Naglieri (2012) replicated that Third Edition (WJ-III; Woodcock et al. 2001 ), the study with another group of students referred for WISC-IV (Wechsler 2003 ), the KABC and reading problems and found CAS Full Scale KABC-II (Kaufman and Kaufman 1983 , 2004), scores that differed by less than one point and a and the CAS (Naglieri and Das 1997 ) and CAS2 304 J.A. Naglieri

(Naglieri et al. 2014 ). The results for the WISC-IV effective ways of using the abilities a student has are reported by O’Donnell (1009), for the SB-IV and managing any limitations in ability. The stu- by Wasserman (2000), and the WJ-III results are dent is seen as an active participant who inter- from Edwards and Oakland ( 2006 ). The race dif- prets information that is received, relates it to preferences for the KABC normative sample were viously acquired facts, organizes and stores it for reported in that test’s manual (Kaufman and later use, develops ways of doing things, and Kaufman 1983 ), and the fi ndings for the KABC-II critically examines information. Because this were summarized by Lichenberger, Sotelo- approach puts emphasis on both the academic Dynega, and Kaufman (2009 ). Differences for the skills the child must learn as well as the cognitive CAS were reported by Naglieri, Rojahn, Matto, abilities the child uses in the act of learning, know- and Aquilino ( 2005 ) and in the test manual for the ing the cognitive ability profi le of a student is a CAS2 by Naglieri, Das, and Goldstein (2014 ). critical element in a complex process that leads to The results of research on race differences for effective teaching and learning (Naglieri 1999 ). the KABC-II and CAS illustrate that second- The relationships between cognitive abilities as generation tests, in contrast to traditional IQ tests, measured by the CAS and academic instruction provide a more equitable way to assess diverse have been reported in a series of research papers. populations of children. The fi ndings suggest that There are several resources for applying the as a group, traditional IQ tests showed differ- PASS theory to academic instruction and reme- ences in ability scores between the races that are diation. The PASS Remedial Program (PREP; about twice as large as that found for second- Das 1999 ) is an option as is the planning strategy generation tests. All of the traditional tests instruction, also known as the planning facilita- included in this table have verbal/quantitative/ tion method, described by Naglieri and Pickering nonverbal content, and two of these three types of ( 2003). Other resources include, for example, questions demand knowledge that is very similar Kirby and Williams’ ( 1991 ) Learning Problems: to that required by standardized achievement A Cognitive Approach, Cognitive Strategy tests (see Naglieri and Bornstein 2003 ). It is rea- Instruction That Really Improves Children’s sonable to conclude that the approach to concep- Academic Performance – Second Edition by tualizing and measuring intelligence taken by the Pressley and Woloshyn (1995 ), Helping Students authors of these second-generation ability tests Become Strategic Learners (Scheid 1993 ), and resulted in smaller race difference without a loss Naglieri and Pickering’s (2010 ) book Helping of prediction to achievement or in the case of the Children Learn: Intervention Handouts for Use CAS sensitivity to learning problems, both of in School and Home . The fi rst two methods are which are critical components of validity. based on empirical studies, while the remaining books contain cognitive approaches to academic interventions. The methods use structured and Do Second-Generation Tests Have directed instructions based on PREP or struc- Relevance to Academic Intervention? tured but not scripted planning strategy instruction. In order to provide more details and research One important purpose of assessment of ability is underlying strategy instruction and PREP, these to help decide how a student learns best and what two methods will be discussed in more detail. obstacles to learning may exist. Knowing the cognitive profi le of an individual student should Strategy Instruction. The connection between inform instruction. This means that in addition to planning from PASS and interventions to improve teaching knowledge and skills, tailored instruc- the use of strategies has been examined in a series tion should help children, for example, “to plan of studies. These investigations have involved and control, to think and inquire, to evaluate and both math and reading comprehension and have refl ect” (Scheid 1993 , p. 3). This kind of approach focused on the concept that children can be taught teaches children knowledge and skills as well as to be more strategic when they complete academic 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 305 tasks and that the facilitation of plans positively weakness improved only marginally. Similarly, impacts academic performance. The essential children with cognitive weaknesses in simultane- concept was based on the idea that teaching stu- ous, successive, and attention showed substan- dents to use strategies directly was not as advan- tially lower rates of improvement. The importance tageous as encouraging students to approach of this study was that the fi ve groups of children their work strategically. The method is designed responded very differently to the same interven- so that children discover the value of strategy use tion. Stated another way, the PASS processing without being specifi cally instructed to do so. scores were predictive of the children’s response The students are encouraged to examine the to this math intervention (Naglieri and Johnson demands of the task in a strategic and organized 2000 ). manner. Research on this intervention method The effects of planning strategy instruction on and its relationship to PASS scores on the CAS reading comprehension were reported by Haddad, has been carefully examined in a number of Garcia, Naglieri, Grimditch, McAndrews, and important research studies. Eubanks (2003 ). This study assessed whether an The fi rst two studies using planning strategy instruction designed to facilitate planning would instruction showed that children’s performance in have differential benefi ts on reading comprehen- math calculation improved substantially (Naglieri sion and if improvement was related to the PASS and Gottling 1995, 1997 ). The children in these processing scores of each child. The researchers two studies attended a special school for those used a sample of general education children with learning disabilities. Students completed sorted into groups based on their PASS scale pro- mathematics worksheets in sessions over about a fi les using the CAS. Even though the groups did two-month period. The method designed to teach not differ by CAS Full Scale scores or pretest planning was applied in individual 1 on 1 tutoring reading comprehension scores, children with a sessions (Naglieri and Gottling 1995 ) or in the planning weakness benefi ted substantially (effect classroom by the teacher (Naglieri and Gottling size of 1.4) from the instruction designed to 1997 ) two to three times per week in half hour encourage the use of strategies and plans. In con- blocks of time. Students were encouraged to rec- trast, children with no PASS weakness or a suc- ognize the need to plan and use strategies when cessive weakness did not benefi t as much (effect completing mathematic problems during the sizes of .52 and .06, respectively). These results intervention periods. The teachers provided further support previous research suggesting that probes that facilitated discussion and encouraged the PASS profi les are relevant to instruction. the students to consider various ways to be more Iseman and Naglieri ( 2011 ) examined the successful. More details about the method are effectiveness of the strategy instruction for stu- provided by Naglieri and Gottling (1995 , 1997 ) dents with LD and ADHD randomly assigned to and by Naglieri and Pickering (2010 ). an experimental group or a control group which The relationship between strategy instruction received standard math instruction. They found and the PASS profi les for children with learning large pre-post effect sizes for students in the disabilities and mild mental impairments was experimental group (0.85), but not the control also studied by Naglieri and Johnson (2000 ). The group (0.26) on classroom math worksheets, as purpose of their study was to determine if chil- well as standardized test score differences in dren with cognitive weaknesses in each of the Math Fluency (1.17 and .09, respectively) and four PASS processes, and children with no cogni- Numerical Operations (.40 and −.14, respective weaknesses, showed different rates of tively). One year later the experimental group improvement in math when given the same group continued to outperform the control group. These planning strategy instruction. The fi ndings from fi ndings strongly suggested that students with LD this study showed that children with a cognitive and ADHD in the experimental group evidenced weakness in planning improved considerably greater improvement in math worksheets, far over baseline rates, while those with no cognitive transfer to standardized tests of math, and at 306 J.A. Naglieri follow- up 1 year later than the control group. The reading decoding. PREP is also founded on the fi ndings also illustrate the effectiveness of premise that the transfer of principles is best strategy instruction especially for those with low facilitated through inductive, rather than deduc- planning scores on the CAS. tive, inference. The program is, therefore, struc- The results of these planning strategy instruc- tured so that tacitly acquired strategies are likely tion studies using academic tasks suggest that to be used in appropriate ways. For example, the changing the way aptitude is conceptualized tasks teach children to focus on the sequences of (e.g., as the PASS rather than traditional IQ) and information included in a variety of tasks, includ- measured (using the CAS) increases the proba- ing reading. bility that an aptitude-by-treatment interaction Support for PREP summarized elsewhere (ATIs) is detected. Past ATI research suffered (Naglieri 2011) has shown the effectiveness of from conceptualizations of aptitudes based on the the instructional method for children with read- general intelligence model which did not ade- ing decoding problems. Children who received quately differentiate cognitive abilities which are PREP in comparison to a regular reading pro- related to instruction. The traditional IQ approach gram improved signifi cantly on Word Attack and is very different from the PASS theory as mea- Word Identifi cation tests from the Woodcock sured by the CAS. The summary of studies pro- Reading Mastery Test – Revised. Learning dis- vided here are particularly different from previous abled children who were randomly assigned to a ATI research that found students with low gen- PREP training and a control group that received eral ability improve little, whereas those with regular classroom instruction showed signifi cant high general ability respond more to instruction. improvement in reading decoding of real and In contrast, children with a weakness in one of pseudowords. When PREP was compared to a the PASS processes (planning) benefi ted more meaning-based reading program using two care- from instruction compared to children who had fully matched groups of fi rst-grade children, the no weakness or a weakness in a different PASS results showed a signifi cant improvement in process. The results of these studies also suggest reading scores for the PREP group and the gain that the PASS profi les can help predict which in reading was greater than it was for the meaning- children will respond to the academic instruction based control group. Specifi c relevance to the and which will not. This offers an important children’s CAS profi les was also demonstrated opportunity for researchers and practitioners by the fact that those children with a higher level interested in the design of instruction as sug- of successive processing at the beginning of the gested by Naglieri and Pickering (2003 ). program benefi ted the most from the PREP PREP. PREP was developed as a cognitive instruction, but those with the most improvement remedial program based on the PASS theory in the meaning-based program had higher levels (Das et al. 1994 ). These researchers summa- of planning. Taken as a whole, these studies sup- rized research which showed that students could port the effectiveness of PREP in remediating be trained to use successive and simultaneous defi cient reading skills during the elementary processes more effi ciently, which resulted in an school years and the connection between the improvement in their performance on that pro- PASS theory and intervention. cess and transferred to specifi c reading tasks. PREP aims to improve the use of cognitive processing strategies (e.g., simultaneous and suc- Do Second-Generation Ability Tests cessive processes) that underlie reading. The Aid in Determination of a Specifi c tasks in the program teach children to focus Learning Disability? their attention on the sequential nature of many tasks, including reading which helps the chil- One of the greatest challenges to traditional IQ dren better utilize successive processing – a has been the identifi cation of a specifi c learning very important cognitive process needed in disability, defi ned in IDEA 2004 as follows: 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 307

Specifi c learning disability means a disorder in one as measured by the CAS2 is best suited because or more of the basic psychological processes it yields profi les for students with SLD, works involved in understanding or in using language, spoken or written, which may manifest itself in the well with minority students, and has intervention imperfect ability to listen, think, speak, read, write, implications (Haung et al. 2010 ; Naglieri and spell, or do mathematical calculations. Such term Otero 2011 ). There are three main components to includes such conditions as perceptual disabilities, eligibility determination based on this conceptu- brain injury, minimal brain dysfunction, dyslexia, and developmental aphasia. Such term does not alization initially presented by Naglieri ( 1999 ). include a learning problem that is primarily the First, the student has signifi cant intraindividual result of visual, hearing, or motor disabilities, of discrepancy among the PASS scales, and the low- mental retardation, of emotional disturbance, or of est PASS ability score is substantially below environmental, cultural, or economic disadvantage. average. Second, there is a discrepancy between Perhaps the most essential problem with using good PASS scores and weak achievement. Third, a traditional IQ test for this purpose is the fact there is a consistency between poor PASS scores that alignment of the test results to the Federal and academic defi cits (Naglieri 1999 , 2011 ) as regulations has been quite diffi cult. This is espe- illustrated in Fig. 20.2 . cially true because specifi c learning disability is An intraindividual discrepancy is examined defi ned as a disorder in basic psychology pro- by comparing a student’s four PASS scale stan- cesses, but traditional IQ tests were not devel- dard scores to determine if there exists a cogni- oped to measure basic psychological processes. tive weakness. The purpose of these analyses is Even though it is clear that the identifi cation of a to identify PASS cognitive processing strengths disorder in one or more of the basic psychologi- (scores that are signifi cantly greater than the stu- cal processes is essential for SLD eligibility dent’s mean score and fall above the normative determination (Fuchs and Young 2006 ; Hale et al. average) or weaknesses (scores that are signifi - 2006), practitioners have been constrained by the cantly lower than the student’s mean score and content of traditional IQ tests which do not align fall below the normative average). For example, with the defi nition. consider a student has standard scores of 114 It is logical that assessment of basic psycho- (planning), 116 (simultaneous), 109 (attention), logical processes as stipulated in IDEA 2004 be a and 94 (successive). The successive score is part of any comprehensive assessment designed 14.25 standard score points below the child’s to determine if a child has a specifi c learning dis- mean of 108.25 which is signifi cant but that score ability. Moreover, IDEA 2004 requires the use of of 94 is within the average range. Academic a variety of assessment tools that are technically achievement scores are similar to the successive sound and nondiscriminatory to gather functional, score (low portion of the average range). This developmental, and academic information when would not be considered evidence of a disorder in special education eligibility is being determined. one or more basic psychological processes and This information should be integrated with other academic failure. In contrast, a cognitive weak- important fi ndings about the child to ensure that a ness is found when, for example, a student has comprehensive evaluation is obtained. In short, standard scores of 102, 104, 97, and 82 for plan- documentation of a basic psychological process- ning, simultaneous, attention, and successive, ing disorder and academic failure is essential for respectively, in which case the successive score is SLD determination (Hale et al. 2006 ; Naglieri considered a weakness and there are comparable 1999, 2011), and this can be best accomplished academic achievement test scores. The succes- with second-generation ability tests. sive and academic weaknesses in contrast to Essential to the description of a specifi c learn- planning, simultaneous, attention, and academic ing disability (SLD) is the presence of a pattern scores that are average or higher would suggest of strengths and weakness in basic psychological the existence of a disorder in one or more of the processes and academic skills. Of all the possible basic psychological processes with academic tools described in this chapter, the PASS theory failure as described in IDEA. 308 J.A. Naglieri

The method described above is referred to as information alone (p. 462).” Importantly, when the discrepancy/consistency model (Naglieri Huang et al. (2010 ) studied PASS profi les on the 1999 ). This method is useful for the identifi cation CAS for large samples of students in regular edu- of specifi c learning disabilities because it ensures cation (N = 1,692) and those with specifi c learning a systematic examination of variability of both disabilities (N = 367), they found ten core PASS cognitive and academic achievement test scores. profi les for those in regular educational and eight Determining if the cognitive processing scores unique profi les from students with SLD. Huang differ signifi cantly is accomplished using the et al. concluded that “a student with a true LD has method originally proposed by Davis ( 1959 ) and a relatively high chance of being accurately iden- Silverstein (1982 ), popularized by Kaufman tifi ed when using profi les analysis on composite (1979 ), and modifi ed by Silverstein ( 1993 ). This [PASS] scores” (p. 28). They added that their so-called ipsative method determines when the “analysis has provided evidence for the use of the child’s scores are reliably different from the PASS theory and that it appears that it has suffi - child’s average score. This technique has been cient applications for diagnosis for students sus- applied to a number of tests including, for exam- pected of having a LD” (p. 28). ple, the WISC-IV (Naglieri and Paolitto 2005 ), the CAS (Naglieri and Das 1997 ), and the SB5 (Roid 2003 ). It is important to note that in the Section Summary discrepancy/consistency model described by Naglieri (1999 ), the ipsative approach is applied The topics covered thus far provide evidence that to the PASS scales which represent four neuro- second-generation ability tests should be consid- cognitive PASS constructs, not the subtests. This ered viable methods of evaluating children and changes the method from one that relies on a adolescents for three important reasons. First, the clinical interpretation of the meaning of subtest KABC and CAS correlate strongly with achieve- variability to analysis of scales that have been ment even though they do not have academic theoretically defi ned and have higher reliability content, which suggests they have excellent and validity. This distinction is important because validity. Second, the CAS and KABC yield small the criticisms of the ipsative method (McDermott differences between Black and White (CAS and et al. 1990 ) have centered around subtest, not KABC and their second editions) as well as scale-level analysis. Hispanic and White (CAS, KABC-II, and CAS2) Naglieri (1999 ) and Flanagan and Kaufman groups which provides evidence that these mea- ( 2004) stressed the importance of recognizing sures are appropriate for non-biased assessment. that because a low score relative to the child’s Third, the evidence presented shows that CAS mean could still be within the average range, add- scores reveal the weakness children with specifi c ing the requirement that the weakness in a pro- learning disability in reading decoding have is cessing test score is also well below average is different from that experienced by those with important. In a study of PASS profi les for the other types of SLD (Haung et al. 2010 ) as well as CAS standardization and validity samples, ADHD and autism (Fig. 20.3 ). Naglieri (2000 ) found that those students who had The remainder of this chapter will focus on a PASS weakness were likely to have signifi cantly one of the two second-generation ability tests. lower achievement scores and more likely to have Although the KABC and CAS both provide sub- been identifi ed as exceptional. That study was stantial advantages beyond traditional IQ, only described by Carroll ( 2000 ) as one which illus- the CAS has demonstrated specifi c PASS profi les trated what a more successful profi le method for students with disabilities—it yields the small- could be. Davison and Kuang ( 2000 ) suggested est differences by race/ethnicity, and there is a that “adding information about the absolute level history of research showing the relevance of of the lowest score improves identifi cation over PASS scores to academic instruction. For these what can be achieved using ipsative profi le pattern reasons the remainder of this chapter will provide 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 309

Fig. 20.3 Structure of the Cognitive Assessment System – Second Edition a more detailed discussion of how the four PASS needed (Naglieri et al. 2014 ). This includes control abilities can be measured by the CAS2. of actions and thoughts so that effi cient solutions to problems can be achieved. Planning provides the means to solve problems for which no method Operationalization of the PASS or solution is immediately apparent and may Neurocognitive Abilities involve retrieval of information as well as utilization of the other PASS abilities to process the The PASS theory was fi rst operationalized by information. Planning ability is also important Naglieri and Das in 1997 with the publication of when individuals refl ect on events, recognizing the CAS (and more recently the CAS2 (Naglieri what worked, and what did not work, and consid- et al. 2014 )). Some of the research about the scores ering better problem solving in the future. The obtained from the tests developed to measure PASS frontal lobes of the brain are directly involved in was described earlier in this chapter. Although the planning ability (Naglieri and Otero 2011 ). PASS theory is more fully described elsewhere The essence of tasks that measure planning is (Naglieri 1999 ; Naglieri and Das 1997; Naglieri that the student must solve novel problems for and Otero 2011 ; Naglieri et al. 2014 ) and in Otero’s which there is no previously acquired strategy chapter in this book, the remainder of this chapter and there should be minimal constraints placed includes a description of how each of the four on the way the student completes the task. The PASS abilities are measured (see Fig. 20.3 ) in the score a planning test yields should refl ect effi - CAS2 and how this relates to traditional IQ. ciency, measured by how a student went about completing the tests and how effective the solutions were. The following tasks are used in Planning the CAS2 to evaluate planning ability:

In PASS theory, planning is a neurocognitive abil- Planned Codes. This subtest contains four ity used to determine and apply strategies to solve items, each with its own set of codes and particu- problems and self-monitor and self-correct as lar arrangements of rows and columns. A legend 310 J.A. Naglieri at the top of each page shows which letters cor- Attention respond to which codes (e.g., A, B, C, D with OX, XX, OO, XO, respectively). Just below the Attention is a neurocognitive ability used to legend are seven rows and eight columns of let- selectively focus on a particular stimulus while ters without the codes. Children write the corre- inhibiting responses to competing stimuli pre- sponding codes in empty boxes beneath each of sented over time (Naglieri et al. 2014 ). Attention the letters. The items differ in the correspondence is a basic component of intelligent behavior of letters to codes and the position of the letters involving allocation of resources and effort. on the page. Students have 60 s per item to com- Arousal, attention, effort, and capacity are con- plete as many empty code boxes as possible. cepts that have a complex relationship and importance for understanding behavior. Luria stated Planned Connections. The Planned Connec- that optimal conditions of arousal are needed tions subtest requires the student to connect num- before the more complex forms of attention bers in sequence that appears in a quasi-random involving “selective recognition of a particular order (e.g., 1–2–3, etc.). For the more complex stimulus and inhibition of responses to irrelevant items, the child connects numbers and letters in stimuli” (Luria 1973 , p. 271) can occur. Attention sequential order, alternating between numbers is conceptualized as a mental activity that pro- and letters (e.g., 1–A–2–B, etc.). Any errors made vides focused, selective cognitive activity over by the child are corrected as they progresses time and resistance to distraction. The process is through the task. The items are constructed so involved when a person must demonstrate that children never complete a sequence by cross- focused, selective, sustained, and effortful activ- ing one line over the other. This provides a means ity. The longer the attention needed, the more the of reducing the areas to be searched when look- activity necessitates vigilance. Intentions and ing for the next number or letter. The subtest goals mandated by the planning process control score is based on the total amount of time used to attention, whereas knowledge and skills play an complete the task. integral part in the process as well. Tasks that measure attention include target Planned Number Matching. The student’s and nontarget stimuli that are multidimensional task is to fi nd and underline two numbers that are with the requirement that the person has to iden- the same in each row. Each item is composed of tify one aspect of the target (e.g., the color blue) eight rows of numbers, with six numbers per and resist responding to distractions (e.g., a word row. Two of the six numbers in each row are the red written in blue ink) as in the Stoop test (Lezak same. The length of numbers differs on the vari- 1995). This kind of a task requires selective focus ous rows. Numbers increase in digit length from of attention over time, an ability that is necessary one digit on the fi rst row of item 1 to seven digits for learning to take place. The following tasks are on the eighth row of item 4. There are four rows used in the CAS2 to evaluate attention: for each digit length and a total of four pages of numbers. Children ages 5–7 are provided an Expressive Attention. The Expressive Attention example followed by two test items. Ages 8–18 subtest consists of two age-related sets of three are provided an example and two test items. items. Students ages 5–7 years are presented with Each row of numbers was carefully developed to three items consisting of seven rows that each maximize benefi ts of strategy usage in the iden- contain six pictures of common animals, with tifi cation of correct matches. This approach each picture depicted as either big (1 in. by 1 in.) resulted in items with some rows that contain or small (1/2 in. by 1/2 in.). In each of three numbers that start with unique numbers, some items, the student is required to identify whether rows that include numbers with similar digit the animal depicted is big or small in real life, strings, and some rows that contain numbers that ignoring the relative size of the picture on the end with similar numbers. page. In item 1, the pictures are all the same size. 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 311

In item 2, the pictures are sized appropriately whole. This ability is involved in visual-spatial (i.e., big animals are depicted with big pictures, tasks as well as those language activities that and small animals are depicted with small pic- require comprehensive of grammatical structures). In item 3, the realistic size of the animal tures. The spatial aspect of simultaneous ability often differs from its printed size. Students ages involves both the perception of stimuli as a group 8–18 years are presented with three items con- or whole and the formation of visual images. The sisting of eight rows of fi ve words each. In item 1, grammatical dimension of simultaneous process- students are asked to read four black-and- white ing allows for the integration of words into ideas color words (blue, yellow, green, and red) that are through the comprehension of word relation- presented in random order. In item 2, students are ships, prepositions, and infl ections, so the person asked to name the colors of four colored rectan- can obtain meaning. gles (printed in blue, yellow, green, and red) that Tasks designed to measure simultaneous pro- are presented in random order. In item 3, the four cessing often have visual-spatial content. One color words are printed in a different-color ink well-known measure of simultaneous processing than the color-word name and are presented in is progressive matrices. Traditional intelligence random order. In this item, students are required tests often include subtests that use the progres- to name the color of the ink in which the word is sive matrix format, as do many nonverbal intelli- printed rather than read the word. gence tests such as the Naglieri Nonverbal Ability Test (Naglieri 2011 ). These tests are often cate- Number Detection. Each Number Detection gorized as perceptual reasoning or nonverbal, but item presents the student with a page of approxi- from PASS, matrices measure simultaneous abil- mately 200 numbers. Students are required to ity. This ability can also be measured using ver- underline specifi c numbers (ages 5–7 years) or bal content which requires comprehension of the specifi c numbers in a particular font (ages 8–18 grammatical components of language such as years) on a page with many distractors. There are comprehension of word relationships and under- four pages of numbers, each of which is scored standing of prepositions and infl ections (Naglieri for the number correct, number of false detec- 1999 ). The Verbal-Spatial Relations subtest on tions, and time. the CAS is an example of this type of a subtest (Naglieri et al. 2014 ). This arrangement of sub- Receptive Attention. The Receptive Attention tests allows for measurement of simultaneous subtest consists of four item sets, each containing ability across verbal and nonverbal contents. The 60 picture pairs (ages 5–7 years) or 180 letter tests used to evaluate simultaneous neurocogni- pairs (8–18 years). Both versions require the stu- tive ability on the CAS2 are as follows: dent to underline pairs of objects or letters that either are identical in appearance or are the same Matrices. Matrices is a multiple-choice subtest from a lexical perspective (i.e., they have the that utilizes shapes and geometric elements that same name). There are four pages of numbers, are interrelated through spatial or logical organi- each of which is scored for the number correct, zation. Students are required to analyze the rela- number of false detections, and time. tionship among the parts of the item and solve for the missing part by choosing the best of fi ve options. The raw score is the total number of Simultaneous items correctly answered.

Simultaneous is a neurocognitive ability used to Verbal-Spatial Relations. Verbal-Spatial Relations integrate separate stimuli into a single whole or is a multiple-choice subtest in which each item con- interrelated group (Naglieri et al. 2014 ). The sists of six drawings and a printed question at the essence of simultaneous processing is that sepa- bottom of each page. The examiner reads the ques- rate elements must be combined into a conceptual tion aloud, and the child is required to select the 312 J.A. Naglieri option that matches the verbal description. The items using the correct series of words (Sentence require the evaluation of logical grammatical rela- Repetition) as well as comprehension of sen- tionships (e.g., “which picture shows a ball in a bas- tences that are understood only by appreciating ket under a table?”), which demands simultaneous the sequence of words (Sentence Questions). processing with verbal content. The raw score is the Additionally, CAS2 has a visual digit span test, total number of items correctly answered. allowing for measurement of successive processing across auditory and visual modalities. The Figure Memory. For each Figure Memory item, tests used in the CAS2 to measure successive the examiner presents the student with a two- or processing ability are as follows: three- dimensional geometric fi gure for 5 s. The picture is then removed, and the student is pre- Word Series. The Word Series subtest utilizes sented with a response page that contains the nine single-syllable, high-frequency words: original fi gure embedded in a large, more com- book, car, cow, dog, girl, key, man, shoe, and plex geometric pattern. The student is required to wall. The examiner reads aloud a series of two to trace the original fi gure with a red pencil in the nine of these words at the rate of one word per Figure Memory Response Form. The raw score is second. The student is required to repeat the the total number of items correctly answered. words in the same order as stated by the examiner. The raw score is the total number of items correctly answered. Successive Sentence Repetition. The Sentence Repetition Successive is a neurocognitive ability used to subtest (administered only to ages 5–7 years) work with information that is arranged in a spe- requires the student to repeat syntactically cor- cifi c serial order where each part follows the rect sentences containing little meaning, such as other in a strictly defi ned order (Naglieri et al. “The blue is yellowing.” The raw score is the 2014). Successive processing is involved in the total number of items correctly answered. perception of stimuli in sequence as well as the formation of sounds and movements into a spe- Sentence Questions. The Sentence Questions cifi c order. This type of ability is necessary for subtest (administered only to ages 8–18 years) the recall of information in order as well as pho- requires the student to listen to sentences that are nological analysis and the syntax of language syntactically correct but contain little meaning (Das et al. 1994 ). Defi cits with successive pro- and answer questions about the sentences. For cessing are also associated with early reading example, the student is read the sentence “The problems in young children, as it requires a child blue is yellowing” and then asked the following to learn sounds in a sequential order. question: “Who is yellowing?” The raw score is Tasks used to measure successive processing the total number of items correctly answered. include digit span forward (as well as the recall of numbers, words, or hand movements) which is Visual Digit Span. Visual Digit Span subtest found on many tests of ability. These tests are requires the student to recall a series of numbers sometimes described as measures of working in the order in which they were shown using the memory or sequential processing (a concept very Stimulus Book. Each item that is 2–5 digits in close to successive processing in PASS theory). length is exposed for the same number of seconds Sometimes a backwards version is included as there are digits. Items with six digits or more which involves successive as well as planning are all exposed for a maximum of 5 s. The raw processing abilities (Schofi eld and Ashman score is the total number of items correctly 1987 ). The successive tasks included in the CAS answered. and CAS2 provide a way to measure this ability The CAS2 subtests described above can be using tests that demand repeating a sentence combined into an 8-subtest Core Battery or a 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 313

12-subtest Extended Battery to yield four scores following the PASS theory, planning, attention, simultaneous, and successive, and a total score called the Full Scale. The subtests are all individually administered tests designed explicitly to yield scores to evaluate the four PASS neurocognitive abilities for children and adolescents aged 5 years 0 months through 18 years 11 months. It was normed on a representative sample of 1,342 Fig. 20.4 Intelligence achievement continuum students. The test manual provides a complete summary of reliability and validity of the CAS2 as well as interpretive and intervention informa- that such tests are often indistinguishable from tion. See Naglieri et al. (2014 ) for more details. measures of achievement , users of traditional IQ tests have ignored this problem and compounded the issue by calling such tests measures of verbal Closing Thoughts intelligence . Fourth, about 80 years after the birth of tradi- The purpose of this chapter was to organize 100 tional IQ, a second generation of intelligence years of progress in the area of IQ tests, twentieth- tests appeared. The fi rst was the KABC (Kaufman century traditional ideas about intelligence, and a and Kaufman 1983 ) and second was the CAS second generation of intelligence tests. The (Naglieri and Das 1997 ). These tests were essence of the discussion has been about the tools designed with a conceptualization of intelligence and concepts we have used in this most important (KABC) or a specifi c theory of intelligence fi eld of applied psychology. I have argued that (CAS); and importantly, they did not include the there are several important issues that need to be verbal and arithmetic test items found in tradi- recognized and will be reiterated here. tional IQ. On a continuum from pure intelligence First, traditional IQ began July 20, 1917, with to pure achievement, the second-generation tests the development of the verbal (quantitative) and were clearly distinct from tests of achievement nonverbal IQ test format. This format has domi- and traditional IQ (see Fig. 20.4). Traditional IQ nated the IQ testing industry since that time and tests share some overlap with second-generation has been used in all individual- and group- IQ tests (mainly the nonverbal portion of tradi- administered IQ tests. tional IQ tests and simultaneous scales of the Second, the tests developed for the US Army CAS and KABC). The verbal and quantitative were designed to test many recruits in the short- portions of traditional IQ share overlap with est amount of time and with the least amount of achievement tests. The academic content of tradi- effort needed for scoring. There was no theory of tional IQ tests poses considerable problem for intelligence that guided the selection or develop- test validity and assessment of diverse popula- ment the Army Alpha and Beta tests. These tests tions as well as those with learning diffi culties. have been accepted as measures of intelligence Fifth, research has clearly supported second- and in fact the IQ score has become synonymous generation intelligence tests over traditional with the term intelligence. IQ. Newer tests offer several advantages including Third, traditional IQ tests include questions (a) a theory that can be used to create scales on a that are very similar to tests found in achievement test that represent a psychological construct, (b) tests, especially, for example, vocabulary, word greater fairness to minorities and to those with lim- analogies, and math word problems. The role of ited academic skills, (c) scores that represent differ- knowledge needed to answer these types of ques- ent abilities according to the theory upon which the tions was recognized as undesirable by the origi- test was developed, (d) greater ability to identify nal authors of the Army Alpha. Despite the fact special populations of individuals with intellectual 314 J.A. Naglieri disorders related to behavioral or academic dis- Flanagan, D. P., & Kaufman, A. S. (2004). Essentials of abilities, and (e) enhanced ability to link second- WISC-IV assessment . Hoboken: Wiley. Fuchs, D., & Young, C. L. (2006). On the irrelevance of generation intelligence test scores to interventions. intelligence in predicting responsiveness to reading The challenge faced by second-generation intel- instruction. Exceptional Children, 73 , 8–30. ligence tests, despite their clear advantages over Goldberg, E. (2009). The new executive brain: Frontal traditional IQ, is inertia. Traditional IQ has 100 lobes in a complex world. New York: Oxford University Press. years of use and acceptance and countless numbers Haddad, F. A., Garcia, Y. E., Naglieri, J. A., Grimditch, of research studies and books written about them M., McAndrews, A., & Eubanks, J. (2003). Planning and their interpretation. But as Neil deGrasse facilitation and reading comprehension: Instructional Tyson, author of the new guide to the Cosmos, relevance of the PASS theory. Journal of Psychoeducational Assessment, 21 , 282–289. recently commented (2014) on the value of tradi- Hale, J. B., Kaufman, A., Naglieri, J. A., & Kavale, K. A. tional wisdom “In practically every idea we have as (2006). Implementation of IDEA: Integrating response humans, the older version of it is not better than the to intervention and cognitive assessment methods. new version” (p. 80). It is time for the fi eld of intel- Psychology in the Schools, 43 (7), 753–770. Hart, B., & Risley, T. R. (1995). Meaningful differences in ligence testing to embrace new ideas of what intel- the everyday experience of young American Children . ligence may be and how best to measure it. Baltimore: Brookes. Haung, L. V., Bardos, A. N., & D’Amato, R. C. (2010). Identifying students with learning disabilities: Composite profi le analysis using the cognitive assess- References ment system. Journal of Psychoeducational Assessment, 28 , 19–30. IDEA. (2004). Individuals with Disabilities Education Barkley, R. A. (1997). ADHD and the nature of self- Improvement Act of 2004 (P.L. 108–446). control . New York: Guilford. Iseman, J., & Naglieri, J. A. (2011). A cognitive strategy Boake, C. (2002). From the Binet-Simon to the Wechsler- instruction to improve math calculation for children Bellevue: Tracing the history of intelligence testing. with ADHD: A randomized controlled study. Journal Journal of Clinical and Experimental Neuropsychology, of Learning Disabilities, 44 , 184–195. 24 , 383–405. Jensen, A. R. (1980). Bias in mental testing . New York: Carroll, J. B. (2000). Commentary on profi le analysis. Free Press. School Psychology Quarterly, 15 , 449–456. Kaufman, A. S. (1979). Intelligent testing with the Das, J. P. (1999). PASS reading enhancement program . WISC-R . New York: Wiley. Deal: Sarka Educational Resources. Kaufman, A. S., & Kaufman, N. L. (1983). Kaufman Das, J. P., Naglieri, J. A., & Kirby, J. R. (1994). Assessment assessment battery for children . Circle Pines: of cognitive processes . Boston: Allyn & Bacon. American Guidance. Das, J. P., Janzen, T., & Georgiou, G. K. (2007). Correlates Kaufman, A. S. (1994). Intelligent testing with the WISC- of Canadian native children’s reading performance: III. New York: Wiley. From cognitive styles to cognitive processes. Journal Kaufman, A. S., & Kaufman, N. L. (2004). Kaufman of School Psychology, 45 , 589–602. assessment battery for children second edition. Circle Davis, F. B. (1959). Interpretation of differences among Pines, MN: American Guidance. averages and individual test scores. Journal of Kaufman, A. S., & Kaufman, N. L. (2004a). Kaufman Educational Psychology, 50 , 162–170. assessment battery for children (2nd ed.). Circle Pines: Davison, M. L., & Kuang, H. (2000). Profi le patterns: American Guidance Service. Research and professional interpretation. School Kaufman, A. S., & Kaufman, N. L. (2004b). Kaufman test Psychology Quarterly, 15 , 457–464. of educational achievement comprehensive form (2nd Dombrowski, S. C., & Watkins, M. W. (2013). Exploratory ed.). Circle Pines: American Guidance Service. and higher order factor analysis of the WJ-III full test Kaufman, A. S., & Lichtenberger, E. O. (1999). Essentials battery: A school-aged analysis. Psychological of WAIS-III assessment . New York: Wiley. Assessment 2013, 25 (2), 442–455. Kaufman, A. S., Lichtenberger, E. O., Fletcher-Janzen, E., Edwards, O. W., & Oakland, T. D. (2006). Factorial & Kaufman, N. L. (2005). Essentials of KABC-II invariance of Woodcock-Johnson III scores for African assessment . New York: Wiley. Americans and Caucasians Americans. Journal of Kirby, J. R., & Williams, N. H. (1991). Learning prob- Psychoeducational Assessment, 24 , 358–366. lems: A cognitive approach . Toronto: Kagan and Woo. Fagan, J. R. (2000). A theory of intelligence as process- Klinger, L., O’Kelly, S. E., & Mussey, J. L. (2009). ing: Implications for society. Psychology, Public Assessment of intellectual functioning in Autism Policy, and Law, 6 , 168–179. spectrum disorders. In S. Goldstein, J. A. Naglieri, & 20 Hundred Years of Intelligence Testing: Moving from Traditional IQ… 315

S. Ozonoff (Eds.), The assessment of autism spectrum and ADHD in adolescence and adulthood (2nd ed., disorders (pp. 209–252). New York: Guilford. pp. 137–160). New York: Wiley. Lichtenberger, E. O., Sotelo-Dynega, M., & Kaufman, A. Naglieri, J. A., & Gottling, S. H. (1995). A cognitive edu- S. (2009). The Kaufman assessment battery for chil- cation approach to math instruction for the learning dren – second edition. In J. A. Naglieri & S. Goldstein disabled: An individual study. Psychological Reports, (Eds.), A practitioner’s guide to assessment of intelli- 76 , 1343–1354. gence and achievement . New York: Wiley. Naglieri, J. A., & Gottling, S. H. (1997). Mathematics Luria, A. R. (1963). Restoration of function after brain instruction and PASS cognitive processes: An inter- injury . New York: Pergamon Press. vention study. Journal of Learning Disabilities, 30 , Luria, A. R. (1966/1980). Higher cortical functions in 513–520. man (2nd ed., Revised and Expanded). New York: Naglieri, J. A., & Johnson, D. (2000). Effectiveness of a Basic Books. cognitive strategy intervention to improve math calcu- Luria, A. R. (1973). The working brain: An introduction lation based on the PASS theory. Journal of Learning to neuropsychology . New York: Basic Books. Disabilities, 33 , 591–597. Markwardt, F. C. (1997). Peabody individual achievement Naglieri, J. A., & Kaufman, A. S. (2008). IDEIA 2004 and test-revised (PIAT-R). San Antonio: The Psychological specifi c learning disabilities: What role does intelligence Corporation. play? In E. Grigorenko (Ed.), Educating individuals with Matarazzo, J. (1972). Wechsler’s measurement and disabilities: IDEIA 2004 and beyond (pp. 165–195). appraisal of adult intelligence (5th ed.). Baltimore: New York: Springer Publishing Company. Williams & Wilkins. Naglieri, J. A., & Otero, T. (2011). Cognitive assessment McDermott, P. A., Fantuzzo, J. W., & Glutting, J. J. system: Redefi ning intelligence from a neuropsycho- (1990). Just say no to subtest analysis: A critique on logical perspective. In A. Davis (Ed.), Handbook of Wechsler theory and practice. Journal of pediatric neuropsychology (pp. 320–333). New York: Psychoeducational Assessment, 8 (3), 290–302. Springer. Naglieri, J. A. (1986). WISC-R and K-ABC comparison Naglieri, J. A., & Paolitto, A. W. (2005). Ipsative com- for matched samples of Black and White children. parisons of WISC-IV index scores. Applied Journal of School Psychology, 24 , 81–88. Neuropsychology, 12 , 208–211. Naglieri, J. A. (1999). Essentials of CAS assessment . Naglieri, J. A., & Pickering, E. B. (2003). Helping chil- New York: Wiley. dren learn: Intervention handouts for use at school Naglieri, J. A. (2000). Can profi le analysis of ability test and home . Baltimore: Brookes Publishing. scores work? An illustration using the PASS theory Naglieri, J. A., & Pickering, E. (2010). Helping children and CAS with an unselected cohort. School Psychology learn: Intervention handouts for use in school and at Quarterly, 15 , 419–433. home (2nd ed.). Baltimore: Brookes. Naglieri, J. A. (2003). Fair assessment of gifted minority Naglieri, J. A., & Rojahn, J. R. (2004). Validity of the children using the NNAT. Gifted Education Press PASS theory and CAS: Correlations with achieve- Quarterly, 17 , 2–7. ment. Journal of Educational Psychology, Naglieri, J. A., & Conway, C. (2009). The cognitive 96 ,174–181. assessment system. In J. A. Naglieri & S. Goldstein Naglieri, J. A., Rojahn, J. R., Matto, H. C., & Aquilino, (Eds.), A practioner’s guide to assessment of intel- S. A. (2005). Black white differences in intelligence: ligence and achievement (pp. 3–10). New York: A study of the PASS theory and cognitive assessment Wiley. system. Journal of Psychoeducational Assessment, Naglieri, J. A. (2011). The discrepancy/consistency 23 ,146–160. approach to SLD identifi cation using the PASS theory. Naglieri, J. A., Goldstein, S., Delauder, B. Y., & In D. P. Flanagan & V. C. Alfonso (Eds.), Essentials of Schwebach, A. (2006a). WISC-III and CAS: Which specifi c learning disability identifi cation (pp. 145– correlates higher with achievement for a clinical sam- 172). Hoboken: Wiley. ple? School Psychology Quarterly, 21 , 62–76. Naglieri, J. A., & Bornstein, B. T. (2003). Intelligence and Naglieri, J. A., Rojahn, J., & Matto, H. (2007). Hispanic achievement: Just how correlated are they? Journal of and non-Hispanic children’s performance on PASS Psychoeducational Assessment, 21 , 244–260. cognitive processes and achievement. Intelligence, 35 , Naglieri, J. A., & Das, J. P. (1997). Cognitive assessment 568–579. system . Itasca: Riverside Publishing Company. Naglieri, J. A., Otero, T., DeLauder, B., & Matto, H. Naglieri, J. A., & Das, J. P. (2001). The Das-Naglieri (2007). Bilingual Hispanic children’s performance on Cognitive Assessment System in theory and practice. the English and Spanish versions of the cognitive In J. Andrews (Ed.), Ability, achievement, and behav- assessment system. School Psychology Quarterly, 22 , ior assessment: A practical handbook (pp. 36–68). 432–448. Naglieri, J. A., & Goldstein, S. (2011). Assessment of Naglieri, J. A., Das, J. P., & Goldstein, S. (2014). Cognitive cognitive and neuropsychological processes. In assessment system (2nd ed.). Austin: Pro-Ed. S. Goldstein, J. A. Naglieri, & M. DeVries (Eds.), O’Donnell, L. (2009). The Wechsler intelligence scale for Understanding and managing learning disabilities children. In J. A. Naglieri & S. Goldstein (Eds.), A 316 J.A. Naglieri

practitioner’s guide to assessment of intelligence and Suzuki, L. A., & Valencia, R. R. (1997). Race-ethnicity achievement (4th ed., pp. 153–190). New York: Wiley. and measured intelligence. American Psychologist, Otis, A. S. (1918). An absolute point scale for the group 52 , 1103–1114. measurement of intelligence. https://ia600401.us. Wechsler, D. (1939). Wechsler-Bellevue intelligence archive.org/28/items/cu31924073896650/ scale . New York: Psychological Corporation. cu31924073896650.pdf . Reprinted from the Journal Wechsler, D. (1991). Wechsler intelligence scale for chil- of Educational Psychology, IX. dren (3rd ed.). San Antonio: The Psychological Otis, A. S., & Lennon, R. T. (1979). Otis-Lennon school Corporation. ability test . New York: Psychological Corporation. Wechsler, D. (2003). Wechsler intelligence scale for chil- Pintner, R. (1923). Intelligence testing. New York: Henry dren (4th ed.). San Antonio: The Psychological Holt. Corporation. Pressley, M. P., & Woloshyn, V. (1995). Cognitive strategy Wechsler, D. (2005). Wechsler individual achievement test instruction that really improves children’s academic (WIAT II) (2nd ed.). London: The Psychological Corp. performance (2nd ed.). Cambridge, MA: Brookline. Wechsler, D. (2014). Wechsler intelligence scale for chil- Prifi tera, A., & Saklofske, D. (1998). WISC-III clinical dren (5th ed.). San Antonio: The Psychological use and interpretation: Scientist-Practitioner perspec- Corporation. tives . New York: Academic. Weiss, L. G., Saklofske, D. H., Prifi tera, A., & Holdnack, Ramirez, R. R., & de la Cruz, G. P. (2002). The Hispanic J. A. (2006). WISC-IV advanced clinical interpreta- population in the United States: March 2002 (Current tion . New York: Elsevier. population reports, P20-545). Washington, DC: Woodcock, R. W., & Johnson, M. B. (1989). Woodcock- U.S. Census Bureau. Johnson revised tests of cognitive ability: Standard Roid, G. (2003). Stanford-Binet (5th ed.). Itasca: and supplemental batteries . Itasca: Riverside Riverside. Publishing. Scheid, K. (1993). Helping students become strategic Wendling, B. J., Mather, N., & Shrank, F. A. (2009). learners . Cambridge, MA: Brookline. Woodcock-Johnson III tests of cognitive abilities. In Schofi eld, N. J., & Ashman, A. F. (1987). The cognitive J. A. Naglieri & S. Goldstein (Eds.), A practitioner’s processing of gifted, high average, and low average guide to assessment of intelligence and achievement ability students. British Journal of Educational (pp. 191–232). New York: Wiley. Psychology, 57 , 9–20. Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Silverstein, A. B. (1982). Pattern analysis as simultaneous Woodcock-Johnson III tests of cognitive abilities . statistical inference. Journal of Consulting and Itasca: Riverside Publishing Company. Clinical Psychology, 50 , 234–240. Yerkes, R. M. (1921). Memoirs of the National Academy Silverstein, A. B. (1993). Type I, Type II, and other types of Sciences Volume XV . Washigton: Government of errors in pattern analysis. Psychological Assessment, Printing Offi ce. 5 , 72–74. Yoakum, C. S., & Yerkes, R. M. (1920). Army mental Sattler, J. M. (1988). Assessment of children (3rd ed.). San tests . New York: Henry Holt and Company. Diego: Jerome M. Sattler. The Relationship Between Theories of Intelligence 21 and Intelligence Tests

W. Joel Schneider and Dawn P. Flanagan

In the fi eld of cognitive ability assessment, the In truth, the “waves” overlapped by many years, separate spheres of theory, research, and prac- and the practices of earlier waves never died out. tice are not that separate, evolving in chaotic First-wave practices are alive and well in many fi ts of individuation and rapprochement. Their applications of intelligence tests, as are second- relationship is complicated: Research can be and third-wave practices. Each wave describes a particularly cruel to theory, theory makes kind of central tendency of each era, but there are impossible demands of practice, and practice many outliers we could point to as exceptions to can go for long stretches acting as if research is the trends. We borrow extensively from the his- not even in the room. Yet, they are family and torical analysis and vocabulary of Kamphaus and they need each other. Research surprises the- colleagues, but our parsing of history differs ory with thoughtful gifts, theory gives sound slightly from theirs on several minor points. advice when practice is troubled, and practice helps research to get out of the lab and meet people. Their story is too long and complex to Wave 1: General Ability (1904–Present) tell in full, but there are helpful ways of making sense of their relationships. After numerous historical anticipations from Kamphaus, Winsor, Rowe, and Kim ( 2012 ) Galton, J. M. Cattell, and many others, British have presented the history of intelligence test researcher Charles Spearman ( 1904 ) inaugurated interpretation as a series of waves of new the era of scientifi c models of intelligence. practices. This is, of course, only a metaphor, but Almost simultaneously, French scholars Alfred it communicates historical truth reasonably well. Binet and Théodore Simon (1905 ) published the The metaphor is possibly misleading because it fi rst well-constructed, reasonably normed, and suggests that each wave was discrete and com- practical method of identifying children who pletely displaced the effects of the previous wave. were in need of special education services because of low intelligence. W. J. Schneider (*) Binet ( 1905 ) believed that Spearman’s theo- Department of Psychology , Illinois State University , ries were overly simplistic, and Spearman ( 1930 ) Campus Box 4620 , Normal , IL 61790-4620 , USA considered Binet’s tests to be rather crude. e-mail: [email protected] Nevertheless, despite this mutual criticism, D. P. Flanagan , Ph.D. Spearman’s ideas and Binet’s tests aligned well Department of Psychology , St. John’s University , Queens , NY , USA enough to launch a movement. Because of the e-mail: fl [email protected] efforts of IQ evangelists such as Cyril Burt,

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 317 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_21, © Springer Science+Business Media New York 2015 318 W.J. Schneider and D.P. Flanagan

Godfrey Thomson, Henry Goddard, and Louis Qualifying Test (a measure of literacy and numer- Terman, Binet’s tests were translated, refi ned, acy strongly correlated with IQ) is allowed to join and normed in countries all over the world. the U.S. military. The U.S. military also uses a For most of the twentieth century, the primary broader collection of aptitude tests (the Armed focus of cognitive ability testing was to deter- Services Vocational Aptitude Battery) to facilitate mine a person’s level of general intelligence. initial placement of recruits and to qualify appli- Researchers devoted an extraordinary amount of cants for specialized training. effort exploring diverse theoretical and practical Organizational psychology has a long tradi- questions about general intelligence and its cor- tion of using IQ tests for personnel selection. The relates. This literature is now truly enormous, primary emphasis on general intelligence in per- with hundreds of thousands of studies making sonnel selection persists to this day, largely use of IQ tests. In contrast with some of the deli- because it has been diffi cult to demonstrate the cate and fi ckle fi ndings that characterize some differential validity of specifi c abilities in job subdomains of psychology, the archives of IQ performance (Schmidt & Hunter 2004 ). research are replete with robust results that reli- Commercially available tests of general ability ably replicate. High IQ is not only predictive of are used by government agencies and large fi rms degrees and grades but also wealth and health to make hiring and promotion decisions. These (Gottfredson 1997 ). tests are still used in a manner that is largely con- Because of the early successes of IQ research sistent with “fi rst-wave” practices. (and other, less laudable motivations), many In educational settings, intelligence tests have institutions began to use them to make deci- been used to identify the extremes of general sions about individuals. The use of IQ tests to intelligence so that children can be placed in make high-stakes decisions aroused strong neg- classrooms deemed to be better suited to their ative sentiments right from the beginning learning speed. In the 1920s, college aptitude (Chapman 1988 ): tests (close relatives of intelligence tests) began At times the rhetoric of the testers was elitist, to be used by selective institutions to identify hereditarian, and defensive. By overselling their academically talented students who did not have reform, they provoked an angry response. By the advantage of having been educated in elite claiming for themselves and their instrument an prep schools. extraordinary power to predict human destiny, they met sharp resistance from those who believed in The focus on the general factor of intelli- America as a place of opportunity, not self- gence is alive and well both in clinical research fulfi lling prophecies. (p. 176) and in practice. Many scholars believe that the clinical use of intelligence tests should focus Nevertheless, the value of IQ tests to important primarily on the general factor of intelligence as institutions in education, government, industry, and measured by overall IQ. These scholars do not the military was so great that the use of IQ tests has deny that other factors exist or that they are been resilient to multiple waves of scholarly criti- associated with important outcomes (Glutting cism and public scrutiny (Kaufman 2000 ). et al. 2006 ). However, they argue that we mea- Intelligence tests were famously involved in sure few abilities with suffi cient validity to use helping the U.S. military with an effi cient means such measurements to make helpful decisions of fi nding an initial placement for millions of about individuals (Canivez 2013 ). These con- recruits during World War I. The success of the clusions are, of course, passionately disputed by initial attempts of psychologists to help in the war many scholars and practitioners in the fi eld. effort has been disputed (Samelson 1977 ), but it is Nevertheless, despite the vituperations of parti- clear that the U.S. military still fi nds value in their sans, the research is still ambiguous enough that continued use. Currently, no one who scores sub- either position is still intellectually respectable stantially below average on the Armed Forces (Schneider 2013a ). 21 The Relationship Between Theories of Intelligence and Intelligence Tests 319

Their approach was to fi rst look at an evaluee’s Wave 2: Clinical Profi le Analysis scores but also to look between the scores, (1940s–Present) beneath the scores, and beyond the scores to assess psychological processes operating idio- Almost as soon as Spearman proposed his theory syncratically in the evaluee: of general intelligence, researchers (including First, we found that one must consider not only Spearman himself) began to produce evidence every subtest score, but every single response and that broad abilities other than g exist (Spearman every part of every response, as signifi cant and rep- 1927). Although the research demonstrating the resentative of the subject. Naturally, many of the intelligence-test responses are highly convention- existence of multiple broad factors was extremely alized and that a subject knows who was president sophisticated, the application of this knowledge of the United States at a particular time merely in practice was somewhat haphazard and clumsy adds to his general score. But where the response at fi rst. Practicing psychologists began to use deviates from the conventional, the deviation does not merely fail to add to his score; it must also be multiple scores from IQ tests to better understand considered as a characteristic which may give us their evaluees, but they tended to use insights material toward the understanding of the drawn from clinical lore rather than from demon- subject…. strated research fi ndings. Second, we found that one may gain some understanding of the subject by comparing the suc- Practitioners were trained to evaluate ability cesses and failures on a given type of test item. profi les in what today is known to be a psycho- Thus, if a subject knows how many pints there are metrically naïve manner, without fully account- in a quart, but does not know what the Koran is, ing for score reliability and stability. this will give us merely an idea of his range of information. But if he knows what the Koran is and Unrealistically specifi c test profi les were said to asserts that a quart has four pints, we must consider be characteristic of specifi c psychiatric diagno- the presence of a temporary ineffi ciency, and if he ses. It would be wonderful if IQ tests were capa- insists that the capital of Italy is Constantinople or ble of making such useful distinctions. that the Vatican is a robe, psychotic maladjustment will have to be considered…. Unfortunately, although it is true that certain In general, one might say that this approach to diagnostic groups do have weak tendencies intelligence testing requires a very different attitude toward certain kinds of profi les, the positive pre- toward tests than does routine intelligence testing, dictive power of such profi les to identify specifi c which hinges upon correct appraisal of whether a response is to be considered passing or failing…. disorders is very low (Watkins et al. 1997 ). On the part of the tester, it requires a great deal of We should not be too quick to cast stones at attention to any type of deviation from the usual run second-wave scholars and their experientially of test performances. (pp. 67–68) and rationally derived clinical lore. Though it is true we do not live in houses made entirely of The second-wave tradition was extended by glass, our homes still have many large and fragile many scholars, including Kaufman (1979 ), windows. Furthermore, those windows are not Sattler ( 1974), and especially Kaplan with her merely decorative but provide useful views of Boston Process Approach (Milberg et al. 1996 ). reality. Much of what practitioners still do is nec- Recent advances in this tradition can be found in essarily based on rationally derived principles, many sources (e.g., Hale and Fiorello 2004 ; clinical experience, common sense, and basic McCloskey et al. 2008 ). human dignity. It would be desirable to reinforce those glass panes with strong research fi ndings, but true progress is diffi cult and thus slow. Wave 3: Psychometric Profi le Much of what is called “naïve psychometrics” Analysis (1970s–Present) can be considered empathic wisdom on the part of practitioners. Among the most thoughtful and Although research on cognitive ability had articulate scholars to promote second-wave prac- always made use of sophisticated statistical tices were Rapaport, Gil, and Schafer (1945–1946 ). methods, practitioners were rarely trained to take 320 W.J. Schneider and D.P. Flanagan advantage of them. In the years following World War II, Luria was able to study and treat a large War II, practical textbooks on psychometrics number of men with diverse, and in many cases began to be published (e.g., Gulliksen 1950 ), focal, brain injuries. The discipline of neuropsy- factor analytic studies of the Wechsler scales chology made rapid advances during this time. showed that non-g abilities could be measured Luria had legendary clinical skills, often mak- (Cohen 1959 ), and fi ndings from neuropsychol- ing up new neuropsychological tests on the spot ogy showed that estimating these non-g abilities to test hypotheses about individual evaluees. was not a waste of time (Halstead 1947 ). Although the creative genius of Luria’s approach The early writings of Sattler (1974) and cannot be duplicated, many of his ideas were Kaufman (1979 ) were extremely persuasive in later used by others. The Kaufman Assessment showing that it was worthwhile to use psycho- Battery for Children (K-ABC; Kaufman & metrics to extract meaningful information from Kaufman 1983 ), the Cognitive Assessment test scores beyond the g factor. Not only were System (CAS; Naglieri & Das 1997 ), and the Sattler and Kaufman persuasive and practical, Luria-Nebraska Neuropsychological Battery their comprehensive and “intelligent” approach operationalized portions of Luria’s methods and to intelligence test interpretation made the pro- theories. Das, Naglieri, and Kirby (1994 ), in par- cess interesting and meaningful to practitioners. ticular, developed and extended Luria’s theories Sattler’s and Kaufman’s early writings were not with their planning, attention, simultaneous, purely psychometric nor were they atheoretical, and successive (PASS) theory and operational- but compared to later efforts (including their ized PASS constructs with the CAS. own), they were less theoretically driven. Much Although Cattell’s Gf-Gc theory had been of their earlier work was devoted to retrofi tting proposed in 1941, it was not tested directly until existing clinical measures not designed to cleanly the 1960s with the help of Cattell’s student, John measure any theoretical entities other than g with Horn (Cattell 1963 ; Horn & Cattell 1966). Horn theoretical interpretations derived from and Cattell, sometimes separately, sometimes Guilford’s (1967 ) structure of-intellect model together, refi ned the theory and subjected it to and Osgood’s (1963 ) psycholinguistic model. critical tests throughout the 1960s and 1970s. By Even at the time, these connections to such theo- the late 1980s, the evidence for the theory was so ries were made with ample warning of their spec- strong that Woodcock restructured his battery of ulative nature. tests to be an admirably comprehensive operationalization of Gf-Gc theory (Woodcock & Johnson 1989 ). The Stanford-Binet Intelligence Wave 4: Application of Theory Scale: Fourth Edition (SB:FE; Thorndike et al. to Interpretation (1980s–Present) 1986 ) and the Kaufman Adolescent and Adult Intelligence Test (KAIT; Kaufman & Kaufman In third-wave approaches, clinical measures were 1993) were also partial operationalizations of mapped onto then-current theories of intelli- Gf-Gc theory. Gf-Gc theory’s status was greatly gence; in fourth-wave approaches, new clinical enhanced when Carroll’s (1993 ) massive reanal- measures were created (or old ones adapted) to ysis of the world’s cognitive ability research operationalize successful theories of intelligence. demonstrated persuasively that the model of Such measures had existed for many decades, but broad abilities specifi ed by the Horn-Cattell they were almost exclusively used for research model was largely correct. Nearly all subse- (e.g., Thurstone’s Primary Ability Battery). quently published multifactor cognitive ability The writings of Soviet neuropsychologist tests have been based explicitly or implicitly on Alexander Luria (1966 , 1973 ) excited a new gen- the ideas of Cattell, Horn, and Carroll (Alfonso eration of cognitive ability researchers and test et al. 2005 ), with Naglieri and Das’s (1997 ) CAS developers. Because of the tremendous number as the primary exception. A number of scholars of casualties suffered by the Soviets during World see considerable merit in both CHC theory and 21 The Relationship Between Theories of Intelligence and Intelligence Tests 321

PASS theory and believe that a successful inte- one battery or another, practitioners can fl exibly gration of the two would be of great theoretical pick tests from any battery to meet specifi c situ- and practical benefi t (Flanagan, Alfonso, & ational demands. Dixon 2014; Kaufman & Kaufman 2004 ). With these theoretical updates to test batteries, new frameworks for test interpretation were Psychometric Forerunners articulated. Building on the work of Woodcock of CHC Theory ( 1990 ), McGrew ( 1997 ) proposed a unifi ed framework for the separate but closely related As shown in Fig. 21.1 , CHC theory has three par- theories of Cattell, Horn, and Carroll, later termed ents ( g f - g c theory, extended Gf-Gc theory, and CHC theory. Also building on Woodcock’s three-stratum theory) and at least two grandpar- (1990) pioneering work, the cross-battery ents (two-factor theory and primary mental approach (Flanagan & McGrew 1997 ; Flanagan abilities). In addition, it has two important fi rst et al. 2013b) was developed to help practitioners cousins once removed (hierarchical Group Factor combine and interpret information from multiple theory and triadic theory). test batteries in a psychometrically defensible manner according to CHC theory. What characterizes fourth-wave test interpre- Two-Factor Theory of Ability tations is the confi dence with which tests can be interpreted according to well-developed theo- Spearman’s (1927 ) two-factor theory of ability is ries such as CHC theory and PASS theory. The the simplest theoretical justifi cation for creating confi dence is reasonably warranted because of a composite score from diverse question types. many replicated fi ndings in cross-battery confi r- The two-factor theory assumes that, aside from matory factor analyses (Flanagan et al. 2013a; error, test scores are determined by two kinds Reynolds et al. 2013). Instead of being bound to of ability: general and specifi c. General ability

Cattell Spearman

gf-gc Theory Two-Factor Theory Direct g g g General Abilities Forerunners f c

of CHC Theory Tests

s1 s2 s3 s4 s5 s6 s1 s2 s3 s4 s5 s6 Specific Abilities

Thurstone Horn Carroll Primary Mental Extended Gf-Gc Three-Stratum Theory Abilities Theory g

I V S N P F M Gf Gc Gv Ga Gs TSR SAR Gf Gc Gv Ga Gr Gy Gs Gt Broad Abilities

Narrow Abilities

Tests Specific Abilities

Fig. 21.1 Direct forerunners of CHC theory 322 W.J. Schneider and D.P. Flanagan

(g ) infl uences all test scores, whereas specifi c predicts that their observed correlation will be abilities infl uence just one kind of test each, as 0.56. shown in Fig. 21.2 . The reason that Spearman One of Spearman’s many insights was that believed that the two-factor theory had merit is even if g has a small infl uence on any particular that in many data sets, as predicted by the the- test item, g can be estimated reasonably well by ory, the product of the g loadings of any two adding many test scores together. In Fig. 21.3 , it tests was close to the observed correlations. can be seen that the sum of two ability tests is a For example, if two tests have g loadings of 0.7 better measure of g than either of the two tests and 0.8, respectively, then the two-factor theory would be if used by themselves. This provides test developers with a rationale for making IQ tests that select from a wide variety of test g General ability formats, contents, and task demands. Spearman’s greatest contribution to CHC the- ab c d g loadings ory, surprisingly, is not the theoretical entity g . Galton’s ( 1888 ) invention of the correlation coef- A B C D Tests fi cient was a major breakthrough, allowing s loadings researchers to ask questions about pairs of vari- sA sB sC sD Specific abilities ables instead of just one variable at a time. What Galton did with pairs of variables, Spearman did ABCD A 1 ab ac ad with whole sets of variables, simultaneously. Implied B 1 bc bd Spearman moved beyond bivariate correlations C 1 cd correlations D 1 to the evaluation of whole correlation matrices. That is, Spearman directed our focus away from Fig. 21.2 Spearman’s two-factor theory of ability. Note : piecemeal analyses to omnibus evaluations of a All observed and latent variables are depicted as z-scores theoretical model as a whole. Spearman’s inven-

The Relationship between g and IQ

.40 .40 Outcome 1 g Outcome 2

.70 .80 .30

49% g .65 64% g .45 s1 Test 1 Test 2 s2 42% s1 20% s2 16% e 9% e1 2 .30 .40 Correlate IQ g .85 s .37 e1 e2 1 s2 .25 e1 .17 e2 .23 72% g Test 1 .88 92% Reliable Test 2 .88 (g + s1 + s2) IQ Outcome 1 .45 20% Specific (s1 + s2) Outcome 2 .34 08% Error (e1 + e2)

Fig. 21.3 The relationship between g and IQ 21 The Relationship Between Theories of Intelligence and Intelligence Tests 323

Table 21.1 Similarities across psychometric models of ability Primary abilities Triadic theory Extended Gf-Gc Three-stratum Thurstone Cattell Horn Carroll Cattell-Horn-Carroll Verbal Crystallized Acculturation Crystallized Comprehension intelligence knowledge intelligence knowledge Reasoning Fluid intelligence Fluid intelligence Fluid intelligence Fluid reasoning Perceptual General Cognitive speed Broad cognitive Processing speed speed perceptual speed speediness Correct decision Reaction time Reaction and decision speed decision speed speed Word fl uency General retrieval Broad retrieval Retrieval fl uency capacity ability Memory General memory Tertiary storage General memory Learning effi ciency capacity and retrieval and learning Short-term apprehension and retrieval Short-term memory Space Visualization Visualization and Broad visual Visual processing capacity spatial orientation perception Auditory capacity Listening and Broad auditory Auditory processing hearing perception Number Quantitative knowledge tion of a procedure that could do this (a rudimen- which he gave to diverse samples at different tary version of factor analysis) is impressive ages. Many of his tests are still used in research, enough, but the real breakthrough was the idea but no clinical instruments were explicitly based that one could and should do this. In a sense, all on Thurstone’s primary mental abilities battery. individual difference researchers are still playing Thurstone’s list of abilities varied somewhat the game that Spearman invented. from study to study, but a stable subgroup of ability factors was found in almost every study. As seen in Table 21.1 , these factors are in most cases Primary Mental Abilities directly analogous to broad abilities in Cattell’s ( 1987) triadic theory, Horn’s extended Gf-Gc Spearman’s rudimentary factor analysis was a theory (Horn & Blankson 2005 ); Carroll’s (1993 ) powerful tool, but it required researchers to three-stratum theory, and Cattell-Horn-Carroll extract only one factor from a correlation matrix theory (McGrew 2009; Schneider & McGrew at a time, with each step requiring hundreds of 2012 ). Thurstone’s research identifi ed a number calculations. Spearman’s methods could account of narrower primary abilities not shown in for group factors, but the calculations were Table 21.1 such as vocabulary, reading compre- tedious. Thurstone (1934 ) invented multiple fac- hension, induction, deduction, and closure tor analysis, which allowed all common factors to (Guilford 1972 ), all of which are represented in be extracted simultaneously via matrix algebra, CHC theory. Thus, CHC theory is not so much a facilitating researchers’ ability to test more com- new theory but an elaboration of very robust fi nd- plex models of intelligence. Thurstone ( 1938 ) did ings that were fi rst discovered by Spearman, not deny the existence of g but emphasized a Thurstone, and many other early researchers. small number of what he called primary mental abilities, which are neither general across all tests nor specifi c to each test (see Fig. 21.1 ). g f -g c Theory Thurstone (1938 , 1947) conducted an extraordinary amount of research developing a compre- Even in his fi rst paper on general intelligence, hensive battery of over 50 cognitive ability tests Spearman (1904 ) was open to the possibility 324 W.J. Schneider and D.P. Flanagan that there were multiple general factors. Three-Stratum Theory of Cognitive Spearman’s student, Raymond Cattell, found Abilities evidence for two general factors of intelligence (Cattell 1941 , 1943 ) which he called fl uid Carroll (1993 ) systematically reanalyzed every intelligence ( g f) and crystallized intelligence data set he could fi nd that would help him gener-

(g c). What athletic talent is to the body, fl uid ate a new taxonomy of abilities. After evaluating intelligence is to the brain. It represents the the results, he found that he had largely recon- speed, power, effi ciency, and overall integrity structed Horn’s model with a number of impor- of the cerebral cortex. In Cattell’s (1987 ) think- tant differences. Carroll believed that there was ing, g f is not an ability itself but an infl uence on strong evidence for the g factor, whereas Horn many abilities, particularly those abilities that did not. As seen in Table 21.1 , Horn and Carroll require controlled attention and on-the- spot parse memory differently. Neither theorist is problem solving. g c is acquired knowledge, wrong; they simply have different emphases. It is particularly information stored in declarative true that nature must be carved at its joints, but memory. Whereas g f is easily damaged (or even different carvers can fi nd equally valid ways to temporarily disrupted by fatigue or intoxica- slice complex domains. tion), g c is relatively robust to the effects of disease, aging, and injury. Triadic Theory of Ability

Extended Gf-Gc Theory Cattell’s (1971 , 1987 ) later work is called the triadic theory of ability (see Fig. 21.4 ). He Under Cattell’s (1963 ) supervision, John Horn hypothesized that some mental capacities (fl uid conducted the first empirical test of Cattell’s intelligence, memory, retrieval fl uency, and pro- theory of intelligence. Some predictions held cessing speed) are not localized in specifi c up well, but the theory needed to be refined regions of the brain. These general capacities and expanded. The new extended Gf-Gc the- depend on the overall integrity of the cortex. ory (Horn and Cattell 1966 ) did not just bor- Some mental capacities, called provincial row Thurstone’s primary abilities, it also powers, mostly depend on regions of the brain provided Thurstone’s ever-lengthening list associated with specifi c sensory modalities. For with some much needed order. That is, it example, individual differences in visualization showed that Thurstone’s primary abilities had capacity result from better functioning in the pri- a higher-order structure and that different primary and secondary visual association cortex mary abilities were differentially subject to regions in the occipital lobe and the rear portions Gf, Gc, and other general factors such as Gv of the parietal and temporal lobes. Each sensory (general visualization), Gs (general speedi- modality with a specifi c region of primary and ness), and Gr (general memory fluency). Later, secondary association cortex is hypothesized to broad abilities were added by Horn’s research have its own provincial power. At the time, independent of Cattell. Cattell’s own theoriz- Cattell had good evidence for the existence of ing and research also continued independently visualization and auditory capacity. He hypothe- of Horn, though the two continued to publish sized that other sensation-specifi c ability factors together from time to time. Confusingly, the would be isolated by future research (e.g., olfac- term g f -g c theory refers to Cattell’s ( 1941 ; tory, gustatory, tactile, and cerebellar). 1943 ) original idea, and the (extended) Gf-Gc Whereas general capacities and provincial theory (aka the Horn-Cattell model) was powers represent infl uences on abilities, agencies refined mostly by Horn (1985 ). As described are developed abilities. In the triadic theory, below, Cattell’s later work is known as the triadic capacities like fl uid intelligence are not true abili- theory of ability. ties but are features of a person’s brain that put 21 The Relationship Between Theories of Intelligence and Intelligence Tests 325

General, Provincial, Agentic: Cattell’s triadic theory of ability

Provincial Triadic General Powers Theory Capacities

Gustatory Visual Processing Retrieval Speed Capacity Olfactory Cerebellar Fluid Memory Auditory Tactile Intelligence Capacity Crystallized Intelligence

Interests

Experience Agencies

Mechanical Spatial Numerical Verbal

Note: Each arrow represents all possible effects from one group to the other

Fig. 21.4 General, provincial, agentic: Cattell’s Triadic Theory of Abilities. Note : Each arrow represents all possible effects from one group to the other

limits on the sorts of developed abilities a person Looking back over the history of the fi eld, I would judge this book to be the single most ambitious and can acquire. The infl uence of general capacities original contribution to the study of mental abili- and provincial powers is always indirect and thus ties since Charles Spearman’s The Ability of Man can only be inferred. Therefore, all abilities (1927 ). Cattell’s book is simply unrivaled in its observed directly are learned abilities. There are scope. Whatever the reader’s overall assessment of the book’s virtues and faults, few who study it thor- many agencies (more than what is shown in oughly will fail to think of it as the work of a bril- Fig. 21.4 ), but one of them (crystallized intelli- liant and encyclopedic intellect. (p. 291) gence) has such a large infl uence on so many tasks that it acts as if it were a general capacity. It feels odd to wholeheartedly agree with

Because of the success of g f- g c theory (Cattell Jensen about anything, but we are happy to 1943 ; Horn and Cattell 1966), it is not a hard sell make an exception in this case. We also agree to suggest that Cattell should be read in the origi- with Jensen’s evaluation of the larger 1987 nal. The rewards of doing so are unexpectedly sequel, Intelligence: It’s Structure, Growth, and fabulous. Cattell was a remarkable communicator Action , written when Cattell was a spry 82 years and a dazzlingly creative scholar. Amazingly, of age. It, too, is a remarkable summary of Cattell’s primary research focus was personality, not Cattell’s thinking but is only barely updated intelligence. In praise of Cattell’s many-splendored from 1971; many references from the 1950s and talents, Carroll (1984 , p. 300) noted that Cattell’s 1960s are referred to as “recent.” Furthermore, work on abilities was “a sideline, conducted, as it Cattell seems to have been so busy with his own were, with his left hand.” Consider this review by research that he does not seem particularly Jensen (1974 ) of Cattell’s (1971 ) Abilities: Their aware of anyone else’s contributions to the fi eld structure, growth, and action : after the 1960s. Nevertheless, Cattell’s 1971 326 W.J. Schneider and D.P. Flanagan masterwork on intelligence and its 1987 sequel Horn and Blankson (2005) and by Cattell ( 1971 ) are fi lled with page after page of insights and make similar (though not identical) distinctions. fascinating details that are rarely mentioned A common misperception of Vernon’s model anywhere else by anyone else. For example, his is that all abilities are subsumed by the v:ed/k:m explanation of how crystallized intelligence clusters. Not at all! Vernon (1961 ) believed that a becomes functionally autonomous from fl uid separate auditory and musical talent cluster of intelligence is detailed and remarkably inge- abilities (similar to Ga) was distinct from both nious (Cattell 1987 , pp. 146–152). v:ed and k:m (see Fig. 21.5). He also believed that sensory discrimination factors and various forms of imagery were also distinct. This is but Hierarchical Group Factors one way in which Vernon’s model has been lamentably oversimplifi ed. The work of British psychologist Philip Vernon A t fi rst glance, Vernon’s model looks a lot like never disappeared from view, but it is now resur- many other hierarchical models. However, there is gent because of the recent successes of its succes- an important difference. Vernon’s model is not a sor, the v erbal- p erceptual-image r otation model straight reporting of factor analytic results. Like (VPR; Johnson and Bouchard 2005 ). Most Carroll’s model, it is instead a summary judgment descriptions of Vernon’s model point out that by Vernon of many factor analyses. In most hier- between the g factor and more specifi c abilities, archical models of ability, including Carroll’s, the abilities can be broadly categorized as verbal/ hierarchies are arranged neatly into two or three educational (v:ed) or spatial/mechanical (k:m). strata with each lower-order ability having one Although not an explicit feature of Carroll’s and only one “parent” ability. In Vernon’s model, model, this was a distinction Carroll (1993 , p. 60) the hierarchy is not neatly arranged, having a believed was valid, noting that there are probably vaguely organic appearance. Abilities diverge and many such intermediate strata in the structure of reconverge at different levels of the hierarchy, ability. The later elaborations of Gf-Gc theory by with some abilities having multiple “parents.”

auditory

musical inductive reasoning psychomotor

physical divergent verbal p thinking ﬂuency creative n spatial mechanical v

reading clerical scientiﬁc & linguistic spelling mathematical technical

Fig. 21.5 Vernon’s hierarchical group factor theory 21 The Relationship Between Theories of Intelligence and Intelligence Tests 327

This is incredibly sensible and important, particu- They wanted a label for the common ground on larly with respect to the achievement factors, which both models stood and both Horn and which are clearly multiply determined. Carroll agreed to have their models yoked As Horn and Noll (1997 , p. 84) famously together under the CHC banner (McGrew 2005 ). warned, a mature theory of cognitive abilities is In CHC theory, g is present but with a de- unlikely to have the rectilinear system of factors emphasized and uncertain status. This has worked implied by factor analysis, but is likely to be to create an engaged community of scholars, described by the “rounded and irregular struc- researchers, and practitioners who can talk about tures of mother nature.” As an example of an cases, discuss research fi ndings, and suggest overly regular model, early diagrams of CHC refi nements to the model without having to theory depict reading/writing ability (Grw) and refi ght constant battles about the existence of g . quantitative ability (Gq) as being connected to Second, now that all three original theorists other factors solely by g , though this was never have passed away, their models will calcify if not what verbal descriptions of the theory implied continuously updated. Without the CHC label (McGrew 1997 , 2005 , 2009 ). Grw and Gq were and the ecumenical space it creates, each scholar always hypothesized to have direct relationships with a slightly different opinion (i.e., everyone) with Gc, Gf, and other abilities. Nevertheless, might feel the need to rename the model, causing this confusion led Major, Johnson, and Deary unnecessary confusion and fragmentation (e.g., ( 2012 ) to test and reject the models implied by Schneider’s revised McGrew-Horn-Flanagan the misleading fi gures. model). Since CHC theory was introduced Before Major, Johnson, and Deary’s paper (McGrew 1997), it has been updated twice in was published, steps had already been taken to response to new research (McGrew 2009; provide a clearer picture of CHC theory, showing Schneider & McGrew 2012). It is likely that that some broad abilities were more closely more updates will follow as needed. related than others (Schneider & McGrew 2012 ). CHC theory recognizes g at the top of the hier- Further clarifi cations of the exact nature of the archy, 16 broad abilities below g , and about 80 relations among CHC broad abilities are still narrow abilities nested within the broad abilities. needed. As it stands, CHC theory and the VPR In several places, there are intermediate catego- model are now explicitly more alike in several ries between the three strata. Some parts of the ways (though not identical). If the VPR model taxonomy are more settled than others. CHC continues its run of empirical successes, practi- theory has largely maintained Carroll’s (1993) tioners will have increasing reason to explore its nomenclature, with departures from the original forerunner, Vernon’s model. detailed in Schneider and McGrew (2012 ). In Fig. 21.6, there are “level” factors (abilities defi ned by the diffi culty of the task) in the top Cattell-Horn-Carroll Theory row and “speed” factors (abilities defi ned by the of Cognitive Abilities rate at which simple tasks can be completed). However, the distinction between level and speed If we have the separate works of Cattell, Horn, is not a true dichotomy because there are speed- and Carroll, why do we need the Cattell-Horn- accuracy trade-offs in most tasks. It is not known Carroll theory of cognitive abilities? First, Horn whether the speed factors are merely descrip- and Carroll produced very similar theories in part tively similar or if there is a superordinate speed because they were based on the same data sets. factor that is distinct from g . Some aspects of They differ primarily in that Horn’s model has no cognitive speed are diffi cult to measure distinctly; place for g , whereas Carroll puts g in a place of for example, some rapid naming tasks likely prominence in the hierarchy. Woodcock and involve a murky mix of perception speed, mem- McGrew recognized that the debate about the ory retrieval fl uency, oral-motor articulation validity of the g factor was not likely to end soon. speed, and attentional fl uency. 328 W.J. Schneider and D.P. Flanagan

Fig. 21.6 Broad ability groups in CHC theory

The perceptual abilities in Fig. 21.6 are example, in different places or in fundamentally grouped together for conceptual economy rather different ways depending on whether one is a than for structural reasons. That is, they do not historian. However, the densely interconnected correlate more with each other than with abili- networks of association in expert knowledge ties in other categories. Each broad perceptual make possible a phenomenon called expertise ability refers to the effi ciency and power of per- wide-span working memory (Horn and Blankson ceptual reasoning a person is capable of within 2005), in which an expert can hold in mind and a particular modality. It is not known whether process much more information in working mem- each perceptual ability will require its own ory than is otherwise possible with information speed factor or if there are common constraints outside the expert’s specialty. Gkn is hexagonally on perceptual speed. shaped in Fig. 21.6 because specialized knowl- The connections among the knowledge abili- edge is partly infl uenced by career interests, ties in Fig. 21.6 are not merely descriptive. For which do have this structure (Holland 1985 ). example, language ability and general knowledge However, the hypothesis that domain-specifi c directly infl uence reading comprehension and knowledge has the same structure as career inter- written expression. Gc and Gkn (domain- specifi c ests is still untested, and the analogy between the knowledge) are not sharply distinct abilities, and two domains is likely to be only approximate. Gkn is not “an ability” at all but a diverse, frag- In early work on fl uid intelligence, what are mented collection of abilities. Gc and Gkn repre- now termed working memory capacity tests (tests sent two ways in which knowledge serves that require simultaneous short-term storage and different functions. Gc is broad knowledge that processing of information) were used as markers facilitates communication across an entire cul- of fl uid ability. Although this practice has been ture. Gkn is specialized and deep knowledge that discontinued, the connection between Gsm and is primarily useful within a particular fi eld, often Gf is intimate (Kane et al. 2004 ), most likely an extremely narrow one. It is unlikely that the because of a shared dependence on the ability brain stores information about Charlemagne, for to control the focus of attention (Unsworth & 21 The Relationship Between Theories of Intelligence and Intelligence Tests 329

McMillan 2014 ). At the core of fl uid intelligence gence but as Binet’s notion (Tuddenham 1962 ) is the ability to perceive new relationships of intelligence in general . If there are a few between objects and concepts held temporarily in broad abilities that infl uence performance in working memory. Better working memory capac- many tasks (e.g., verbal ability, spatial ability, ity is not synonymous with better reasoning logical reasoning, and so forth), a simple aver- capacity, but reasoning is much facilitated by age of these important abilities is a useful sum- working memory capacity. mary of a person’s capacity to act intelligently. Current conceptions of working memory This is true whether g exists or not. (Baddeley 2006 ; Unsworth & Engle 2007 ) emphasize the tight connection between short- and long-term memory. In Fig. 21.6, Gsm and IQ Versus Weighted Predictors Glr are misleadingly far apart. In both Cattell’s triadic model and Carroll’s model, they are A fi nely tuned multiple regression equation with grouped in the same superordinate category, multiple abilities as predictors typically forecasts memory. Clean measures of any one memory outcomes somewhat better than IQ (an equally process are probably impossible to design. All weighted composite score). However, the advan- memory tasks involve attention, encoding, pro- tage of a specifi c multiple regression equation cessing, storage, and retrieval but are designed over IQ is not usually large, typically in the range with different information processing bottlenecks of 1–6 % additional variance explained (Carretta so that different mixtures of memory abilities & Ree 2000 ; Hunter 1986 ). This fact has long determine performance. been known and, under plausible conditions in the ability domain is mathematically inevitable (Wainer 1976 ; Wilks 1938). Dramatically better Doing Wave 1 Well predictions will come from identifying dramatically better predictors; fi ne-tuning regression On many occasions, Spearman (1915 ) took Binet coeffi cients with existing predictors will only and other test developers to task for creating IQ improve predictions a little bit. tests with diverse item types but denying the What little advantage fi nely tuned regression existence of g. If all test scores are determined equations might have had over equally weighted solely by specifi c abilities, it is hard to justify the IQ scores has been reduced because abilities that practice of calculating IQ. Why? If only specifi c have the highest correlations with important out- abilities are measured, then a person’s IQ would comes tend to be overrepresented in IQ batteries. depend on the arbitrary choices made by test For example, there are ten subtests in the developers as to which abilities are measured. WISC-IV FSIQ. From the perspective of CHC Different IQ tests measuring nonoverlapping theory (McGrew 1997, 2005, 2009; Schneider sets of specifi c abilities would not correlate with and McGrew 2012 ), there are three tests of verbal each other. comprehension and knowledge (information, However, Spearman’s objections can be vocabulary, and similarities), two tests of fl uid overcome if we posit the existence of a rela- intelligence (matrix reasoning and picture con- tively small number of abilities that are broad cepts), two tests of working memory (digit span (i.e., neither general nor specifi c) and predict a and letter-number sequencing), two tests of pro- wide variety of criteria, such as academic and cessing speed (coding and symbol search), and occupational success. For example, Cattell’s one test of visual-spatial processing (block 16PF had a single- score intelligence measure design). This unequal mix of abilities (3 Gc:2 even though Cattell did not believe that there Gf:2 Gsm:2 Gs:1 Gv) approximates the relative was a single entity called g . This practice is jus- sizes of the regression coeffi cients one would tifi able if IQ is seen not as a measure of fi nd in a multiple regression equation prediction Spearman’s (1904 ) notion of general intelli- of academic and occupational success. 330 W.J. Schneider and D.P. Flanagan

Fig. 21.7 Conﬁ dence intervals for g and FSIQ when WISC-IV FSIQ = 85

Finally, there is a practical advantage of IQ ysis is done, g will likely get all the credit in its scores: It is an inconvenient truth that a different prediction of Outcome 1. multiple regression equation must be computed IQ scores are famously reliable—and reliabil- for each outcome. An IQ score, which is only ity, famously, is not the same as validity. High reli- slightly suboptimal as a predictor, must be com- ability means that a true score can be located with puted only once. Thus, whether one believes in g precision. However, an IQ’s true score is not so or not, an IQ score is still a remarkably effective much an estimate of g as it is an estimate of the and practical predictor of important outcomes. entire mix of reliable sources of variance. The WISC-IV FSIQ, across age groups, has a classical test reliability coeffi cient of 0.97. Using a model The Relationship Between IQ and g of the WISC-IV developed by Weiss, Keith, Zhu, and Chen ( 2013), it can be deduced that g makes Although the terms “IQ” and “g ” are sometimes up 81 % of the variance in the FSIQ, with 16 % of used interchangeably, they are not quite the the variance attributed to other abilities and 3 % of same. At fi rst glance, their relationship is quite the variance attributed to error (Schneider 2013b ). simple: Observed IQ is an imperfect estimate of Does this distinction between g and the true the latent variable g. In truth, IQ can be a com- score matter? It does if you think that FSIQ is an plex mix of many things besides g. For example, estimate of g . Consider a person with a WISC-IV in Fig. 21.3, the IQ score is infl uenced by the FSIQ of 85. In Fig. 21.7 , it can be seen that the specifi c abilities measured by Tests 1 and 2. If 95 % confi dence interval for the true score is those specifi c abilities have correlations with an between 80 and 90, a 10-point range. In contrast, outcome, the IQ score will have a greater cor- g cannot be estimated with the same level of pre- relation with the outcome than would a more cision; its 95 % confi dence interval is between 74 pure estimate of g. For example, in Fig. 21.3 , and 99. This is 2.5 times larger than the confi - both Outcome 1 and Outcome 2 are equally cor- dence interval for the true score! Although a related with g (r = .40). However, because 25-point confi dence interval covers a wide range, Outcome 1 is also correlated with the specifi c it is not so wide that the estimate is useless; in ability in Test 1, IQ is more strongly correlated this case, the estimate allows us to be confi dent with Outcome 1 (r = .45) than with Outcome 2 that the person’s g is lower than the population ( r = .34). Unfortunately, unless the correct anal- mean but not extremely so. 21 The Relationship Between Theories of Intelligence and Intelligence Tests 331

Not All General Ability Scores Individualizing Individual Estimate g Equally Well Difference Models

Although most cognitive batteries estimate g rea- One of the central insights of second-wave schol- sonably well, some do it more precisely than oth- ars is that the way in which a person completes a ers. The Stanford-Binet 5 is especially noteworthy task matters. In some cases, the manner in which in that it has many highly g -loaded subtests. the person approaches the test completely alters Based on a simple 1-factor model using the stan- what is being measured. Although second-wave dardization sample norms (ages 11–16), it is esti- scholars frequently reiterate this point, rarely do mated that g explains 92 % of the variance in their analyses go further than assertions and anec- FSIQ (more realistic models result in slightly dotes. However, it is quite possible for these lower estimates around 90 %). When the SB5 hypotheses to be operationalized and tested empir- FSIQ is used to estimate g , the 95 % confi dence ically (e.g., Hegarty 2010 ; Kyllonen et al. 1984 ). interval is only 16 points wide, noticeably more To illustrate, imagine that there is a test that precise than the 25-point interval on the WISC-IV. requires the agile use of controlled attention to It can be estimated that about 72 % of the vari- respond fl uently to a sequence of visual images. ance in the WJ III GIA Extended is due to g , Such tests are usually thought of as measures of resulting in a 95 % confi dence interval for g that is processing speed (Gs). However, unanticipated by 31 points wide. The reason that the SB5 is so good the test developers, suppose there is an alternate at estimating g and that the WJ III is not is that the strategy by which the task can be completed very SB5 fails to measure any other factor with preci- quickly, even if the person has low Gs. Suppose that sion (Canivez 2008 ). In contrast, the WJ III was this alternate strategy, though not diffi cult to imple- explicitly designed to measure abilities with low g ment, requires spatial visualization and is typically loadings. Note, however, the fact that the WJ III only discovered by evaluees with high Gv. For eval- GIA Extended is less correlated with g does not uees who use the “spatial strategy,” Gs is not mea- mean that it is less reliable or a worse predictor of sured at all, but instead performance mostly depends important outcomes. On the contrary, it is about on Gv. For individuals who do not adopt the spatial as reliable as the SB5 FSIQ, and it predicts out- strategy (even if their Gv is high), the test is primar- comes with non-g infl uences. Thus, outcomes that ily a measure of Gs, as shown in Fig. 21.8 . are infl uenced by g alone (e.g., Outcome 2 in For the sake of simplicity, suppose that evalu- Fig. 21.3 ) are best predicted by highly g -loaded ators can easily tell when this spatial strategy is tests like the SB5. The WJ III shines in its ability employed, and they check a box when the telltale to predict outcomes explained by abilities other signs of the strategy’s use are observed. Therefore, than g (e.g., Outcome 1 in Fig. 21.3 ). “spatial strategy” is a binary variable for which 1 and 0 denote the strategy’s presence and absence, respectively. As shown in Fig. 21.8 , the probabil- Taking Wave 2 Practices Seriously ity that the spatial strategy will be used is a function of Gv. Only about 1.7 % of people use the There will never be a complete list of the mani- spatial strategy. Of strategy users, Gv is, on aver- fold infl uences on test performance nor should age, 1.75 standard deviations above the mean. there be. If someone were to try to make one, it However, even most people with high Gv do not would be as useful as comedian Steven Wright’s discover the strategy. For example, only 16 % of full size maps (“Scale: 1 mile = 1 mile”). The best people with Gv exactly 2 standard deviations we can do (i.e., should do) is generate a useful list above the mean use the strategy, and only 50 % of of categories that orients our thinking toward pro- people with Gv 3 standard deviations above the cesses and infl uences that are consequential and mean use it. Among non-strategy users, Gv is leads us to achieve the goals of our assessments. essentially average (mean = −0.03). 332 W.J. Schneider and D.P. Flanagan

.4 .7

Spatial Strategy? Gs P(Yes)= Gv Φ(Gv−3) −.8 .8

.8 .0 Correlations with Observed Test Scores “Gs” Sample g Gv Gs Test All (100%) .34 .26 .79 No Strategy (98.3%) .32 .22 .80 Spatial Strategy (1.7%) .41 .70 .15

Fig. 21.8 A test of processing speed (Gs) is sometimes a measure of visual-spatial ability (Gv). Note: Φ is the cumulative distribution function of the standard normal distribution

Fig. 21.9 A “Gs” test measures Gv when a spatial strategy is employed

In Fig. 21.8 , the loading of Gs on the test is individuals with low Gs tended to perform well normally 0.8 but is 0 when the spatial strategy is on the test if the spatial strategy was employed. used. Likewise, the loading of Gv on the test is What conclusions would come from the typi- normally 0 but is 0.8 when the spatial strategy is cal correlational study in which the effect of the used. Thus, among spatial strategy users, the test spatial strategy is not modeled? The direct effect is not a measure of Gs but of Gv. Otherwise, the of Gv would probably be close to 0 because test functions as it was intended. Because typi- among non-strategy users the observed correla- cally only people with high Gv discover the strat- tion between Gv and the “Gs” test is due only to egy, test scores among spatial strategy users are the shared infl uence of g . Clinicians would persist on average 1.4 standard deviations above the in arguing that this “spatial strategy” phenome- mean. In Fig. 21.9 , 10,000 cases were simulated non is real and that it matters. Researchers who according to the model. It can be seen that even only look at the correlational data would counter 21 The Relationship Between Theories of Intelligence and Intelligence Tests 333 that Gv’s effect is small and thus might treat what facts does more good than learning true ones. clinicians say about the spatial strategy as However, there will never be a time in which the “unsubstantiated clinical lore.” insights of the second-wave scholars will be Many validity-altering test behaviors are obviated fully by scientifi c fi ndings. The under- impossible to study because they occur exactly appreciated genius, William Stern (1900, as cited once in a practitioner’s career (e.g., an extremely in Lamiell 2003 ), said it best: bright examinee might score low on similarities [E]very individual is a singularity, a one-time because she believes that she can only give existing being, nowhere else and never before single- word responses). Because there is an infi - present. To be sure, certain law-like regularities nite number of ways to be unusual, practitioners apply to him, certain types are embodied in him, but the individual is not exhausted by these laws observe rare behaviors regularly. In such cases, and types; there remains ever something more, there is no need to let the hobgoblin of little through which the individual is distinct from oth- minds dictate a consistent by-the-book interpre- ers who conform to the same laws and types. And tation of the test score. Unusual behaviors are this last kernel of being, which reveals the individual to be thus and so, distinct from all others, is diffi cult to interpret precisely because we have no not expressible in the language of scientifi c con- experience with them. It is probably impossible cepts, it is unclassifi able, incommensurable. In to avoid generating explanations for unusual this sense, the individual is a limiting concept, behavior, but it is generally safer to not take our toward which theoretical investigation strives but can never reach; it is, one could say, the asymptote interpretations too seriously. Testing the same of science. (pp. 15–16) ability with another test is generally the best course of action when the validity of a test has We can only add that, as humanists who value possibly been compromised by unusual test the dignity of individuals, our sentiments resonate behaviors. with the poetic rhythm of Stern’s rhetoric; as scientists, we are rather restless to see just how much we can close the gaping chasm between our cur- When Will the Second Wave End? rent understanding of individual cognitive processes and whatever asymptotes there might be. Well-substantiated research fi ndings can never be The upside of the second-wave approach is as rich as the individually tailored case conceptu- that it acts as a fail-safe for our incomplete and alizations of thoughtful practitioners because partially incorrect theories and measures. We do practitioners can always take the latest fi ndings not yet have mature theories and measures that and incorporate them into their assessments. account for the incredibly diverse approaches What defi nes second-wave scholars is not a lack people take while completing cognitive tasks. As of empirical evidence but their focus on individu- mentioned above, much of the validation efforts als. Their insights can be taken seriously and for our measurement tools assume a one-size- studied rigorously. One of the biggest obstacles fi ts-all interpretation (e.g., “block design mea- to investigating the claims of process-oriented sures visualization”). Of course no researcher scholars is that researchers from the psychomet- actually believes that each test means the same ric tradition tend to rely on latent variable model- thing for everyone, but we do not yet have coming, and only fairly recently has SEM software pelling research that allows us to construct psy- (e.g., Mplus, OpenMx) made it convenient to chometrically sophisticated models that account evaluate the kinds of if-then relationships that for countless possible moderators of performance process-oriented scholars tend to highlight. on our tests. It is a scientifi c victory every time that “clini- All practitioners have stories about how a cal lore” is found to be substantiated by well- standard interpretation of a test score would have designed research. Indeed, there is reason to resulted in a wildly inaccurate case conceptual- celebrate when such lore is found to be an unsub- ization and perhaps an injustice to the evaluee stantiated myth: Sometimes unlearning false (Schneider 2013a ). Second-wave approaches 334 W.J. Schneider and D.P. Flanagan allow us to combine psychometric data and nary stages of theory validation, it is compara- research-based theoretical understandings with tively easy to conduct several dozen exploratory our intuition, empathy, and sometimes creativity studies (with college sophomores by the hundreds) so that such misunderstandings and injustices are than it is to publish a single instrument with a large less likely to occur. nationally representative standardization sample. The downside of second-wave approaches is This situation means that there will always be that they are only as good as the person using some amount of pressure to retrofi t older tests with them. Many accounts of the clinical exploits of new meanings. Theoretical explanations come and the masters of these methods have uncomfortable go, but digit span is forever! amounts of hagiography and hero worship. It is In the excitement of the early work on cross- reasonable to assume that among practitioners battery assessment (XBA; Flanagan and McGrew there are large individual differences in talent for 1997; Flanagan & Ortiz 2001 ), a fair amount of empathic observation. Second-wave approaches such retrofi tting necessarily occurred. From the probably magnify the good that can be done by beginning, in communicating with practitioners, the gifted but also magnify the possible harmful care was taken to distinguish between tests placed effects of the interpretations and actions of the in a particular category because of strong research incompetent. How can we know to which group fi ndings, weak research fi ndings, or expert con- we belong? As one of Tolstoy’s characters in sensus. Most early speculations about which tests Anna Karenina remarked, “No one is satisfi ed would load on which factor were later substanti- with his fortune, and everyone is satisfi ed with ated, but some were not. For example, for decades his wit.” By which standards can an individual- the Wechsler block design subtest was thought to ized interpretation be challenged? Directly chal- be primarily a measure of fl uid reasoning (Gf). lenging another professional is usually considered Theory-driven cross-battery factor analyses later rude, if not unethical. revealed that for most people block design Even if we are sincerely open to correction, primarily measured visual processing (Gv; what feedback mechanisms are available to eval- Flanagan et al. 2013a ). Conversely, the Wechsler uators who wish to identify and learn from their Arithmetic subtest has been interpreted variously interpretive errors? Second-wave scholars do as a measure of working memory capacity (Gsm), have some answers to such questions, but there quantitative ability (Gq), verbal ability (Gc), and has never been a compelling account of how ordi- fl uid reasoning (Gf), depending on the develop- nary practitioners can use these kinds of mental level of the sample and the diversity of approaches to make more valid interpretations ability measures included in the analyses. than could have been made with more purely Although much progress has been made in psychometric approaches. On balance, however, making broad ability classifi cations, many nar- we estimate that second-wave approaches have row ability classifi cations of clinical tests are still done far more good than ill. Nevertheless, there is untested hypotheses. Part of the reason for this is much room for improvement in second-wave that the most convincing data from Carroll’s methods and training practices. work is at the broad ability level, and, as Carroll (1998 ) warned the fi eld, the taxonomy of narrow factors is not at all settled. Since the publication Lessons Learned from Wave 3 of Carroll’s ( 1993) masterwork, no one has attempted a systematic reevaluation of the struc- Just as it is easier for second-wave scholars to ture of the narrow abilities. speculate than it is for researchers to validate those If retrofi tting must be done, we have some ideas, it is easier for applied researchers to gener- advice: ate new fi ndings than it is for test developers to 1. Be excited about new theories. Use them incorporate new ideas into well-normed batteries with gusto! However, remember that they are that can be used by practitioners. At the prelimi- probably doomed to sound silly a decade or 21 The Relationship Between Theories of Intelligence and Intelligence Tests 335

two from now. It is hard to convey to graduate whether a test measures lexical knowledge, students how thrilling some of the now- perceptual speed, memory span, and the like neglected theories of intelligence were in their cannot be our ultimate concern. What matters day. Maintain a healthy sense of history: Most is what each test can tell us about individuals. new ideas are not dramatic paradigm shifts, If we do not know what a low score on the WJ but are actually old ideas with new names and III Visual Closure means for the future well- a bit of twist. Often those twists represent real being of an individual, the test’s place in a tax- progress. However, some developments are onomy is of minor importance. heralded with so much fanfare that it is inevi- 6. New scientifi c terms are introduced when table that they will ultimately disappoint us. cracks and gaps in our understanding are 2. Kaufman and Sattler had the right idea (and discovered. Some new terms are simply still do)—if you must retrofi t old tests with labels for the holes. In contrast, some scien- new interpretations, all fl ights of fancy must tifi c ideas actually fi ll in the holes and con- take off and land on the runway of psychomet- tribute to soundness of the whole structure. ric constraints. For example, if two tests are Maybe that is why we call them constructs . allegedly measuring the same thing based on It is important not to confuse these two your task analysis, they should probably cor- types of labels, heeding Melville’s warning relate with each other more highly than they that mouthing hard words is not the same as do with other tests. understanding hard things (nor is specifying 3. The human mind has trouble grasping nature’s a latent variable model). One sign that preference for continua over dichotomies. As something is amiss is when an earwormish a shorthand for communication, it is common bit of jargon jingles and jangles itself into to label a test as belonging to a certain cate- every aspect of a theoretical domain. For gory (“the WISC-IV matrix reasoning subtest example, it is unclear at this point what abil- measures Gf and the WJ III Visual Closure ity is not an “executive function.” A con- test measures Gv.”). In reality, tests refl ect one struct that is invoked to explain everything or more factors, each to varying degrees (e.g., is as good as one that explains nothing. An matrix reasoning has a secondary loading on admirable attempt to rescue the term execu- Gv that is slightly larger than visual closure’s tive function from excessive generality can primary loading.). The important question is be found in the work of McCloskey and col- not so much “ Does this test measure factor leagues (2008 ). X?” but “How well does this test measure fac- 7. When retrofi tting old tests with new theories, tor X? For whom? Under what conditions?” there is a tendency for one’s understanding of 4. There are more things in heaven and earth, an idea to morph and drift over time. In dis- Horatio, Than are dreamt of in your philoso- cussions of these constructs, particularly phy. Recognize that the latest psychological among novices, it is common to hear Ga con- constructs are probably not going to be mea- fl ated with Gc, Gsm with Glr, and Gv with Gf sured cleanly by any tests that predate the new and/or Gq. One must read and reread source ideas. It is going to be messy, and there is no materials regularly to repair and refresh leak- point in trying to make things fi t snugly. For ing and lapsing memories. example, Wechsler’s picture arrangement subtest is a great test of general ability. Many have attempted to classify it into one theoreti- The Maturation of the Fourth-Wave cal category or another, but none have been Approach terribly convincing. 5. Although classifying tests can be fun, it is a Imagine a musical artist who selects the fi nest stale enterprise when it becomes an end in musical instruments, the best musicians, and itself. It is easy to lose sight of the fact that state-of-the-art recording facilities to record a 336 W.J. Schneider and D.P. Flanagan great new song. The music is recorded with the be used to give clinicians a far more accurate utmost care, and then months are spent meticu- picture of not only what is being measured but lously improving the recording with sophisti- the precision with which inferences can be made. cated postproduction software. All this effort is Ideally, instead of rough rules of thumb, clini- wasted if people can only listen to the music on cians would have guidance about how many tests old cassette players with distorted tinny speakers. would be needed to measure a particular ability It is likewise a waste of effort for cognitive ability with a preferred level of precision. Before such researchers to develop increasingly refi ned psy- information could be available to practitioners, chometric models if clinicians are not given bet- much more cross-battery research will need to be ter tools (both conceptual and technical) with conducted such as the large multi-battery study which to make use of those models at the point of by Reynolds and colleagues (2013 ). contact with individuals. This emphasis on precision of measurement With powerful and comprehensive theories of can be extended to prediction. For example, it is intelligence such as CHC theory and PASS the- widely known that phonological awareness defi - ory, well-trained clinicians can describe an indi- cits are causally related to reading decoding dif- vidual’s cognitive ability profi le with much more fi culties (Wagner & Torgesen 1987 ). If we fi nd confi dence than was possible in the recent past. that a child of otherwise average intelligence has Wave 4 models and methods are an advance, but low reading decoding ability and low phonologi- there is still much room for improvement. cal processing, it is easy to believe that one’s Flanagan, Ortiz, and Alfonso ( 2013 ) present a set work is done—an explanation has been found! of practical guidelines for constructing a battery Indeed it has…partially. In most cases, though, that will measure broad and specifi c abilities in a there is more to the story. In fact, most people theoretically sound and psychometrically defen- with average general ability but with low phono- sible manner. For example, each broad ability is logical processing do not read poorly (Schneider to be measured with two or three qualitatively 2013a ). The vast majority of people with low different tests, preferably from different narrow phonological processing abilities (but average abilities. Such guidelines will become increas- intelligence) actually read at an average level or ingly nuanced with time; in order to achieve the better! same level of precision, ability clusters that are Though low phonological processing may not less correlated (e.g., visual-spatial processing) be a suffi cient explanation for poor reading skills, need to be measured with more tests than do abil- it is true that it substantially elevates the risk of ity clusters that are more highly correlated (e.g., having reading decoding problems. Therefore, crystallized intelligence). when we encounter a child with reading decod- In most clinical validation research, the ing problems, the question we should ask is not emphasis has been on which factor a test mea- “ Does low phonological processing account for sures, not on how well it does so. In some cases, this child’s reading diffi culties?” It probably the tools and data sets needed to answer how well does, at least a little. The better question to ask is, we measure various abilities are not just within “How much of this child’s underperformance in our grasp, but already in our hands. It is possible reading is explained by low phonological pro- to use existing data sets and models to calculate cessing?” In what way is this question better? It how well we are measuring the theoretical con- is actually rare that phonological processing is so structs we intend to measure. For example, using low that it is a suffi cient explanation of reading a model of the WISC-IV developed by Weiss and problems. If an evaluee’s reading underperfor- colleagues (2013 ), Schneider (2013b ) illustrated mance remains mostly unexplained (and it is usu- a method of calculating latent variable scores for ally is), we are prompted to search for additional individuals along with appropriate confi dence explanations, including other relevant cognitive intervals around those scores. Such methods, defi cits. Often these additional explanations combined with large cross-battery models could involve the child’s personality, family system, 21 The Relationship Between Theories of Intelligence and Intelligence Tests 337 past learning history, current learning environ- done with the aid of computer software. Instead ment, and a host of other considerations. A fuller of loosely applying structural models of cogni- accounting of what caused the child’s reading tive abilities to test interpretation in our heads or diffi culties is likely to lead to substantially better using rigid cookbook-style interpretive software, interventions. Focusing solely on phonological clinicians will increasingly use software to inter- awareness defi cits might lead to insuffi cient and actively and fl exibly apply complex mathemati- ill-fi tting interventions that are likely to fail, even cal models directly to individual test data though phonological processing really is part of (Schneider 2010 , 2013b ). the problem (Schneider 2013a ). It is likely that cognitive ability test interpreta- Methods for applying complex prediction tion will be directly incorporated into academic equations are available, but their use is not yet progress monitoring. That is, traditional cogni- common, in part because the methods are not tive ability assessments and the response-to- fully developed and also because they can be intervention (RTI) approach will become mathematically daunting. A number of scholars integrated into a coherent and unifi ed interpretive are attempting to make it easier to apply complex framework. Complex prediction equations prediction equations in the assessment of indi- involving cognitive and academic abilities will viduals (e.g., Crawford et al. 2012 ; Flanagan include important covariates such as past perfor- et al. 2013b; Schneider 2010 , 2013b ). mance, time on task, task persistence, and quality of instruction. This framework is likely to be increasingly informed by dynamic brain imaging Wave 5? The Future of Cognitive techniques and well-developed cognitive infor- Ability Test Interpretation mation processing models. That is, we will be able to observe information processing defi cits Predictions of the future are often laughable in and abnormalities in real time as evaluees per- retrospect, but we will nevertheless hazard a few form academic tasks. Furthermore, we will be predictions about how cognitive ability tests will able to monitor directly whether interventions be interpreted in the coming decades. The initial succeed in normalizing the processing defi cits we hopes of a successful integration of information identify. processing theories and intelligence theories Like the fi rst automobiles, airplanes, and com- have not yet been realized (Hunt 2011 ). Many puters, these innovations initially will be researchers have proposed various models by extremely expensive and inequitably distributed which individual differences in cognitive abilities and probably will not work very well. If clini- can be seen as parameters of information pro- cians, researchers, and entrepreneurs are allowed cessing (Kyllonen 2002 ; Lohman 2000 ; to innovate, these procedures will become Schneider & McGrew 2013 ; Woodcock 1993 ). cheaper and universally available and, most As yet, these models are mostly untested hypoth- importantly, will actually work. eses and are not yet ready to be applied in everyday practice (Floyd & Kranzler 2012 ). However, ultimately we hope to have a consilient account References of the philosophy of mind, neuroscience, universal cognitive processes, and individual differ- Alfonso, V. C., Flanagan, D. P., & Radwan, S. (2005). The ences in intelligence. impact of the Cattell-Horn-Carroll theory on test development and interpretation of cognitive and aca- One of the barriers to practitioners using the demic abilities. In D. Flanagan & P. L. Harrison (Eds.), latest theories and methods from cognitive ability Contemporary intellectual assessment: Theories, research is that cognitive ability theories are now tests, and issues (2nd ed., pp. 185–202). New York: almost exclusively expressed in terms of complex Guilford. Baddeley, A. (2006). Working memory: An overview. In mathematics. Therefore, it is likely that test inter- S. J. Pickering (Ed.), Working memory and education pretation, like test scoring, will be increasingly (pp. 1–31). New York: Elsevier. 338 W.J. Schneider and D.P. Flanagan

Binet, A. (1905). Analyse de C. E. Spearman, the proof Flanagan, D. P., & McGrew, K. S. (1997). A cross-battery and measurement of association between two things approach to assessing and interpreting cognitive abili- and general intelligence objectively determined and ties: Narrowing the gap between practice and cogni- measured. L’Année Psychologique, 11 , 623–624. tive science. In D. P. Flanagan, J. L. Genshaft, & P. L. Binet, A., & Simon, T. (1905). New methods for the diag- Harrison (Eds.), Contemporary intellectual assess- nosis of the intellectual level of subnormals. L’annee ment: Theories, tests, and issues (pp. 314–325). Psychologique, 12 , 191–244. New York: Guilford Press. Canivez, G. L. (2008). Orthogonal higher order factor Flanagan, D. P., & Ortiz, S. O. (2001). Essentials of cross- structure of the Stanford-Binet Intelligence Scales–for battery assessment . Hoboken: Wiley. children and adolescents. School Psychology Quarterly, Flanagan, D. P., Alfonso, V. C., & Reynolds, M. R. 23 , 533–541. (2013a). Broad and narrow CHC abilities measured Canivez, G. L. (2013). Psychometric versus actuarial and not measured by the Wechsler scales: Moving interpretation of intelligence and related aptitude bat- beyond within-battery factor analysis. Journal of teries. In D. H. Saklofske, C. R. Reynolds, & V. L. Psychoeducational Assessment, 31 , 202–223. Schwean (Eds.), The oxford handbook of child psycho- Flanagan, D. P., Ortiz, S. O., & Alfonso, V. C. (2013b). logical assessments (pp. 84–112). New York: Oxford Essentials of cross-battery assessment (3rd ed.). University Press. Hoboken: Wiley. Carretta, T. R., & Ree, M. J. (2000). General and specifi c Flanagan, D. P., Alfonso, V. C., & Dixon, S. G. (2014). cognitive and psychomotor abilities in personnel Cross-battery approach to the assessment of executive selection: The prediction of training and job perfor- functions. In S. Goldstein & J. Naglieri (Eds.), mance. International Journal of Selection and Handbook of executive functioning (pp. 379–409). Assessment, 8 , 227–236. New York: Springer. Carroll, J. B. (1984). Raymond B. Cattell’s contributions Floyd, R. G., & Kranzler, J. H. (2012). Processing to the theory of cognitive abilities. Multivariate approaches to interpretation of information from cog- Behavioral Research, 19 , 300–306. nitive ability tests: A critical review. In D. Flanagan & Carroll, J. B. (1993). Human cognitive abilities: A survey P. Harrison (Eds.), Contemporary intellectual assess- of factor-analytical studies. Cambridge, UK: ment: Theories, tests, and issues (3rd ed., pp. 497– Cambridge University Press. 523). New York: Guilford Press. Carroll, J. B. (1998). Human cognitive abilities: A cri- Galton, F. (1888). Co-relations and their measurement, tique. In J. J. McArdle & R. W. Woodcock (Eds.), chiefl y from anthropometric data. Proceedings of the Human cognitive abilities in theory and practice Royal Society of London, 45 , 135–145. (pp. 5–24). Mahwah: Erlbaum. Glutting, J. J., Watkins, M. W., Konold, T. R., & Cattell, R. B. (1941). Some theoretical issues in adult McDermott, P. A. (2006). Distinctions without a dif- intelligence testing. Psychological Bulletin, 38 , 592. ference: The utility of observed versus latent factors Cattell, R. B. (1943). The measurement of adult intelli- from the WISC-IV in estimating reading and math gence. Psychological Bulletin, 40 , 153–193. achievement on the WIAT-II. Journal of Special Cattell, R. B. (1963). Theory for fl uid and crystallized Education, 40 , 103–114. intelligence: A critical experiment. Journal of Gottfredson, L. (1997). Why g matters: The complexity of Educational Psychology, 54 , 1–22. everyday life. Intelligence, 24 , 79–132. Cattell, R. B. (1971). Abilities: Their structure, growth, Guilford, J. P. (1967). The nature of human intelligence . and action . Boston: Houghton Miffl in. New York: McGraw-Hill. Cattell, R. B. (1987). Intelligence: Its structure, growth, Guilford, J. P. (1972). Thurstone’s primary mental abili- and action . New York: Elsevier. ties and structure-of-intellect abilities. Psychological Chapman, P. D. (1988). Schools as sorters: Lewis M. Bulletin, 77 (2), 129–143. Terman, applied psychology, and the intelligence test- Gulliksen, H. (1950). Theory of mental tests . New York: ing movement, 1890–1930 . New York: New York Wiley. University Press. Hale, J. B., & Fiorello, C. A. (2004). School neuropsy- Cohen, J. (1959). The factorial structure of the WISC at chology: A practitioner’s handbook . New York: ages 7–6, 10–6, and 13–6. Journal of Consulting Guilford Press. Psychology, 23 , 285–299. Halstead, W. C. (1947). Brain and intelligence: A quanti- Crawford, J. R., Garthwaite, P. H., Denham, K. A., & tative study of the frontal lobes. Chicago: University of Chelune, G. J. (2012). Using regression equations Chicago Press. built from summary data in the psychological assess- Hegarty, M. (2010). Components of spatial intelligence. In ment of the individual case: Extension to multiple B. H. Ross (Ed.), The psychology of learning and moti- regression. Psychological Assessment, 24 , 801–814. vation (Vol. 52, pp. 265–297). San Diego: Academic. doi: 10.1037/a0027699 . Holland, J. L. (1985). Making vocational choices: A the- Das, J. P., Naglieri, J. A., & Kirby, J. R. (1994). Assessment ory of vocational personalities and work environment . of cognitive processes: The PASS theory of intelli- Odessa: Psychological Assessment Resources. gence. Boston: Allyn and Bacon. 21 The Relationship Between Theories of Intelligence and Intelligence Tests 339

Horn, J. L. (1985). Remodeling old models of intelli- Lamiell, J. T. (2003). Beyond individual and group differ- gence. In B. B. Wolman (Ed.), Handbook of intelli- ences: Human individuality, scientifi c psychology, and gence (pp. 267–300). New York: Wiley. William Stern’s critical personalism . Thousand Oaks: Horn, J. L., & Cattell, R. B. (1966). Refi nement and test Sage. of the theory of fl uid and crystallized general Lohman, D. F. (2000). Complex information processing intelligences. Journal of Educational Psychology, 57 , and intelligence. In R. J. Sternberg (Ed.), Handbook of 253–270. intelligence (pp. 285–340). New York: Cambridge Horn, J. L., & Noll, J. (1997). Human cognitive capabili- University Press. ties: Gf-Gc theory. In D. P. Flanagan, J. L. Genshaft, & Luria, A. R. (1966). Human brain and psychological pro- P. L. Harrison (Eds.), Contemporary intellectual cesses . New York: Harper & Row. assessment: Theories, tests and issues (pp. 53–91). Luria, A. R. (1973). The working brain: An introduction New York: Guilford Press. to neuropsychology . New York: Basic Books. Horn, J. L. & Blankson, N. (2005) Foundations for better Major, J. T., Johnson, W., & Deary, I. J. (2012). Comparing understanding of cognitive abilities. In D. Flanagan & models of intelligence in Project TALENT: The VPR P. Harrison (Eds.) Contemporary Intellectual model fi ts better than the CHC and extended Gf–Gc Assessment: Theories, tests, and issues (2nd ed., pp. models. Intelligence, 40 (6), 543–559. 41–68). New York, NY: Guilford Press. McCloskey, G., Perkins, L. A., & Van Diviner, B. (2008). Hunt, E. (2011). Human intelligence . New York: Assessment and intervention for executive function Cambridge University Press. diffi culties . New York: Routledge. Hunter, J. E. (1986). Cognitive ability, cognitive aptitudes, McGrew, K. S. (1997). Analysis of the major intelligence job knowledge, and job performance. Journal of batteries according to a proposed comprehensive vocational behavior, 29 , 340–362. Gf-Gc framework. In D. P. Flanagan, J. L. Genshaft, & Jensen, A. R. (1974). Review of abilities: Their structure, P. L. Harrison (Eds.), Contemporary intellectual growth and action. American Journal of Psychology, assessment: Theories, tests, and issues (pp. 151–179). 87 , 290–296. New York: Guilford. Johnson, W., & Bouchard, T. J. (2005). The structure of McGrew, K. S. (2005). The Cattell-Horn-Carroll (CHC) human intelligence: It is verbal, perceptual, and image theory of cognitive abilities. Past, present and future. rotation (VPR), not fl uid and crystallized. Intelligence, In D. Flanagan, & Harrison (Eds.), Contemporary 33 (4), 393–416. intellectual assessment. Theories, tests, and issues Kamphaus, R. W., Winsor, A. P., Rowe, E. W., & Kim, S. (2nd ed., pp.136–181). New York. (2012). A history of intelligence test interpretation. In McGrew, K. S. (2009). Editorial: CHC theory and the D. P. Flanagan & P. L. Harrison (Eds.), Contemporary human cognitive abilities project: Standing on the intellectual assessment (3rd ed., pp. 56–70). New shoulders of the giants of psychometric intelligence York: Guilford. research. Intelligence, 37 , 1–10. Kane, M. J., Hambrick, D. Z., Tuholski, S. W., Wilhelm, Milberg, W. P., Hebben, N., & Kaplan, E. (1996). The O., Payne, T. W., & Engle, R. W. (2004). The general- Boston Process Approach to neuropsychological ity of working memory capacity: A latent variable assessment. In I. Grant & K. M. Adams (Eds.), approach to verbal and visuospatial memory span and Neuropsychological assessment of neuropsychiatric reasoning. Journal of Experimental Psychology: disorders (2nd ed., pp. 58–80). New York: Oxford General, 133 , 189–217. University Press. Kaufman, A. S., & Kaufman, N. L. (1983). KABC: Naglieri, J. A., & Das, J. P. (1997). Cognitive assessment Kaufman Assessment Battery for Children . Circle system . Itasca: Riverside. Pines, MN: American Guidance Service. Osgood, C. E. (1963). Psycholinguistics. In S. Koch (Ed.), Kaufman, A. S. (1979). Intelligent testing with the Psychology: A study of a science (Vol. 6, pp. 244– WISC-R (p. 1979). New York: Wiley-Interscience. 316). New York: McGraw-Hill. Kaufman, A. S. (2000). Intelligence tests and school psy- Rapaport, D., Gil, M., & Schafer, R. (1945–1946). chology: Predicting the future by studying the past. Diagnostic psychological testing (2 vols.) . Chicago: Psychology in the Schools, 37 , 7–16. Year Book Medical. Kaufman, A. S., & Kaufman, N. L. (2004). Kaufman Reynolds, M. R., Keith, T. Z., Flanagan, D. P., & Alfonso, Assessment Battery for Children-Second Edition V. C. (2013). A cross-battery, reference variable, con- (KABC-II) . Circle Pines: American Guidance Service. fi rmatory factor analytic investigation of the CHC tax- Kyllonen, P. C. (2002). Knowledge, speed strategies, or onomy. Journal of School Psychology, 51 , 535–555. working memory capacity? A systems perspective. In Sattler, J. M. (1974). Assessment of children’s intelli- R. J. Sternberg & E. L. Grigorenko (Eds.), The general gence. Philadelphia: Saunders. factor of intelligence: How general is it? (pp. 415– Samelson, F. (1977). World War I intelligence testing and 465). Mahwah: Erlbaum. the development of psychology. Journal of the History Kyllonen, P. C., Lohman, D. F., & Woltz, D. J. (1984). of the Behavioral Sciences, 13 , 274–282. Componential modeling of alternative strategies for Schmidt, F. L., & Hunter, J. (2004). General mental ability performing spatial tasks. Journal of Educational in the world of work: Occupational attainment and job Psychology, 76 , 1325–1345. 340 W.J. Schneider and D.P. Flanagan

performance. Journal of personality and social psy- Thurstone, L. L. (1947). Multiple factor analysis . chology, 86 (1), 162–173. Chicago: University of Chicago Press. Schneider, W. J. (2010). The compositator 1.0 . Olympia: Tuddenham, R. D. (1962). The nature and measurement WMF Press. of intelligence. In L. Postman (Ed.), Psychology in the Schneider, W. J. (2013a). Principles of assessment of apti- making: Histories of selected research problems tude and achievement. In D. Saklofske, C. Reynolds, (pp. 469–525). New York: Knopf. & V. Schwean (Eds.), Oxford handbook of psychologi- Unsworth, N., & Engle, R. W. (2007). The nature of indi- cal assessment of children and adolescents (pp. 286– vidual differences in working memory capacity: 330). New York: Oxford University Press. Active maintenance in primary memory and con- Schneider, W. J. (2013b). What if we took our models trolled search from secondary memory. Psychological seriously? Estimating latent scores in individuals. Review, 114 , 104–132. Journal of Psychoeducational Assessment, 31 , Unsworth, N., & McMillan, B. D. (2014). Trial-to-trial 186–201. fl uctuations in attentional state and their relation to Schneider, W. J., & McGrew, K. S. (2012). The Cattell- intelligence. Journal of Experimental Psychology: Horn-Carroll model of intelligence. In D. Flanagan & Learning, Memory, and Cognition . Advance online P. Harrison (Eds.), Contemporary intellectual assess- publication. doi: 10.1037/a0035544 . ment: Theories, tests, and issues (3rd ed., pp. 99–144). Vernon, P. E. (1961). The structure of human abilities New York: Guilford. (2nd ed.). London: Methuen. Schneider, W. J., & McGrew, K. S. (2013). Cognitive per- Wagner, R. K., & Torgesen, J. K. (1987). The nature of formance models: Individual differences in the ability phonological processing and its causal role in the to process information. In S. Ortiz & D. Flanagan acquisition of reading skills. Psychological Bulletin, (Sec. Eds.), Section 9: Assessment Theory, in B. Irby, 101 , 192–212. G. Brown, & R. Laro-Alecio, S. Jackson (Vol. Eds.), Wainer, H. (1976). Estimating coeffi cients in linear mod- Handbook of educational theories (pp. 767–782). els: It don’t make no nevermind. Psychological Charlotte: Information Age Publishing. Bulletin, 83 , 213–217. Spearman, C. (1904). General intelligence, objectively Watkins, M. W., Kush, J. C., & Glutting, J. J. (1997). determined and measured. American Journal of Discriminant and predictive validity of the WISC‐III Psychology, 15 , 201–293. ACID profi le among children with learning disabili- Spearman, C. (1915). The measurement of intelligence. ties. Psychology in the Schools, 34 , 309–319. The Eugenics Review, 6 , 312–313. Weiss, L. G., Keith, T. Z., Zhu, J., & Chen, H. (2013). Spearman, C. E. (1927). The abilities of man: Their nature WISC-IV and clinical validation of the four- and fi ve- and measurement . London: Macmillan. factor interpretative approaches. Journal of Spearman, C. (1930). Autobiography. In C. Murchison Psychoeducational Assessment, 31 , 114–131. (Ed.), A history of psychology in autobiography (Vol. Wilks, S. S. (1938). Weighting systems for linear func- 1, pp. 299–334). Worchester: Clark University Press. tions of correlated variables when there is no depen- Stern, W. (1900). Über Psychologie der individuellen dent variable. Psychometrika, 3 , 23–40. Differenzen (Ideen zu einer “Differentiellen Woodcock, R. W. (1990). Theoretical foundations of the Psychologie”) [On the psychology of individual differ- WJ-R measures of cognitive ability. Journal of ences (Toward a “differential psychology”)] . Leipzig: Psychoeducational Assessment, 8 , 231–258. Barth. Woodcock, R. W. (1993). An information processing view Thorndike, R. L., Hagen, E. P., & Sattler, J. M. (1986). of the Gf-Gc theory. In Journal of Psychoeducational Stanford-Binet intelligence scale (4th ed.). Chicago: Assessment Monograph Series: Woodcock-Johnson Riverside. Psycho-Educational Assessment Battery—Revised Thurstone, L. L. (1934). The vectors of mind. (pp. 80–102). Cordova: Psychoeducational Corporation. Psychological Review, 41 , 1–32. Woodcock, R. W., & Johnson, M. B. (1989). Woodcock– Thurstone, L. L. (1938). Primary mental abilities . Johnson psycho-educational battery–revised . Allen: Chicago: University of Chicago Press. Developmental Learning Materials. Intelligence and Culture: History and Assessment 22

Donald H. Saklofske , Fons J. R. van de Vijver , Thomas Oakland , Elias Mpofu , and Lisa A. Suzuki

Human behavior occurs in spatiotemporal and assessment of intelligence, including explaining sociohistorical contexts. This chapter discusses race and ethnic group differences in intelligence intelligence in light of some of its contextual fea- scores. The second section provides an overview tures, including the link between intelligence and of contemporary practices, in particular cultural culture. The issues are discussed in three parts: conceptions of intelligence (i.e., how intelligence historical infl uences, cultural issues in conceptual- is defi ned in different cultures, cultural issues in izing and assessing intelligence, and intelligence intelligence assessment) and trends in test use test adaptation and development in the interna- across cultures (e.g., which tests are used in vari- tional community. The fi rst section on historical ous cultures). The third section highlights issues of perspectives discusses early conceptions of the equivalence and bias in intelligence tests followed by procedures used to adapt tests or their administration to increase their cultural suitability and D. H. Saklofske (*) issues important to the translation and adaptation Department of Psychology , University of Western of intelligence tests. Ontario, Social Science Centre (7th fl oor) , London , ON N6A 5C2 , Canada e-mail: [email protected] F. J. R. van de Vijver Historical Infl uences in Intelligence Department of Culture Studies , Testing Tilburg University , P. O. Box 90153 , 5000 LE Tilburg , The Netherlands Both the construct of intelligence and its measure- e-mail: [email protected] ment predate the establishment of psychology as a T. Oakland scientifi c discipline and most likely have a history as University of Florida , 1921 SW 8th Drive , Gainesville , FL 32601 , USA long as human civilization. A search through the lexi- e-mail: Oakland@ufl .edu con of the world’s languages both past and present E. Mpofu will uncover words and descriptions that capture Faculty of Health Sciences , University of Sydney , individual differences in human behavior that refl ect Cumberland Campus Room T-428 , Lidcombe , intellectual and other cognitive abilities. The overlap NSW 2141 , Australia as well as the uniqueness in how intelligence has e-mail: [email protected] been and is described, defi ned, observed, measured, L. A. Suzuki , Ph.D. valued, and used by cultures, societies, and countries Department of Applied Psychology , New York University , New York , NY 10003 , USA is likely to be of interest to those interested in the e-mail: [email protected] concept and measurement of intelligence.

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 341 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_22, © Springer Science+Business Media New York 2015 342 D.H. Saklofske et al.

dawn of the twentieth century ushered in new Early Efforts to Defi ne and Measure measurement methods based on different theo- Intelligence retical views of intelligence (e.g., knowledge, comprehension, judgment, and reasoning) that The fi rst recorded evidence of test development were also sensitive to age differences. Alfred and use occurred in China approximately 3000 Binet (1857–1911), Victor Henri (1872–1940), years ago. The tests had a decidedly practical and Theodore Simon (1873–1961) led this effort purpose, namely, to help identify civil servants. after discarding tests of palmistry and head cir- Assessments included essays on civil law, mili- cumference. Binet and Simon’s tests, developed tary studies, agriculture, and geography. between 1904 and 1911, were intended to be Subsequent measures of visuospatial perception, used in Paris schools to help identify special creativity, problem solving, and divergent think- needs students. The content of test items was ing were used somewhat prominently in China. A modeled closely on school-like tasks and used for number of these qualities are consistent with cur- timely selection for remediation on the basis of a rently identifi ed components of intelligence. learners measured mental age. Empirical evi- Both the early Roman and Greek civilizations dence that the tests correlated with academic inferred differences among its citizens in what achievement added to their value (Wolf 1973 ). they knew and did. For example, ancient Greek Interest in ability assessment intensifi ed with the society focused on self-care and community widespread emergence of public education, fi rst engagement. People were thought to be of nor- in grammar schools (i.e., grades 1–6) and later mal intelligence if the level of these qualities secondary schools. Their emergence was in were similar to their peers and to display dimin- response to large increases in the number of stu- ished intelligence if they were similar to younger dents, many of whom came from families that persons (Oakland and Harrrison 2008 ). Aristotle’s lacked a history of formal education. Educators description of reason or the proposed brain- sought methods that provided information on stu- intelligence connections described by dent’s general academic aptitude. The large Hippocrates and Plato were forerunners of later increase in immigrants into the United States thought. during the nineteenth and early twentieth centu- In Europe during the Middle Ages, various ries gave rise to concerns that the country’s aver- “tests” comprised of a few simple questions were age level of intelligence was being lowered by used to determine the ability or “mentality” of their presence and further spurred the develop- persons. However, it was not until the nineteenth ment and use of intelligence tests for screening, century that the pioneering work of Sir Francis classifi cation, and selection purposes (Tulsky Galton (1822–1911), Wilhelm Wundt (1832– et al. 2003 ). In 1916, Lewis Terman (1877–1956) 1920 ), James McKeen Cattell (1860–1944), and coined the term intelligence quotient (IQ), calcu- others helped to establish the central role of intel- lated by dividing a person’s mental age by his or ligence within the context of the study of indi- her chronological age and multiplying this ratio vidual differences in psychology, including the by 100. Subsequent use of ability tests in the mil- effects of race and social class. Their contribu- itary during World War I seemed to secure a last- tions led to worldwide interest in both the con- ing place for intelligence testing (and the resulting cept and measurement of intelligence and to what IQ) in psychology, education, and the public may be termed “the testing enterprise.” mainstream. During these early years of psychology, defi - The rapid increase in the number of intelli- nitions of intelligence remained elusive and gave gence tests and their widespread use throughout rise to the development of many tests that later the twentieth century gave rise to further efforts to have been discarded as measures of intelligence develop more comprehensive theories and models (e.g., head circumference, visual and auditory that would defi ne the structure of intelligence and discrimination, grip strength). However, the in turn guide the measurement and interpretation 22 Intelligence and Culture: History and Assessment 343 of intelligence. Early models included those proposed by Spearman (i.e., general and specifi c Recognizing Individual and Group factors) and Thurstone (i.e., seven primary mental Differences in Intelligence abilities), and later, during the mid-twentieth century, models from Burt and Vernon (i.e., a hierar- Individual, group, and cultural variability have chical model that included general, group, been apparent from the earliest use of intelli- specifi c, and error factors), Jensen (i.e., level 1, gence tests, refl ected largely in mean score dif- memory, and level 2, mental manipulation or g ), ferences. For example, the US military developed Guilford (i.e., a structure of intellect comprising the Army Alpha and Beta tests to facilitate the 120–180 abilities), and Wechsler (i.e., verbal and assessment of recruits who were either literate or visual perceptual qualities that contribute to a lacking in language use and expression, a tradi- general IQ). Later twentieth-century models were tion carried on by Wechsler. As the use of intel- proposed by Gardner (i.e., multiple intelligences) ligence tests increased in other parts of the world and Sternberg (i.e., triarchic theory). Cattell and together with an increase in immigration, the Horn initially proposed two kinds of intelligence, issue of observed within- and between-group crystallized and fl uid intelligence, that later were differences came to the forefront. The major incorporated into the work by John Carroll, giving question was whether there were “real” differ- rise to CHC theory (named after Cattell, Horn, ences between the peoples of different countries, and Carroll) (i.e., nine broad stratum abilities and ethnicities, and cultures or whether the observed more than 70 narrow abilities). Other notable con- score differences were a function of the very tributions come from Europe (e.g., Das-Luria’s tests used to assess intelligence. In the mid- PASS model and Piaget’s stages of cognitive twentieth century, Raymond Cattell (1940 ) development) and Israel (Feuerstein’s instrumen- developed a “culture-free” (later referred to as a tal enrichment model). culture-fair) measure of intelligence. In the mid- While many defi nitions of intelligence remain 1970s, Mercer and Lewis ( 1977 ) published the in vogue, most include attention to abstract rea- System of Multicultural Pluralistic Assessment soning, problem solving, and the acquisition of in response to a need for more culturally fair knowledge (Snyderman and Rothman 1988 ). In methods of assessing children from different 1994 in response to controversies generated by cultural and ethnic backgrounds. Bracken and the publication of The Bell Curve: Intelligence McCallum’s ( 1998 ) Universal Nonverbal and Class Structure in American Life (Hernstein Intelligence Test is a recent attempt to minimize and Murray 1994 ), 52 experts on intelligence test content imbued by culture and language. endorsed a statement published in the Wall Street However, among measures of intelligence, the Journal that summarized for public understand- Wechsler scales are used most frequently to ing the scientifi c fi ndings regarding intelligence. assess both child and adult intelligence. Intelligence was defi ned as “a very general men- Test authors and publishers of well-known tal capability that, among other things, involves measures of intelligence along with others in the the ability to reason, plan, solve problems, think scientifi c community have been asked increas- abstractly, comprehend complex ideas, learn ingly to provide evidence of the extent to which quickly and learn from experience” (Gottfredson test content and results provide an accurate 1994, p. 14). This defi nition, along with the well- assessment when used outside of the context researched CHC model that has infl uenced many where it was developed and standardized. Even of the tests used in the Western world, has been in neighboring countries in which many persons marketed in the majority world.1 use the same language, such as the United States (US) and Canada, the question of “how well do intelligence tests travel?” was being raised about 1 We use this term to refer to developing or emerging non- Western countries that constitute the majority of human- tests such as the Wechsler intelligence tests that kind and defi ne community in terms of what it is. were developed in the United States but used 344 D.H. Saklofske et al. extensively in Canadian schools. This spurred the differences between Black and White 12-year- olds, search for answers to whether intelligence is a in favor of Whites, dropped from 15 points to 9.5 universal construct, whether it varies across points over the past three decades (Nisbett 2009 ). groups, countries, and cultures, and how to most Other research found similar results for a broader accurately and fairly measure it. age group (e.g., Edwards and Oakland 2006 ; Weiss In the United States, the validity of intelligence et al. 2010 ). Some believe the performance gaps on tests when used with persons from minority tests of intelligence for different races depend upon groups, specifi cally Blacks and Hispanics from the complexity of the task (Flynn 2012 ; Jensen lower socioeconomic levels, has been questioned 1972 , 1973 , 1980 ). For example, subtests with for at least 100 years. The accusation of bias gen- greater g loadings (i.e., greater complexity) yield erally is based on an incorrect, outdated model of larger group differences. However, more complex test bias, one that highlights mean score differ- test items generally require more implicit and ences. Modern defi nitions of test bias do consider explicit knowledge of the culture of the test (i.e., on differences in mean scores as prima fascia evi- tests that have higher cultural loadings) (Helms- dence of test bias. Nevertheless, these allegations Lorenz, Van de Vijver, and Poortinga 2003 ). As a gave rise to concerns by the public, educational consequence, the performance difference may be institutions, and psychologists as to whether intel- infl uenced more by knowledge of the culture ligence tests are fair and unbiased when used with within which the test was developed than in the minority groups, especially Blacks and English- complexity of qualities assessed by the test. language learners (Oakland 1977 ). Following Moreover, other data indicate scores by Blacks challenges in the courts of both law and public increased on a variety of intelligence subtests (i.e., opinion, the prevailing conclusion among scholars Wechsler), including those with high cognitive was that intelligence tests are “…not culturally complexity (Weiss et al. 2010 ). biased against American Blacks or other native Although intelligence and culture are linked born, English speaking peoples in the U.S.” and closely, many aspects of cognitive functioning are “…members of all racial-ethnic groups can be universal (Lonner 1980 , 2011 ). For example, most found at every IQ level” (Gottfredson 1994 , p. 14). individuals are unlikely to survive if they had not While group differences in intelligence initially acquired basic cognitive skills (e.g., Piagetian con- were attributed to genetic and other familial infl u- servation skills). However, results from cross-cul- ences, environmental and cultural infl uences also tural research on intellectual assessment also make were recognized as being an important determi- clear that these basic skills often are diffi cult to nant of score variability. The debate regarding the identify, notably in non- Western cultures. Thus, degree to which cultural, genetic, and environmen- interpretations of cross- cultural differences in tal factors account for racial group differences in intellectual skills remain controversial and present intelligence continues (e.g., Rushton and Jensen complex challenges to test developers, research- 2005 ). Studies, however, continue to support the ers, and practicing psychologists. Genetic and cul- malleability of intelligence based upon environ- tural differences together with assessment issues mental and cultural factors. For example, a study cannot be overlooked. of group differences on the Wechsler tests of intelligence provided a relatively unbiased measure of intelligence for African Americans, Caucasians, Cultural Issues in Conceptualizing and Hispanics in the United States. Demographic and Assessing Intelligence qualities, including parents’ education and parental expectations for their children and education, Contemporary Practices accounted for much of the between-group differences (Prifi tera et al. 2008 ; Weiss et al. 2010 ). Despite spectacular advances in assessment tech- The IQ gap between Blacks and Whites has nar- nology within Western societies, recognition of rowed somewhat in recent years. For example, IQ intellectual abilities that matter to the majority 22 Intelligence and Culture: History and Assessment 345 non-Western world and ways to measure them (Sternberg 1985). Implicit theories about remain underdeveloped (Mpofu and Ortiz 2009 ; behaviors are those that people use to guide their Serpell and Haynes 2004 ). A reason for this lag conduct in a broad range of situations, typically in development is due, in part, to the relatively representing shared underlying meanings and limited research investment in understanding the social judgments they make on their and others’ nature and types of abilities that are valued in behavior. An understanding of the implicit theo- these settings. Additionally, there has been a ries of ability held by those engaged in a culture long-standing belief that concepts of intelligence provides a productive way to begin the journey that emanate from Western views are transport- toward understanding and appreciating the diver- able to non-Western settings, sometimes referred sity in human abilities that are valued. to as an “etic” perspective. While this view Knowledge of implicit theories of intelligence underlies efforts to discover and map universal held by a cultural community is prerequisite to traits in intellectual abilities, there is no cross- constructing explicit (or formal) theories for the cultural consensus about the nature of human design and use of appropriate assessments for intellectual abilities and how these should be rec- such abilities. As an example, Sternberg et al. ognized or measured (Marfo 2011 ; Mpofu and (2001 ) observed tacit abilities to be highly valued Nyanungo 1998 ; Pence 2011; Serpell 2011 ). in a Kenyan indigenous Luo cultural-linguistic Thus, the results of studies that examine the community. On the basis of that understanding, transportability of Western developed tests to the authors developed the test of tacit knowledge emerging countries indicate their limited incre- for natural herbal medicines for use with Luo mental value and that their results often merely community children to assess their knowledge of highlight mean score differences between common illnesses and standard herbal treatments Western and non-Western persons rather than for those illnesses in that community. The test is shed light on the nature of intellectual abilities of designed to access children’s acquired knowl- persons in non-Western cultures. This latter emic edge about the illnesses and herbal treatment perspective is illustrated in the following regimens through their experience and informal sections. observation. The test presentation format uses 22 stories that require children to identify an illness that a story depicted and the appropriate herbal Cultural Values and Human Abilities treatment for it. Storytelling and interpretations are valued procedures to tap into tacit knowledge Cultural values play a signifi cant role in the held by those living in sub-Saharan African ways human abilities are recognized, displayed communities. in context, and measured. The meanings that a culture assigns to displayed abilities and other Structure and Function of Intelligence. An behaviors infl uence the behaviors that defi ne examination of both ancient and modern views intelligence in those settings. An understanding from across the globe immediately shows that of the important intellectual abilities in a cultural there is not a universal consensus about what setting requires the framing of the abilities from intelligence is, what causes it, and how to best the perspective of those who display and fi nd measure it. Past and contemporary views of meaning in a particular context (Nsamenang intelligence in China refl ect some major 2006 ; Mpofu et al. in press ; Serpell 2008 , 2011 ). differences from Western perspectives (Yan et al. This includes learning and knowing the cultural 2009 ). For example, key positive human qualities contexts that defi ne those abilities and the typi- are more akin to linking Western descriptions of cal indicators of them. intelligence with personality and conative factors Concepts of intelligence in the non-Western (Yan and Saklofske 2004). Descriptions of world are represented best by implicit theories intelligence in majority world settings may be about abilities that the culture prioritizes less about the “what” (structure) than the “how” 346 D.H. Saklofske et al.

(function) of abilities, and of course, what a culture and group participation in culturally signifi cant values. This may be explained, in part, by daily activities. For example, a person’s prioritizing lived (and thus a personal understanding intelligence could be described on the basis of his of practical-oriented) intelligence rather than its or her abilities to achieve practical outcomes for hypothesized conceptual structure. For example, oneself and important others (e.g., family or clan). an investigation of concepts of intelligence among To illustrate, Ugandan Baganda and Batoro the Luo of Kenya identifi ed four terms that referred villagers regard intelligence as socially oriented to intelligence: rieko , luoro , winjo, and paro behavior that benefi ts the social collective (Wober (Grigorenko et al. 2001 ). Rieko refers to the 1974). This view is consistent with current context- and task-specifi c knowledge or skill; thus, concepts of social and emotional intelligence school-related rieko refers to ability to perform even if various cognitive capacities may be school-related tasks. Rieko mzungu refers to underlying factors. competence in white man’s (mzungu) technology. Observing the social responsiveness of A person’s luoro is expressed by his or her ability actions, processes, and outcomes across cultures to display social responsibility by being considerate provides further indications that intelligence is of others, observant of appropriate social etiquette, viewed differently from current Western views and willing to share resources. Winjo refers to that are infl uenced heavily by psychology and behaviors that display appropriate deference to captured in the tests used to assess intelligence. adults, the elderly, and authority fi gures. Paro In many non-Western cultures, the timeliness of a refers to one’s ability to be innovative and creative selected course of action (rather than the speed of and to complete tasks in a timely fashion. action), its social or collective intent (protecting Intellectual abilities underlying luoro and winjo the honor and social respect of others by sharing would be evident in the quality of a person’s social the credit for a job well done with them), and rec- participation, whereas those for reiko and paro ognition of history and traditions are qualities would be evident from the effi ciency in completing that defi ne those with high intelligence. All per- tasks. Some studies (e.g., Grigorenko et al. 2001 ; sons are presumed to be intellectually competent Serpell 2008 , 2011 ) reported performance ratings unless there is compelling and consistent evi- on contrived Western measures of intellectual dence to the contrary. This allows for a high lati- ability that poorly predict success on culturally tude of “ability appreciation” and a discounting embedded intelligence measures. In non-Western process that can be applied whereby behaviors culture settings, higher levels of intelligence often that are less than expected are judged and are aligned with practical or functional intelligence explained by others in light of the context in (Sternberg 1999 ) together with social emotional which the behavior occurred rather than as a and sensitivity (Mayer et al. 2008 ). refl ection of the person’s ability. Measures of wisdom and the ability to associate appropriate Qualities of Intelligence in Majority Culture social worth to actions as well as to decisions of Settings. Six qualities seemingly defi ne the one’s self and others (Sternberg and Jordan 2005 ) intelle ctual functioning in non-Western, majority would appear to be more akin to intelligence in world contexts: interpersonal relationships, these contexts. socially oriented planning, decision-making and problem solving, socially responsible resource management and utilization, cultural historicity and Intelligence Tests and Testing acknowledging highly regarded others who took similar actions, and social worth and productivity The above discussion emphasizes that, while (Dasen 2011 ; Mpofu 2004 ; Mpofu et al. 2007 ; intelligence is widely considered a cornerstone Mpofu et al. in press ). In these settings, the activi- in describing individual and groups differences ties, domains, and contexts for the valid appraisal (and similarities), how intelligence is defi ned of intellectual competence characterize individual and assessed varies across time and cultures. 22 Intelligence and Culture: History and Assessment 347

While such diversity is both acknowledged and psychology, especially school and clinical psy- respected, the Western perspective on intelli- chology, as well as industrial/organizational psy- gence appears to be most widespread and grow- chology especially related to human resource ing. The dominance of the Western perspective in development. studies and models of intelligence has strong his- The steady development of testing resources torical and intellectual roots as well as a strong during the fi rst third of the twentieth century war- economical component. Western developed intel- ranted the emergence of the Mental Measurements ligence tests are used on a massive scale, not just Yearbook that both catalogued and reviewed in Western countries. Therefore, developments in commercially available standardized tests. Later, international testing constitute another place Tests in Print emerged, thus allowing the Mental where culture and intelligence meet. Measurements Yearbook to focus on test reviews. It lists virtually all English-language standard- Setting the Stage. The proliferation of measures ized tests. Information in Tests in Print can pro- to assess intelligence and the importance of tests vide some insights into market-driven needs. The and testing in psychological science and practice 3,003 tests referenced in the eighth edition pub- warrant further discussion before examining lished in 2011 are divided into 20 categories, thus intelligence tests in the context of culture. Test use providing a partial summary of market-driven constitutes the dominant method of measurement forces that help identify tests that may be used in psychology. While varying methods of and thus what concepts command the greatest measuring (e.g., observation, interview) the public and professional interest (Murphy et al. plethora of identifi ed psychological factors (e.g., 2011 ). Among these tests, 21 % assess personal- extraversion, anxiety, resilience, self-concept, and ity, 20 % assess vocations, and 7 % assess intel- emotional intelligence), standardized tests are ligence. Some concepts of personality include likely to be the most common methods and, as intelligence as an important component. discussed above, have their roots in earlier tests Moreover, psychological assessment as practiced developed by Galton, Cattell, and Binet. A test by clinical, school, industrial/organizational, and consists of a standard set of questions or items counseling psychologists as well as neuropsy- designed to assess knowledge, skills, interests, or chologists may include an assessment of intelli- other characteristics of an examinee (VandenBos gence. Thus, this information suggests 2007 ). Minimally, tests should display key applications of the concept of intelligence psychometric qualities, including reliability and through test use are important to psychologists. validity, and result in providing practical ways to assess desired constructs (see Saklofske et al. Test Use Internationally. International surveys 2013). Knowledge that “psychological test of test development and use with persons of all validity is strong and compelling” and that ages have not been conducted, although recent “psychological test validity is comparable to publications such as Pruebas Publicadas en medical test validity” is reassuring (Meyer et al. Española: An Index of Spanish Tests in Print 2001, p. 128). Many psychologists would contend provide an indication of the growth of tests and that psychological tests constitute the fl agship of testing in other countries. However, our applied psychology with standardized intelligence understanding of the international status of test testing as its most widely used invention. development and use with persons from birth to Earlier in our history, most psychological tests death is dependent somewhat on anecdotal were developed in universities to support evidence. Besides psychologists, various other research. Although this practice continues, it now professions use measures of cognitive abilities is overshadowed by testing companies working including psychiatrists, educators, counselors, in partnership with test authors to develop tests and speech pathologists. Attention to global test for clinical and other applied markets. Test use is use also has been addressed by the International the strongest in the applied and practice areas of Test Commission, in part through its Guidelines 348 D.H. Saklofske et al. for Translating and Adapting Tests (see www. Table 22.1 Tests used with children and youth intestcom.org for the 2010 updated version) in Top ten tests in 1990 check date, is it 88, 89 or 90??? terms of context, test development and Wechsler intelligence scales for children adaptation, administration, and documentation/ Ravens Progressive Matrices score interpretation. Bender Gestalt Test Rorschach Inkblot Test Test Use Internationally with Children and Stanford-Binet intelligence scales Youth. Two surveys of test development and use Wechsler adult intelligence scales with children and youth provide some indications Thematic Apperception Test of international test use, including measures of Differential Aptitude Test intelligence, with children and youth (Table 22.1 ). Minnesota Multiphasic Personality Inventory Frostig Developmental Test of Visual Perception The 1989 survey acquired information from Top ten tests in 2012 school psychologists in 44 countries (Hu and Wechsler intelligence scales for children Oakland 1991; Oakland and Hu 1992). The 2012 Ravens Progressive Matrices survey acquired information from school Bender Gestalt Test psychologists in 77 countries. Wechsler Preschool and Primary Scale of Intelligence Kaufman Assessment Battery for Children Summary of the 1989 Survey Data. Of the ten Child Behavior Checklist tests identifi ed in terms of popular usage, fi ve Wide Range Achievement Test were measures of intelligence (Table 22.1 ) while Children’s Memory Scale the remaining tests assessed visuomotor skills, Children’s Apperception Test personality, and psychopathology. Additionally Wechsler Individual Achievement Test some countries reported using the Frostig Developmental Test of Visual Perception to provide a culture-fair assessment of intelligence. Thus, at that time, school psychologists in 3–5, and over 18 years; as expected, tests were international settings used intelligence tests least often used by school psychologists in the somewhat prominently. assessment of infants.

Summary of the 2012 Survey Data. Respon- Implications of the Survey Data. Assuming dents to the more recent survey identifi ed 606 that test use is market driven, we could conclude tests. Their rank ordered areas of focus were as the assessment of intelligence constitutes a follows: intelligence, achievement, language, steadfast public interest and remains a common personality, visuomotor qualities, school readi- feature of the work of psychologists inter- ness, social and emotional qualities, motor skills, nationally, especially school psychologists. and adaptive behavior. Again, among the top During the 23 years covered by these survey ten tests cited among the 77 reporting countries, data, the Wechsler scales for assessing children’s six measure intelligence. The remaining tests intelligence, the Raven’s Progressive Matrices, assessed visuomotor skills, psychopathology, and the Bender Gestalt Test have maintained achievement, and personality. their prominent ranking. The Wechsler scales When asked to indicate the purposes of using are used in all reporting countries. However, 74 tests, the respondents identifi ed diagnosis as countries use older versions of this test than the being most prominent followed by attempts to current WISC-IV. The WISC-V is projected to acquire a better understanding of children’s be published in 2014. The more recent versions achievement, make admissions decisions, sup- of the Wechsler Preschool and Primary Scale port guidance services, estimate academic apti- of Intelligence and the Kaufman Assessment tude, and assist in vocational selection. Test use Battery for Children replaced the previously by school psychologists is most common among revered Stanford-Binet as well as the Frostig persons ages 6–12 years followed by ages 13–18, measure. Two projective-type measures of 22 Intelligence and Culture: History and Assessment 349 personality (e.g., Rorschach Inkblot Test and the transporting of tests from developed to the Thematic Apperception Test) along with the majority world likely occurs through the efforts Minnesota Multiphasic Personality Test dropped of psychologists, often trained abroad, who off the earlier list in lieu of greater use of the return to their developing country and attempt measures of children’s psychopathology, to use or even replicate some of the major tests memory, and achievement seen in the 2012 data. found commonly in the practices of the more In much of the Western world, the CHC developed countries—principally those devel- model of intelligence (McGrew and Flanagan oped in the United States. 1998) provides one of the leading descriptions Test-related conditions often differ in developed of intelligence and often is used to guide the and emerging countries. Emerging countries tend development or revision of commonly used tests to use foreign-developed tests that are older ver- such as the Woodcock-Johnson III Tests of sions, have outdated norms, and are not standard- Cognitive Abilities—a measure that is being ized for use in their countries. They also are more used increasingly in the United States even likely to use fewer tests for multiple purposes. For though it is not included in the 2012 listing. example, in one country, the Wechsler scale is used However, none of the tests listed in Table 22.1 to assess children‘s school readiness, intelligence, are explicitly grounded in CHC theory. Thus, academic achievement, speech/language, social/ current tests seemingly underrepresent the com- emotional development, adaptive behavior, motor plexity of the concept of intelligence, at least as development, and neuropsychological qualities. refl ected by the CHC model. However, clini- School psychologists in emerging countries are cians can use a cross-battery approach to use likely to use fewer tests that assess personality various measures and then to recast their scores measures and, if used, to rely on interpretations to be more consistent with this model (Flanagan gleaned from theory, not from the use of norms. et al. 2013 ). The Wechsler scales assess fi ve Conditions for test development differ major factors that line up with CHC model with- between developed and developing countries. In out trying to replicate it (Weiss et al. 2013a , b ; general, test use is strongest in Australia, see Journal of Psychoeducational Assessment Canada, most Western European countries, and Special Issue, 2013). the United States. These countries display One may wonder why intelligence may be strong beliefs in science, technology, individual valued so highly and measured so widely com- differences, and meritocracy (Oakland 2009 ). In pared to measures that assess other important contrast, both test development and use are psychological constructs and factors. Although lower in Asia, Eastern Europe, and Latin we are not suggesting that measures of intelli- America, among the 22 Arab countries and, gence are overused or do not provide important with the exception of South Africa, lowest in the or relevant information, we acknowledge other 54 African countries. Compared to developed rich and important testing resources that, if and countries, those in the majority world may have when used properly, could enhance our under- smaller populations and thus smaller test mar- standing and promotion of child growth and kets, fewer test development companies, thus development as well as address issues impor- less fi nancial support for test development, tant for understanding infants and adults. together with lower professional standards for Furthermore, with the exception of Mpofu’s those who use tests. They also have fewer uni- important and pioneering teamwork in versity graduate programs that develop skills Botswana (Mpofu et al. 2012; Mpofu et al. in important to test development, fewer scholarly press), ultimate consumers of test results (e.g., journals that feature assessment and test-related children and youth, their parents and teachers) issues, and fewer professional associations that rarely are asked to identify the information they advocate for high legal and ethical standards for would like to obtain through tests. Instead, the test development and use. 350 D.H. Saklofske et al.

(Trimble 2010 , p. 316). Cultural equivalence, a Intelligence Test Adaptation higher-order form of equivalence, is dependent and Development on measures that meet specifi c criteria that indicate a measure is appropriate for possible use Equivalence and Bias in Intelligence with other cultural groups beyond the one for Testing which it originally was developed. Upwards of 50 or more types of equivalence may impact Based on the discussion above, one may presume interpretive and procedural practices in order to that intelligence is understood best in the context establish cultural equivalence (Trimble 2010 ). of culture and the identifi cation of “universals” Various forms of equivalences may need to be across groups (e.g., common factors such as known in order to use a measure of intelligence problem solving, reasoning ability). Thus, we cross-nationally with confi dence (Butcher and should strive to accurately measure and under- Han 1996 ; Helms 1992; Lonner 1985; van de stand universal traits within the cultures in which Vijver 2001 ). These include (a) functional (i.e., they are displayed and whether the traits are cul- whether the construct of IQ as operationally turally meaningful and important. For example, defi ned occurs with equal frequency across the factor structure of the WISC-III was found to groups), (b) conceptual (i.e., item information is be remarkably robust across more than a dozen familiar across groups and the construct of intel- developed countries representing Asia, Europe, ligence means the same thing in various cultures), and North America following careful translation (c) linguistic (i.e., language used has similar of test items or modifi cations of content defi ning meaning across groups), (d) metric (i.e., a scale the factors and subtests (Georgas et al. 2003 ). measures the same behavioral qualities or char- Other evidence suggests that the structure of acteristics and the measure has similar psycho- intelligence is invariant across cultures (Van de metric properties in different cultures), and (e) Vijver 1997 ). However, as we can infer from the scalar (i.e., score differences refl ect the same above discussion, instruments developed for and degree, intensity, or magnitude for different cul- validated in and for one cultural context may not tural groups). Scalar equivalence is established retain their meaning when used in other cultural only when mean differences in scores can be contexts; their meaning and usefulness need to be attributed to genuine cultural differences and not demonstrated empirically. While we agree that to any form of bias (Butcher and Han 1996 ). how intelligence is viewed, manifest, valued, and Thus, evidence is needed that the measure is assessed within and between cultures may vary, operating appropriately in various cultural con- we also contend that there are “universal” or texts. Most measures of intelligence and indeed common aspects that can be described and in turn most psychological measures have not attained used in a sensible, ethical, and culturally sensi- scalar equivalence across diverse cultural groups. tive way with and for the benefi t of the individual Thus, the use of measures of intelligence and the and larger group. The discussion to follow out- meaning of racial and ethnic group differences in lines the rigorous methodology needed to achieve scores obtained on these measures remain some- this outcome. what controversial. Understanding equivalence is imperative in Test developers address issues of equivalence the assessment of intelligence as it addresses through various procedures. They include expert whether the procedures and formats of the mea- panel reviews (i.e., professionals review item sure, results, and interpretations are equally content and provide informed judgments regard- familiar and relevant across diverse populations. ing potential biases), examination of differential Cultural equivalence refers to whether “…inter- item functioning (DIF) between groups, statisti- pretations of psychological measurements, cal procedures that compare psychometric fea- assessments, and observations are similar if not tures of the test (e.g., reliability coeffi cients) based equal across different ethnocultural populations” upon different population samples, exploratory 22 Intelligence and Culture: History and Assessment 351 and confi rmatory factor analysis and structural are unattainable, some instruments reduce the equation modeling (i.e., examination of the simi- impact of culture (i.e., culture-reduced tests). larities and differences of the constructs struc- Ironically, they do not differ appreciably from ture), and mean score differences while taking Cattell’s original measure. Importantly, Jensen into consideration the spread of scores within proposed to combine adequate test design with particular racial and ethnic groups as well as strict psychometric criteria for assessing the between groups. cultural appropriateness of instruments. In State-of-the-art measures of intelligence may contrast, Cattell seemingly believed that careful use expert review panels, racial and ethnic overs- test design alone would be able to resolve the ampling, and inclusion of statistical practices that problem of cultural bias in psychological address multiple forms of equivalence through assessment. the use of various statistical procedures. Despite During the last decades, bias in intelligence sophisticated efforts, problems may remain. For tests has been addressed from three perspectives: example, the Wechsler scales have been trans- conceptual advances in theories of intelligence, lated and renormed in several countries around improved and more abundant statistical methods, the world (e.g., Georgas et al. 2003 ). The reliabil- and conceptual advances in test adaptations. The ity and validity of the measures have been exam- CHC model of intelligence has gained wide ined in their new cultural contexts, and statistical acceptance in intelligence research, including procedures were conducted to establish linguis- cross-cultural intelligence research. Its accep- tic, structural, and conceptual forms of tance is fairly widespread (Carroll 1993 ; McGrew equivalence. However, concerns continue to arise 2005), and the model has demonstrated a robust- as clinicians struggle to apply the measure in ness that seems to show it is applicable univer- diverse cultural contexts. For example, even with sally (Berry et al. 2011 ; Van de Vijver 1997 ). procedures deemed appropriate to establish lin- Evidence for this universality comes from numer- guistic equivalence (van de Vijver 2001 ), a ous cross-cultural studies in which factors under- Spanish version of the measure developed in a lying intelligence batteries were compared. Spanish- speaking country may not be suitable for Although most research underlying this model use with Spanish-speaking individuals in the was conducted before the CHC theory was for- United States. Language alone poses a signifi cant mulated, much evidence suggests that its funda- obstacle given vast number of languages cur- mental features of intelligence (e.g., fl uid and rently spoken in the United States. There are cur- crystallized intelligence) can be found in all cul- rently 6,909 languages spoken worldwide and tures. However, this universality should not be 381 coded in the US Census (Ryan 2013 ). taken to imply that all measures of intelligence work equally well in all cultures. Bias in Intelligence Tests. Results from cross- Improved statistical methods are available for cultural studies of intelligence raised concern use in cross-cultural research. For example, struc- about the suitability of such tests when used with tural equation modeling has become popular. persons in a different and unfamiliar culture or Considerable toolsets exist to examine whether with those unfamiliar with the source test’s items, subtests, and tests can be compared across language. The Army Beta was the fi rst test to cultures. Important conceptual advances in test remove the effects of language on the assessment adaptations are reviewed below in considerable of intelligence. As noted above, Cattell (1940 ) detail. These developments, taken together, indi- fi rst attempted to develop a measure free from cate that much progress has been made since cultural bias. Cattell’s seminal work on culture- Cattell formulated the problem of cultural bias in free intelligence tests has led to various intelligence. In addition, few cross-cultural refi nements that interestingly became less researchers believe that our best intelligence mea- ambitious over time. More recently, Jensen sures are culture-free. However, researchers often ( 1980 ) proposed that, although culture-free tests mistakenly believe that adequate test design and 352 D.H. Saklofske et al. psychometric evidence on intelligence tests exam- the cognitive processes and structures may not be ined cross-culturally and cross-nationally alone universal or that problems exist with the cultural are crucial to attempts to understand and reduce suitability of one or more subtests. cultural bias (Mpofu and Ortiz 2009 ). The history of cross-cultural psychology is The term bias is the generic term to explain replete with examples. For example, Porteus cross-cultural differences (Van de Vijver 2003 ; ( 1937) administered his maze test, a paper-and- Van de Vijver and Leung 1997 ). The term refers pencil test in which the examinee must trace a to a problem common in cross-cultural studies way out of a schematically drawn maze, to vari- that data from tests of intelligence and other traits ous cultural groups, including the Khoisan in the obtained from different groups may not refl ect Kalahari Desert. His participants were not used the groups’ traits accurately. If scores are biased, to working with drawings on paper had some dif- their psychological meaning is unique to them fi culty completing the tasks. Porteus concluded (i.e., group dependent). Group differences in that their spatial skills were poorly developed, assessment outcomes may be accounted for, at not realizing that his conclusion contradicted by least to some extent, by other psychological con- their display of the excellent tracking skills in the structs or measurement artifacts (e.g., stimulus relatively cueless Kalahari desert. familiarity). Bias is not an intrinsic property of an Method bias can result from sample incompa- instrument but arises when a particular instru- rability, instrument characteristics, tester effects, ment is used to compare an individual to a larger and the method (mode) of administration. Stimulus group or make between-group comparisons. An familiarity (i.e., knowing the materials used in the instrument that works well in two groups may not test such as pictures) provides a common example work well in a third group. Some instruments are in intelligence tests. Tests administered to children more susceptible than others to bias. For exam- from different cultures are unlikely to have stimuli ple, an examination of the Georgas et al. (2003 ) that are equally familiar to all children. For exam- cross-cultural study of the WISC-III showed that ple, Cattell originally thought that basic geometric the greatest number of changes were made in the shapes (e.g., squares, triangles, and lines) would subtests that comprised the verbal comprehen- be equally familiar to children in all cultures. This sion factor. In contrast, the Raven’s Progressive assumption is incorrect. For example, a child’s Matrices often are considered to be among the early social and environmental contexts can infl u- least culturally biased measures of intelligence ence exposure to geometric shapes. Although and one of the highest in terms of g loading. some problems associated with the lack of familiarity may be overcome by test instructions or pro- Sources of Bias. There are three possible bias viding additional examples prior to commencing sources in cross-cultural research: construct bias, testing, one should not assume that pervasive cul- method bias, and item bias. tural differences in antecedent experiences can be Construct bias occurs when the construct mea- undone through these methods. sured is not identical across groups. For example, Item bias or differential item functioning can this may occur when data suggest the very basis assist in detecting bias due to item level anoma- of intelligence differs for people from, say, the lies (e.g., inadequate item translations). An item People’s Republic of China and Zimbabwe or is biased if persons with the same standing on the Canada and Paraguay. This relativistic viewpoint underlying construct (i.e., they are equally intel- lacks evidence. Another more likely example ligent) who come from different cultural groups occurs when data from an intelligence test admin- do not display the same expected score on an istered in two countries display different factor item. If a geography test administered to pupils in structures. Similarity of factor structures provides Germany and Australia contains the item “What evidence for the identity of underlying structures. is the capital of Germany?” the item would be Thus, when factor structures of test data obtained biased in favor of German students as they can be in different cultures differ, one may conclude that expected to display higher scores on the item 22 Intelligence and Culture: History and Assessment 353 than Australian students when pupils with the than a linguistic exercise. Adapting a test for use same total test score are compared. in another culture that uses a different language Construct and item bias has been studied most requires a combination of linguistic, cultural, and frequently. The reason for this interest is readily psychometric expertise. Thus, while linguistic understandable. Unlike method bias that usually skills remain important, experts with cultural and can be assessed only by adding items or subtests to psychometric knowledge also are needed. For an existing test battery, the statistical analysis of example, the item “Name the president of the construct and item bias is more straightforward and United States” may not be diffi cult to translate does not require additional data. Thus, method bias into other languages. However, we fi rst need to often is a neglected source of bias in cross-cultural know the purpose of the item. Is it to test the test use. Very few cross-cultural studies address examinee’s knowledge of a specifi c person or is it confounding sample differences. For example, intended to assess the name of one of the best- such confounding differences may include previ- known persons in a country? If the purpose is the ous test exposure, educational quality and opportu- latter, then an equivalent item in French could ask nity, and familiarity with the test’s tasks. examinees to name the French president. If this latter item were used in France, one needs to determine whether the American and French Translating and Adapting presidents are equally known in their countries. Intelligence Tests The psychometrically important question is whether the item will perform in the same man- Adaptation, Adoption, and Assembly. ner in the two languages. This example is Conceptual advances in test translation have shifted intended to illustrate that translating a test often signifi cantly during the last decades. Early test requires a combination of professional skills, translations focused largely on changing a test’s often provided by a group of persons, not a single language from the original source test to a target person. test. Decades later, the importance of examining Thus, the translation of an item (or test) can the quality of the translations became more use three methods: adoption, adaptation, and apparent. The introduction of back translation assembly. A “close” translation may yield an procedures marked an important hallmark (Brislin adequate version in the target language (i.e., an 1970). This procedure translates an instrument adoption method; Van de Vijver 2003 ; Van de from its source language to a target language and Vijver and Poortinga 2005 ). Secondly, an item then back from the target test to the original source (or test) may need some modifi cation to make it language (i.e., a forward and then a backward suitable in the target language (i.e., an adaptation translation). Then, the original instrument and the method). For example, the term dollars should be back-translated version are compared. Similarity converted to euros and miles to kilometers . between these two versions is seen as support for Thirdly, an item may be entirely inappropriate in the quality of the translation. This procedure works the target culture. For example, a Dutch intelli- well when all items can be translated closely in gence test has an item about bacon. This item is relation to language content and meaning (e.g., inappropriate for Muslim immigrant children in simple biographical questions can be checked the Netherlands, given the food taboo on pork adequately using this procedure). However, the meat in this group. In such a case, an entirely dif- procedure does not work well when an instrument ferent item needs to be constructed (i.e., an has idiomatic expressions or has other linguistic assembly method). Test translations typically features that are diffi cult to translate (e.g., words have relied on adoption methods. The term adap- such as him or her are not used in the same way tation (e.g., changing parts of the stimulus or across languages). instrument) currently is a generic term used for Dissatisfaction with translation procedures led translations. Assembly methods lead to more sig- to the recognition that a test translation is more nifi cant changes and may result in constructing a 354 D.H. Saklofske et al. new instrument. The above example about the customs, or practices. For example, the transla- maze test may illustrate the use of this method. tion of the English you requires cultural knowl- The measurement of special skills among edge about appropriate modes of address in tests Khoisan through a paper-and-pencil version of when languages distinguish formal and informal mazes test is not appropriate as it has such low addresses (e.g., vous and tu in French). The dis- ecological validity. Therefore, a completely dif- tinction can be simple yet important. For exam- ferent instrument needs to be constructed to mea- ple, Dutch also draws a distinction between these sure their spatial skills. two modes, but the conditions in which the two Thus, although close translation methods have forms are used are not entirely identical in French been relied on somewhat exclusively, current and Dutch. Thus, one would use the informal efforts focus more on preparing test instruments address in a Dutch test for students and a formal that are suitable in their new cultural context. address in a test for adults where the French ver- Adaptations and assembly methods are preferred sion would use the formal address for both when close translation methods are unsuitable. populations. Theory-driven adaptations address changes Types of Adaptation. The need for adaptations based on underlying concept or theory. For exam- can arise from various sources (Malda et al. 2008 ; ple, digit span items should have digit names that Van de Vijver and Leung 2011). Cognitive tests are all of similar length. However, similarity in can be adapted in fi ve ways: construct driven, digit length may be lost when the items are trans- language driven, culture driven, theory driven, lated into another language. and familiarity/recognizability driven. Familiarity/recognizability-driven adaptations Construct-driven adaptations address differ- address differences in the familiarity with task or ences in defi nitions of psychological concepts item characteristics. For example, drawing cords across cultures. For example, there are various on phones is inappropriate for those who are non-Western studies of everyday conceptions of exposed only to mobile phones. children’s intelligence in which obedience and rule compliance are components of intelligence An Illustration of Test Adaptation: The (Goodnow 1984 ; Hess et al. 1980 ). A full assess- KABC-II. As part of a large study of the ment of intelligence in these cultures would infl uence of micronutrients on children’s intel- require the assessment of obedience. lectual skills, a development team (Malda et al. Language-driven adaptations address the 2008 , 2010) adapted the Kaufman Assessment unavailability of synonyms across languages. For Battery for Children, Second Edition (KABC-II; example, a person who displays vigorous energy Kaufman and Kaufman 2004 ) for use with urban in an uncivil manner may be called aggressive in children in Bangalore, India. The children attended English (e.g., an aggressive salesperson). primary school and came from poor families (e.g., Translations in other languages presumably may average monthly income was slightly above $56 amount to a description of the concept rather than USD). Many adults were illiterate or had only a a direct translation in order to avoid the connota- few years of education. Most houses had one or tion of aggression with violence (e.g., the transla- two rooms, and the average number of people in a tion could become “a forceful and energetic household was 5.8. The homes had few or no toys salesperson”). Other adaptations could be a con- and usually no learning materials other than school sequence of language differences using personal books. Most families owned a television. Children references. For example, the English word friend either played in the streets or watched television can indicate both a male and a female person, when not doing chores. while other languages use different words for a The KABC-II (a revised and re-standardized male and a female friend. second edition of the KABC) is an individually Culture-driven adaptations address different administered measure of intelligence designed cultural norms, values, communication styles, for US children and youth ages 3 through 18. 22 Intelligence and Culture: History and Assessment 355

It measures short-term memory, visual processing, sequence as presented by the examiner, with long-term storage and retrieval, fl uid reasoning, series ranging in length from two to nine digits. and crystallized abilities. According to Baddeley’s phonological loop Locally developed test were not available for model (Baddeley et al. 1975 ), the number of items Kannada-speaking children. The choice of the that can be stored in short-term memory varies KABC-II was based on several criteria. The with their phonological length, usually operation- instrument has some attractive features for use in alized by the number of syllables. Shorter items a non-Western context. For example, the instru- are recalled more easily. Thus, Number Recall ment has a fi rm theoretical basis in the Luria neu- will be more sensitive to differences in memory ropsychological model but also one that is capacity when shorter digits are used. As a result, compatible with the CHC model (Carroll 1993 ; maintaining a constant phonological digit length McGrew 2005 ). Both the Luria and the CHC is important. These theoretical concentrations had models were “assumed” to be valid in this popu- a major impact on the instrument in Kannada. All lation. The test administration also is less digits in this language are bisyllabic, except 2 and dependent than other tests on the child’s use of 9 that have three syllables. Therefore, the bisyl- receptive and expressive language skills. labic digits in the Kannada version were used as Additionally, sample items are provided prior to much as possible. Given their increased diffi culty, administering the test to help ensure a child trisyllabic digits were used only toward the end of understands the task demands before commenc- the test. As a consequence of this choice, its items ing the assessment. differ from the original source test The research team was aware that the work Atlantis measures long-term storage and may involve fi ve types of adaptations. The team retrieval. The child is taught nonsense names for attempted to avoid various sources of bias by pictures of fi sh, plants, and shells and then asked selecting the KABC-II. For example, the theo- to point to the corresponding picture in an array retical underpinning of the KABC-II makes it of pictures when it is named. The test requires the unlikely that construct-driven adaptations would child to memorize new phonological information be needed. The intellectual abilities assessed by without the support of the meaning or context of the instrument were assumed to be displayed by the words. Results from a pilot study indicated and salient among the target children. The that the English nonsense names were diffi cult to instrument minimizes the use of language by distinguish for the Indian children. Therefore, the using simple language forms. Thus, language- English nonsense names were replaced with driven adaptations were not expected. The lim- Kannada nonsense names that were suffi ciently ited role of language also minimizes the need distinct for easy discrimination by the children. for culture-driven adaptations. The other two Triangles assesses visual processing. The types of adaptations, theory-driven and famil- child is given several identical foam triangles iarity/recognizability-driven adaptation, were (blue on one side and yellow on the other) and is more urgent, diffi cult, and required considerable asked to put them together so that their fi nal time and effort, as explained below. assembled form is identical to a target picture of an abstract design. For easier items, the child Adapting the Subtests of the KABC-II. Two assembles a set of colorful plastic shapes to KABC subtests required a theory-driven adaptation: match a model constructed by the examiner or Number Recall and Atlantis. Four subtests requi- shown in the test booklet. Results from a pilot red familiarity/recognizability-driven adaptations: study indicated that the children found this task Triangles, Rover, Pattern Reasoning, and Story very diffi cult as there are no similar tasks in their Completion. Two subtests required both types of daily environment (e.g., jigsaws or other tasks adaptation: Word Order and Rebus. where the spatial orientation of components is Number Recall measures short-term memory. important in completing an assembly task). The child is asked to repeat a series of monosyl- Consequently, the examples and fi rst items on the labic digits (1 to 9, excluding 7) in the same instrument were simplifi ed to ensure they could 356 D.H. Saklofske et al. be understood and solved by the children. In Therefore, all illustrated pictures were replaced addition, time constraints were relaxed. These by new items based on those from the Picture adaptations illustrate the infl uence that a lack of Arrangement subset from the Wechsler familiarity with tasks that resemble those on a Intelligence Scale for Children (Wechsler 1949 , test can have on test performance. 1974 , 1991) . Each Picture Arrangement item Rover measures visual processing. The child consists of a series of pictures that depict a story is asked to move a dog toy (named Rover) to a and are presented in an incorrect order. The child bone on a checkerboard-like grid that contains is asked to arrange them in an order that makes a obstacles by making as few moves as possible. sensible story. Pilot study data showed that the children used the Although Picture Arrangement seemed to be direction in which the dog was facing as a cue to less related to a specifi c cultural context, the how they could or were supposed to start. This items still needed considerable modifi cation. The fi nding was both undesirable and unanticipated. number of cards in each item was kept similar to Thus, the dog was replaced with a pawn that did the original Wechsler scales whenever possible. not have a directional indication for the children. Five new themes were introduced (two sample Children did not have much experience with items and three test items), one item from the similar board games. As a consequence, not all original WISC was used, one item from WISC- children in the pilot test understood the moves III was used, and eight items of WISC-R were the dog could make on the board. A sample item adapted. The original Picture Arrangement sub- was changed, and two regular test items were test had only one sample item that did not require converted to sample items to help ensure that the the child’s active engagement. Instead the exam- child understood the principles of the test. As in iner arranges the cards in the correct order, tells Triangles, the original time limits were relaxed. the story, and asks the child whether he or she Pattern Reasoning assesses fl uid reasoning. understood the item. Two sample items that Children are shown a series of stimuli that form a required the child’s active engagement were logical sequence; however, one stimulus in the added. series is missing. The child completes the pattern Word Order examines short-term memory. by selecting the correct stimulus from an array of The examiner fi rst names a set of items that the four to six options depicted at the bottom of the child cannot see and the child then is asked to page. Most stimuli are abstract, geometric shapes, point to a series of silhouettes of common objects whereas some easy items use meaningful pic- in the same order the examiner named them. On tures. The original version requires the assess- the more diffi cult tasks, an interference task (i.e., ment of response times at item level. However, saying the names of colors) is added between the results from fi eld tests revealed the diffi culty of stimulus and the response. Stimuli from the accurately timing responses that often take only a source version of Word Order were selected care- few seconds. Therefore, scoring was changed to fully to ensure that young children with normal the number of items solved correctly. receptive language development would readily Story Completion also measures fl uid reason- identify and label all pictures in an adequate ing . A row of pictures that tell a story is shown manner. The source test only uses objects with but some pictures are missing. The child is given one-syllable names to control for phonological a set of pictures, selects the ones that are needed length and complexity similar to what was previ- to complete the story, and places the missing pic- ously observed for Number Recall (theory-driven tures in their correct sequence. This subtest cre- adaptation). Unlike Number Recall, Word Order ated insurmountable adaptation problems due to does not require the child to respond verbally. the numerous explicit and implicit references to Identifying daily used objects that have one- cultural aspects that were unfamiliar or unknown syllable names in Kannada was diffi cult. Thus, to the children (e.g., having a birthday party, objects with bisyllabic names were selected blowing up balloons, and the use of napkins). (theory-driven adaptation). The additional criteria 22 Intelligence and Culture: History and Assessment 357 for selecting new stimuli were that their names environments. The second component of each and corresponding visual representation (black- iteration involved the administration of subtests and-white drawings) should be unambiguous to samples of children in the target group. The and highly familiar (familiarity/recognizability- adaptation process took several months and led to driven adaptation). One out of the twelve original many major and minor changes to the original stimuli needed to be redrawn (i.e., the original instrument. The adaptations were made by a team drawing of a house contained a chimney—a fea- that included persons with cultural, linguistic, ture unknown to the Indian children). Six of the and psychometric expertise. The absence of any twelve original stimuli had to be replaced. one type of expertise would have led to a different, Drawings of a star, key, hand, moon, heart, and presumably less adequate instrument. shoe were replaced by drawings of a fl ower, The current thinking on test adaptations (e.g., book, leg, sun, chair, and bus, respectively. The Hambleton and Lee 2013 ; Hambleton et al. 2005 ) color interference task (color naming) was used emphasizes the importance of using mixed meth- to measure recall after injecting an interference ods when engaged in test adaptation. During the task. Children had diffi culty naming gray blocks design stage, contacts with experts, piloting, and because Kannada does not have a word for gray. cognitive interviewing are important tools to Thus, this problem was avoided by using blocks develop adequate test adaptations. However, with more familiar colors. these usually qualitative procedures should be Rebus measures associative memory, verbal complemented by acquiring quantitative data learning, and long-term storage and retrieval. The with the new instrument. These data form the examiner states a word or concept and associates core of the quantitative stage that are used to it with a particular drawing. Then the child investigate whether and to what extent the psy- “reads” aloud phrases and sentences composed chometric properties of the original instrument of these drawings (e.g., six different drawings are retained in the new instrument. can form the sentence “The girl and boy play The adapted KABC-II was administered to games”). The research team decided to not use 598 children (Malda et al. 2010 ). Subtest data the Rebus because translating and adapting it were reliable, with values ranging from .70 to would not have been possible. The sentences to .96. Validity was examined using confi rmatory be produced are related strongly to the specifi cs factor analysis. Findings largely replicated the of the local language (e.g., the word order in sen- CHC model underlying the original KABC-II; tences), so that a close (literal) translation was support for Das-Luria’s Planning, Attention- not possible and any modifi cation would produce Arousal, Simultaneous, and Successive (PASS) a considerably different version from the original model was not tested. Relations with demo- item. Therefore, the Verbal Learning Test based graphic characteristics (e.g., sex and age) and a on the Rey Auditory Verbal Learning Test (Rey measure of arithmetic were consistent with 1964 ) replaced the Rebus test. This new test mea- expectations. Thus, the adequacy of the adapta- sures immediate memory, effi ciency of learning, tion was supported. and recall after short and long delays. In summary, the adaptation process comprised a small number of large decisions and a From Test Development to “Testing the Test”. large number of small decisions. Large deci- As can be seen from the above description, sions concerned issues about the suitability of extensive adaptations were needed to design and the subtests. For example, the need to reject and evaluate a test battery that presumably assessed replace the Rebus occurred early in the adapta- the intellectual skills measured by the original tion process. However, considerable time was KABC-II. The adaptations were conducted using devoted to making numerous refi nements to the iterative procedures. Each iteration included instrument. For example, multiple iterations contacts with local psychologists who had expert with sometimes minute changes were needed to knowledge in the local language and children’s develop an adequate version of Triangles. Similarly, 358 D.H. Saklofske et al. designing, drawing, and refi ning items of Picture subtest that went from 3 to 7 to 9 on the standard, Arrangement required much time before the draw- multiple choice and picture vocabulary versions, ings were clear to all children. respectively. Follow-up diagnosis suggested she was an anxious girl with elective mutism and other clinical features that masked her cognitive M o d i fi cations to the IQ Testing abilities under standard testing conditions. Process Nonverbal Measures. Tests that reduce language Various other procedures and instruments have or are fully nonverbal often are used with non- been proposed and implemented to increase test English-speaking (or those with limited English validity when testing members of culturally profi ciency) persons. An increase in the number diverse communities. These include testing the of non-English- speaking immigrants has limits (e.g., following the standardized adminis- increased the need for nonverbal measures of tration, the examiner alters procedures to deter- intelligence. Historically, measures of visual mine if the individual can successfully complete conceptual or perceptual organizational ability items with suspended time limits), demonstrat- found in existing tests (e.g., the subtests that ing and teaching the task to encourage correct previously comprised the Wechsler scale’s performance, incorporating nonverbal mea- Performance IQ) were used for this purpose. sures, and altering and in other ways modifying However, this practice now is deemed inadequate standardized instructions to obtain qualitative given their reliance upon verbal directions. information. Although these alterations jeopar- Nonverbal testing requires an absence of verbal dize a test’s standardized administration poli- instructions and verbal responses (McCallum and cies, they provide an astute and informed Bracken 2013 ; Naglieri 2003 ). The use of psychologist the opportunity to use the test nonverbal tests with persons from racial and framework to draw clinical inferences about the ethnic minority groups decreases yet does not person’s cognitive abilities. Further modifi ca- eliminate IQ differences. Thus, these scales may tions to the assessment process as well as test be referred to as culturally reduced measures. adaptations and alternative methods are described in Table 22.2 . Further discussion of Cross-Battery Assessment. As noted earlier, “testing the limits” can be found in Sattler cross-battery assessment methods may provide (2008 ). The following provides a brief overview important information regarding the use of of some test adaptations that may assist in the intelligence tests with diverse cultural groups assessment of intelligence when culture, lan- through the use of the culture- language interpretive guage, and cultural differences may be salient. matrix (C-LIM; Flanagan et al. 2013; Ortiz 2013 ) and culture-language test classifi cations (C-LTC). Integration of Modifi ed Instructional This system classifi es each subtest within a test Procedures. The WISC-IV Integrated (Kaplan battery in light of its cultural loading and linguistic et al. 2004 ) goes beyond the traditional demand (i.e., low, medium, high). For example, administration of the standardized format to the Wechsler scales’ Matrix Reasoning subtest is enable a clinician to gather information regarding rated low on both linguistic demand and cultural underlying processing issues that may impact loading. However, the Information, Similarities, performance on its 16 subtests. For example, the Vocabulary, and Comprehension subtests are rated standard vocabulary subtest of the WISC-IV can high on both linguistic demand and cultural be further supplement with both a multiple choice loading. The classifi cations are based upon and picture vocabulary versions. For example, a empirical data, when available, or expert consensus young client who was referred for possible procedures and judgment. The use of these systems intellectual retardation and tested on the WISC-IV may assist a clinician when selecting tests and revealed a pattern of scores on the vocabulary interpreting their data. 22 Intelligence and Culture: History and Assessment 359

Table 22.2 Potential indicators and measures of intelligence for individuals from diverse cultural backgrounds Options Instruments/process Use of “other” markers of intelligence Developmental milestones (e.g., motor and speech development in infancy linked to cognitive functioning; e.g., Murray et al. 2007 ) Achievement indicators (e.g., IQ is correlated: .50 with grades; .55 with total years of education; e.g., Neisser et al. 1996 ) Tests of mental chronometry People with higher IQ “apprehend, scan, retrieve, and respond to stimuli (e.g., reaction time) more quickly than those who score lower” (e.g., Neisser et al. 1996 , p. 83) Curriculum-based assessment (CBM) CBM may be a viable alternative to traditional norm- referenced testing for African American students (e.g., Fore et al. 2006 ). CBM provides meaningful information regarding a student’s performance with direct linkage to instructional programs Dynamic assessment Dynamic assessment and instruction yields fi ndings reducing the minority performance achievement gap in math (e.g., Jeltova et al. 2011 ) Use of performance/nonverbal Example: use of perceptual reasoning index of Wechsler scales reasoning subtests Use of interpreter Trained interpreters for the process of assessment may be used with caution given that issues of equivalence (as noted earlier) will impact fi ndings Use of nonverbal measures Nonverbal intelligence tests such as the University Nonverbal Intelligence Test (UNIT; 1998); Test of Nonverbal Intelligence Fourth Edition (TONI; Brown et al. 2010 ); Leiter International Performance Scale Third Edition (Leiter-3; Roid et al. 2013 ); Raven’s Standard Progressive Matrices (RSPM; Raven et al. 1998 ). Note some nonverbal measures contain verbal instructions while others do not. Nonverbal measures are identifi ed as culturally reduced given absence of language component Use of measures translated and Many instruments such as the Wechsler Scales have been translated and standardized in country of origin standardized in many countries (e.g., Georgas et al. 2003 )

Biocultural Model. The biocultural model of more routine procedures. Attention to other forms intelligence (Armour-Thomas and GoPaul- of “intelligences” (e.g., musical) also is included. McNicol 1998 ) incorporates a pre-assessment For example, the work of Howard Gardner and ecological evaluation processes to obtain reminds us that our views of intelligence information regarding health, linguistics, and sometimes are narrow, in part, because we prior experiences in a person’s school, home, and seemingly judge the test’s ability to predict community (e.g., educational, psychosocial, school achievement as the expression of a test’s acculturation). Attention to these issues may validity. For example, a First Nations student enable a clinician to understand an examinee’s who was failing in school, withdrawn, and hostile personal and environmental contexts. Using this toward teachers also displayed athletic skills. Her model, psychometric tests fi rst are administered track and fi eld activities (e.g., short- and long- using standardized procedures and then distance running as well as broad and high unstandardized methods designed to identify the jumping) and team sport activities in volleyball examinee’s intellectual potential are used by and soccer were better than her school peers. An suspending the administration time, assessment that focused on the use of her skills contextualizing language vocabulary (e.g., and interest areas in a learning context found that allowing the examinee to use the word in a she could read, communicate, strategize, plan, sentence if they are unable to provide a defi nition), and keep time charts to determine improvements allowing use of paper and pencil, and applying in performance. She also was able to illustrate dynamic assessment procedures (test-teach- human movement and motion using line drawings retest). Thus, in many ways, this process is akin that were quite exceptional. Thus, the collection to testing the limits yet goes beyond many of the and incorporation of additional information 360 D.H. Saklofske et al. gathered through these alternative procedures that one cultural context may not transfer well to may include informal assessment, observation, another context. and interviews may enable a clinician to form a Knowledge of the assessment procedures used more accurate judgment of an individual’s as immigrants were entering the United States at potential and estimated abilities beyond that Ellis Island may help avoid similar practices provided through the use of standardized (Gould 1996 ; Tulsky et al. 2003 ). The inadequacy administrative processes. of the testing procedure combined with the exhaustion of the immigrants after the long boat trip, the effects of fl eeing from war and poverty, Conclusions the unfamiliarity of this new context, and the stress of wanting an opportunity for a better life The structure of intelligence is universal; there is underscores the importance of assessing intellec- compelling evidence through many cross-cultural tual skills in ways that acknowledge the examin- studies of intelligence tests that these measures ee’s social, cultural, and linguistic background are both robust and invariant. Basic features (e.g., and align assessment methods consistent with it working memory, spatial reasoning) are part of along with the context in which the data will be the cognitive repertoire of individuals across the used. Many intelligence tests have been devel- globe. However, this universality could be mis- oped for Western cultural groups and thus may be construed if applied universally in a rigid manner. culturally loaded. A test’s good psychometric While we may agree that crystallized intelligence properties do not compensate for their use in cul- refl ects the capacity to learn, store, and retrieve turally insensitive ways. A theory is neither right information (the terms schema, assimilation, nor wrong, neither good nor bad, and should be accommodation, declarative, semantic and pro- judged whether it is useful or useless in under- cedural knowledge could be used), the examples standing, predicting, and even changing human in this chapter show that beyond the underlying behavior. The tools we create to refl ect the latent processes and structure of intelligence that is traits and human characteristics derived from manifested and what is valued will vary across these theories and models must be viewed in the cultures. Thus, persons engaged in the assess- same light. While one could argue with this view- ment of intelligence cross-nationally should con- point (e.g., Nazi views of Aryan racial superior- sider the following four issue: the nature and ity leading to support for ethnic cleansing and importance of intellectual abilities desired by a genocide programs or restricting opportunity culture or society; whether these qualities differ based on racial views of intelligence), it seems based on a person’s age, gender, and socioeco- much more meaningful if also placed in the con- nomic status; the degree of overlap between these text of ethical practice. Psychologists who use desired intellectual qualities and the current measures of intelligence tests have an important structure of intelligence as more universally responsibility to the individual client and to soci- defi ned such as the CHC model; and the avail- ety, one that should refl ect comprehensive and ability of valid methods to assess intelligence in relevant training and in-depth knowledge of psy- light of this information. chological research as well as practice standards, The repertoire of abilities captured by the cultural sensitivity, and ethics. To paraphrase structure of intelligence is not used uniformly by Anne Anastasi whose brilliant writings on psy- all people. One’s culture and socialization prac- chological testing and assessment informed gen- tices contribute to the promotion of specialized erations of psychologists and others, “we are all and possibly different abilities. Persons acquire of equal value as human beings” (Gottfredson culture-specifi c skills along with universal skills 1997). And if we superimpose on that, the recog- through their socialization. Cultures also rein- nition echoed by Kluckhohn et al. (1953 ) that force the display of biologically based abilities. noted the ways in which we are “like all others, As a result, assessment procedures developed in some others and no others,” the construct of 22 Intelligence and Culture: History and Assessment 361

intelligence and its assessment in a culturally et al 2001). Thus, given its international focus, diverse world will be well served. The information ITC leadership recognized its need to attend to in this chapter is intended to identify and clarify prevailing issues that were impeding test develop- issues that may arise when using measures of ment and use regionally and internationally. The intelligence cross-culturally or cross-nationally. lack of authoritative guidelines can contribute to assessment services that are unreliable, inferior in quality, and reﬂ ect unsuitable standards. Thus, the Appendix A ITC began developing various guidelines intended to promote sound testing practices. Features of International Test Commission four guidelines that may be most pertinent to cog- Guidelines nitive assessment are summarized below. Information on the guidelines can be found on the The use of cognitive and other ability measures ITC website at http://www.intestcom.org . can engender strong negative feelings and opinions—particularly in reference to their misuse Test Adaptation Guidelines. The ITC recog- with groups that may display lower than average nized most countries lack resources needed to mean scores, when tests are misused when assess- develop their own tests and thus will continue the ing immigrants and second-language learners, or practice of obtaining tests, often measures of when tests do not meet suitable psychometric cognitive ability (Hu and Oakland 1991 ), from standards. During the 1970s and 1980s, articles in others. Thus, guidelines were needed to adapt these newspapers and professional journals alleged that tests for local use. The guidelines provide guidance examiners often are poorly prepared; tests typi- in reference to the context of a test’s use, test cally rely on a small sample of behavior, often development and adaptation, test administration, those to which persons from lower-class homes and documentation/scoring interpretations. These lacked exposure; tests overlook functional daily guidelines distinguish two processes: translating/ used qualities; tests are the servants of the estab- adapting existing tests and instruments as well as lished middle class and are designed to discrimi- developing new instruments intended to be used nate against the poor and colored; and that cheating internationally to provide international comparisons. on tests is widespread and invalidates their use. In short, the allegations suggest tests generally are ITC Guidelines for Test Use. These guidelines unfair, culture bound, biased, and used discrimina- were developed to help overcome unsuitable torily (Oakland 1977; Phelps 2008). assessment practices by focusing on test user Many of these issues surfaced in countries competencies. Competencies require acquiring other than the United States during this same basic knowledge (e.g., psychometric principles and period (Oakland 2009; Oakland et al. 2013) fueled, procedures along with technical requirements of in part, by two conditions: (1) the use of pirated tests) and skills (e.g., suitable use of assessment copies of tests, developed originally in the United procedures), displaying suitable professional and States or Western Europe, and translated into the ethical behaviors, assuming responsibility for test local language and without adequate efforts to use, ensuring test materials are secured, ensuring acquire norms or validity data as well as (2) strong that test results are treated conﬁ dentially, evaluating reliance on theory from social psychology that the potential utility of testing, selecting technically expressed disdain for the principle of meritocracy sound tests appropriate for the situation, considering and favoritism toward egalitarian views. issues of fairness, establishing rapport, scoring and The International Test Commission (ITC) was analyzing test results accurately displaying a good established during this time to promote an understanding of the test’s theoretical or conceptual exchange of information on test development and basis, communicating the results clearly and use as well as to address prevailing issues that cut accurately to relevant others, and reviewing the across national boundaries and cultures (Oakland appropriateness of the test and its use. 362 D.H. Saklofske et al.

International Guidelines on Computer-Based behavioral), the importance of psychometric and Internet-Delivered Testing. The use of qualities, interpersonal relationships, and ethical computers to administer scores and report tests issues. The importance of considering the has become somewhat routine. Lamentably, acculturation status of immigrant clients (i.e., individuals with limited ability to develop suitable their orientation and level of adjustment to their tests are able to market their inferior products ethnic culture and the mainstream culture) and online and potentially reach all people who have possible ways to address acculturation issues are Internet access (Friedman 2005 ). These guidelines described. Possible test modifi cations together are intended to promote good practices for with best practice guidelines are suggested. computer-based (CBT) and Internet-delivered These proposed guidelines are intended to assist testing. They address technical issues (e.g., professionals in the provision of clinical hardware and software requirements; the assessment services for persons who are accul- robustness of the CBT/Internet test; human factors turating to a new culture and are second-language in the presentation of materials via a computer learners. The ITC leadership currently is or the Internet; providing reasonable adjustments reviewing these proposed guidelines. to the test’s technical features for test-takers who display disabilities; and providing help, information, and practice items) as well as quality References control issues (e.g., knowledge, competence, and appropriate use of CBT/Internet testing; the Armour-Thomas, E., & GoPaul-McNicol, S. A. (1998). psychometric qualities of CBT/Internet tests; equi- Assessing intelligence: Applying a biocultural model. Thousand Oaks, CA: Sage. valence between paper-and-pencil tests and those Baddeley, A. D., Thomson, N., & Buchanan, M. (1975). delivered via CBT/Internet methods; scoring Word length and the structure of short-term memory. and analyzing CBT/Internet results accurately; Journal of Verbal Learning and Verbal Behavior, 14 , interpret results appropriately, including feedback; 575–589. Berry, J., Poortinga, Y., Breugelmans, S., Chasiotis, A., & and equality of access for all groups). Sam, D. (2011). Cross-cultural psychology. Research and applications (3rd ed.). Cambridge, UK: Proposed ITC Guidelines for the Clinical Cambridge University Press. Assessment of Immigrants and Second- Bracken, B. A., & McCallum, R. S. (1998). Universal nonverbal intelligence test examiner’s manual . Itsaka: Language Learners. An increase in personal Riverside. mobility has led to an increase in the number Brislin, R. W. (1970). Back translation for cross-cultural of persons who hold immigrant status, especially research. Journal of Cross Cultural Psychology, 185–216. in Western countries. They frequently seek Brown, L., Sherbenou, R. J., & Johnsen, S. K. (2010). Test of nonverbal intelligence (4th ed.). San Antonio: Pearson. improved lifestyles, including education, work, Butcher, J. N., & Han, K. (1996). Methods of establishing safety, and other conditions that contribute to cross- cultural equivalence. In J. N. Butcher (Ed.), their health, welfare, and stability. Their International adaptations of the MMPI-2: Research immigrant status implies they are residing in a and clinical applications (pp. 44–63). Minneapolis: University of Minnesota Press. different culture and are likely to lack fl uency in Carroll, J. B. (1993). Human cognitive abilities: A survey the host country’s primary or preferred language of factor-analytic studies. Cambridge, UK: Cambridge and thus are second-language learners. Attempts University Press. to assess their cognitive abilities pose severe Cattell, R. B. (1940). A culture free intelligence test, I. Journal of Educational Psychology, 31 , 176–199. challenges to assessment specialists. These Dasen, P. R. (2011). Culture, cognition and learning. In proposed guidelines consider fi ve issues that may A. Bame Nsamenang & T. S. Tschombe (Eds.), impact the clinical assessment of persons whose Handbook of African educational theories and prac- cultural and linguistic qualities differ from the tices: A generative teacher education curriculum contents (pp. 187–202). Bamenda: Human Development local culture: qualities associated with culture Resource Centre. and language, four personal qualities that commonly Edwards, O., & Oakland, T. (2006). Factorial invariance are assessed (i.e., medical, social, cognitive, and of Woodcock-Johnson scores for African Americans 22 Intelligence and Culture: History and Assessment 363

and Caucasian American. Journal of Psycho- International Test Commission, (2010). International test educational Assessment, 24 , 358–366. commission guidelines for translating and adapting Flanagan, D. P., Ortiz, S. O., & Alfonso, V. C. (2013). tests. Retrieved from http://www.intestcom.org/ Essentials of cross-battery assessment (2nd ed.). upload/sitefi les/40.pdf Hoboken: Wiley. Jeltova, I., Birney, D., Fredine, N., Jarvin, L., Sternberg, Flynn, J. R. (2012). Are we getting smarter? Rising IQ in R. J., & Grigorenko, E. L. (2011). Making instruction the twenty-fi rst century . New York: Cambridge and assessment responsive to diverse students’ prog- University Press. ress: Group administered dynamic assessment in Fore, C., Burke, M. D., & Martin, C. (2006). Curriculum- teaching mathematics. Journal of Learning based measurement: An emerging alternative to tradi- Disabilities, 44 , 381–395. tional assessment for African American children and Jensen, A. (1972). Genetics and education. San Francisco: youth. The Journal of Negro Education, 75 , 16–24. Harper & Row. Friedman, T. (2005). The world is fl at. New York: Farrar, Jensen, A. (1973). Educability & group differences . San Straus and Giroux. Francisco: Harper & Row. Georgas, J., Weiss, L., Van de Vijver, F. J. R., & Saklofske, Jensen, A. (1980). Bias in mental testing . New York: The D. H. (Eds.). (2003). Cultures and children’s intelli- Free Press. gence: A cross-cultural analysis of the WISC-III . Kaplan, E., Fein, D., Kramer, J., Delis, D., & Morris, R. New York: Academic. (2004). WISC-IV integrated. San Antonio: The Goodnow, J.1. (1984). On being judged intelligent. Psychological Corporation. International Journal of Psychology, 19:391–406. Kaufman, A. S., & Kaufman, N. L. (2004). Kaufman Gottfredson, L. S. (1994). The science and politics of assessment battery for children, second edition: race-norming. American Psychologist, 49 , 955–963. Manual . Circle Pines: AGS Publishing. Gottfredson, L. S. (1997). Mainstream science on intelli- Kluckhohn, C. E., Murray, H. A., & Schneider, D. M. gence: An editorial with 52 signatories, history and (Eds.). (1953). Personality in nature, society, and cul- bibliography. Intelligence, 24 (1), 13–23. ture . New York: Knopf. Gould, S. J. (1996). The mismeasure of man . New York: Lonner, W. J. (1980). The search for psychological uni- Norton & Company. versals. In H. C. Triandis & W. W. Lambert (Eds.), Grigorenko, E. L., Geissler, P. W., Prince, R., Okatcha, F., Handbook of cross-cultural psychology (Vol. 1, Nokes, C., Kenny, D. A., Bundy, D. A., & Sternberg, pp. 143–204). Boston: Allyn and Bacon. R. J. (2001). The organization of Luo conceptions of Lonner, W. J. (1985). Issues in testing and assessment in intelligence: A study of implicit theories in a Kenyan cross-cultural counseling. Counseling Psychologist, village. International Journal of Behavioral 13 , 599–614. Development, 25 , 367–378. Lonner, W. J. (2011). The continuing challenge of discov- Hambleton, R. K., & Lee, M. K. (2013). Methods for ering psychological ‘order’ across cultures. In F. J. translating and adapting tests to increase cross- R. Van de Vijver, A. Chasiotis, & S. M. Breugelmans language validity. In D. H. Saklofske, C. R. Reynolds, (Eds.), Fundamental questions of cross-cultural psy- & V. L. Schwean (Eds.), The Oxford handbook of child chology (pp. 64–94). Cambridge, UK: Cambridge psychological assessment (pp. 172–181). New York: University Press. Oxford University Press. Malda, M., Van de Vijver, F. J. R., Srinivasan, K., Transler, Hambleton, R., Merenda, P., & Spielberger, C. (Eds.). C., Sukumar, P., & Rao, K. (2008). Adapting a cogni- (2005). Adapting educational and psychological tests tive test for a different culture: An illustration of quali- for cross-cultural assessment . Mahwah: Erlbaum. tative procedures. Psychology Science Quarterly, 50 , Helms, J. E. (1992). Why is there no study of cultural 451–468. equivalence in standardized cognitive ability testing. Malda, M., Van de Vijver, F. J. R., Srinivasan, K., Transler, American Psychologist, 47 , 1083–1101. C., & Sukumar, P. (2010). Traveling with cognitive Helms-Lorenz, M., Van de Vijver, F. J. R., & Poortinga, tests: Testing the validity of a KABC-II adaptation in Y. H. (2003). Cross-cultural differences in cognitive India. Assessment, 17 , 107–115. performance and Spearman’s Hypothesis: G or C? Marfo, K. (2011). Envisioning an African child develop- Intelligence, 31 , 9–29. ment fi eld. Child Development Perspectives, 5 (2), Hernstein, R., & Murray, C. (1994). The bell curve: 140–147. Intelligence and class structure in American life . Mayer, J. D., Salovey, P., & Caruso, D. R. (2008). New York: The Free Press. Emotional intelligence: New ability or eclectic mix of Hess, R., Kashiwagi, K., Azuma, H., Price, G., & Dickson, traits? American Psychologist, 63 , 503–517. W. P. (1980). Maternal expectations for mastery of McCallum, R. S., & Bracken, B. A. (2013). Intelligent developmental tasks in Japan and the United States. testing of underserved populations. In J. C. Kaufman International Journal of Psychology, 15, 259–271. (Ed.), Intelligent testing: Integrating psychological Hu, S., & Oakland, T. (1991). Global and regional per- theory and clinical practice (pp. 193–219). New York: spectives on testing children and youth: An interna- Cambridge University Press. tional survey. International Journal of Psychology, 26 , McGrew, K. S. (2005). The Cattell-Horn-Carroll theory of 329–344. cognitive abilities: Past, present, and future. In D. P. 364 D.H. Saklofske et al.

Flanagan & P. L. Harrison (Eds.), Contemporary intel- Nsamenang, A. B. (2006). Human ontogenesis: An lectual assessment: Theories, tests, and issues indigenous African view on development and intel- (pp. 136–181). New York: The Guilford Press. ligence. International Journal of Psychology, 41 , McGrew, K., & Flanagan, D. (1998). The intelligence test 293–297. desk reference: Gf-Gc cross-battery assessment . Oakland, T. (Ed.). (1977). Psychological and educational Boston: Allyn & Bacon. assessment of minority children . Larchmont: Mercer, J. R., & Lewis, J. F. (1977). SOMPA, System of Brunner-Mazel. multicultural pluralistic assessment: Parent interview Oakland, T. (2009). How universal are test development manual . New York: Psychological Corporation. and use? In E. Grigorenko (Ed.), Assessment of abili- Meyer, G. J., Finn, S. E., Eyde, L. D., Kay, G. G., ties and competencies in an era of globalization Moreland, K. L., Dies, R. R., Eisman, E. J., Kubiszyn, (pp. 1–40). New York: Springer. T. W., & Reed, G. M. (2001). Psychological testing Oakland, T., & Harrrison, P. (2008). Adaptive behavior and psychological assessment. American Psychologist, and skills: An introduction. In T. Oakland & 56 , 128–165. P. Harrison (Eds.), ABAS-II: Clinical uses and inter- Mpofu, E. (2004). Being intelligent with Zimbabweans: A pretations . New York: Elsevier. historical and contemporary view. In R. J. Sternberg Oakland, T., & Hu, S. (1992). The top ten tests used with (Ed.), International handbook of intelligence children and youth worldwide. Bulletin of the (pp. 364–390). Cambridge/New York: Cambridge International Test Commission, 19 , 99–120. University Press. Oakland, T., Poortinga, Y., Schlegel, J., & Hambleton, Mpofu, E., & Nyanungo, K. R. (1998). Educational and R. K. (2001). International Test Commission: Its his- psychological testing in Zimbabwean schools: Past, tory, current status, and future directions. International present and future. European Journal of Psychological Journal of Testing, 1 , 3–32. Assessment, 14 , 71–90. Oakland, T., Wechsler, S., & Maree, K. (2013). Test use Mpofu, E., & Ortiz, J. (2009). Equitable assessment prac- with children across cultures: A view from three tices in diverse contexts. In E. Grigorenko (Ed.), countries. In K. F. Geisinger (Ed.), APA handbook of Multicultural psychoeducational assessment testing and assessment in psychology (Vol. 3, (pp. 41–76). New York: Springer. pp. 231–258). Washington, DC: American Mpofu, E., Gudyanga, E., & Ngara, C. (2007). Psychological Association. Constructions of giftedness among The Shona of Ortiz, S. O. (2013). Distinguishing difference from disor- Central-Southern Africa. In S. N. Phillipson & der in English learners and guidance on how to use M. McCann (Eds.), What does it mean to be gifted? the Culture-Language Interpretive Matrix. New York: Sociocultural perspectives (pp. 225–252). Mahwah: Cross-Battery Institute, Fordham University. Lawrence Erlbaum. Pence, A. (2011). Early childhood care and development Mpofu, E., Ntinda, K., & Oakland, T. (2012). research in Africa: Historical, conceptual, and struc- Understanding human abilities in Sub-Saharan tural challenges. Child Development Perspectives, African settings. Online Readings in Psychology and 5 (2), 112–118. Culture, Unit 4. Retrieved from http://scholarworks. Phelps, R. (Ed.). (2008). Correcting fallacies about edu- gvsu.edu/orpc/vol4/iss3/2 . ISBN 978-0-9845627-0-1 cational and psychological testing . Washington, DC: Mpofu, E., Oakland, T., Ntinda, K., Seeco, E., Maree, J.G. American Psychological Association. (2014). Constructing a framework for the use of tests Porteus, S. D. (1937). Primitive intelligence and environ- within a developing country’s school system. ment. New York: MacMillan Co. International Perspectives in Psychology: Research, Priﬁ tera, A., Saklofske, D. H., & Weiss, L. G. (Eds.). Practice, Consultation, 3(2), 106–122. doi: 10.1037/ (2008). WISC-IV clinical interpretation and interven- ipp0000015 tion . San Diego: Academic. Murphy, L. M., Geisinger, K. F., Carlson, J. F., & Spies, Raven, J., Raven, J. C., & Court, J. H. (1998). Raven’s R. A. (Eds.). (2011). Tests in print VIII . Lincoln: Buros standard progressive matrices – Parallel form . Oxford: Institute of Mental Measurements. Oxford Psychologists Press. Murray, G. K., Jones, P. B., Kuh, D., & Richards, M. Rey, A. (1964). L'examen clinique en psychologie [The (2007). Infant developmental milestones and subse- clinical examination in psychology] . Paris: Presses quent cognitive function. Annals of Neurology, 62 (2), Universitaires de France. 128–136. Roid, G. H., Miller, L. J., Pomplum, M., & Koch, C. Naglieri, J. A. (2003). Naglieri nonverbal abtlity Test . San (2013). Leiter international performance scale (3rd Antonio: Psychological Corporation. ed.). Torrance: Western Psychological Services. Neisser, U., Boodoo, G., Bouchard, T. J., Jr., Boykin, Rushton, J. P., & Jensen, A. R. (2005). Thirty years of A. W., Brody, N., Ceci, S. J., Halpern, D. F., Loehlin, research on race differences in cognitive ability. J. C., Perloff, R., Sternberg, R., & Urbina, S. (1996). Psychology, Public Policy, and Law, 11 (2), 235–294. Intelligence: Knowns and unknowns. American Ryan, C. (2013, August). Language use in the United Psychologist, 51 (2), 77–101. States 2011: American community survey reports. Nisbett, R. E. (2009). Intelligence and how to get it: Why Retrieved from www.census.gov/prod/2013pubs/acs- schools and cultures count . New York: Norton. 22.pdf 22 Intelligence and Culture: History and Assessment 365

Saklofske, D. H., Reynolds, C. R., & Schwean, V. L. Handbook of culture and psychology (pp. 77–97). (Eds.). (2013). Oxford handbook of child psychologi- Oxford: Oxford University Press. cal assessment . New York: Oxford University Press. Van de Vijver, F. J. R. (2003). Bias and equivalence: Sattler, J. M. (2008). Assessment of children: Cognitive Cross-cultural perspectives. In J. A. Harkness, F. J. foundations (5th ed.). San Diego: Jerome Sattler. R. Van de Vijver, & P. P. Mohler (Eds.), Cross- Serpell, R. (2008). Participatory appropriation and the cultural survey methods (pp. 143–155). New York: cultivation of nurturance: A case study of African pri- Wiley. mary health science curriculum development. In P. R. Van de Vijver, F. J. R., & Leung, K. (2011). Equivalence Dasen & A. Akkari (Eds.), Educational theories and and bias: A review of concepts, models, and data ana- practices from the “majority world” (pp. 71–97). New lytic procedures. In D. Matsumoto & F. J. R. van de Delhi: Sage. Vijver (Eds.), Cross-cultural research methods in psy- Serpell, R. (2011). Social responsibility as a dimension of chology (pp. 17–45). New York: Cambridge University intelligence, and as an educational goal: Insights from Press. programmatic research in an African society. Child Van de Vijver, F. J. R., & Poortinga, Y. H. (2005). Development Perspectives, 5 , 126–133. Conceptual and methodological issues in adapting Serpell, R., & Haynes, B. P. (2004). The cultural practice tests. In R. K. Hambleton, P. F. Merenda, & C. D. of intelligence testing: Problems of international Spielberger (Eds.), Adapting educational tests and export. In R. J. Sternberg & E. L. Grigorenko (Eds.), psychological tests for cross-cultural assessment Culture and competence: Contexts of life success (pp. 39–64). Mawhaw: Erlbaum. (pp. 163–185). Washington, DC: American VandenBos, G. (2007). APA dictionary of psychology . Psychological Association. Washington, DC: American Psychological Association. Snyderman, M., & Rothman, S. (1988). The IQ contro- Wechsler, D. (1949). Wechsler intelligence scale for chil- versy, the media and public policy . New Brunswick: dren . New York: Psychological Corporation. Transaction Publishers. Wechsler, D. (1974). Manual for the Wechsler intelligence Sternberg, R. J. (1985). Implicit theories of intelligence, scale for children-revised. San Antonio: The creativity and wisdom. Journal of Personality and Psychological Corporation. Social Psychology, 49 , 607–627. Weiss, L. G., Saklofske, D. H., Coalson, D., & Raiford, Sternberg, R. J. (1999). The theory of successful intelli- S. E. (Eds.). (2010). WAIS-IV clinical use and inter- gence. Review of General Psychology, 3 , 292–375. pretation . San Diego: Academic. Sternberg, R. J., & Jordan, J. (Eds.). (2005). A handbook Weiss, L. G., Keith, T. Z., Zhu, J., & Chen, H. (2013a). of wisdom: Psychological perspectives . New York: WAIS-IV and clinical validation of the four and fi ve Cambridge University Press. factor interpretative approaches. Journal of Sternberg, R. J., Nokes, C., Geissler, P. W., Prince, R., Psychoeducational Assessment, 31 , 94–113. Okatcha, F., Bundy, D. A., & Grigorenko, E. L. (2001). Weiss, L. G., Keith, T. Z., Zhu, J., & Chen, H. (2013b). The relationship between academic and practical intelli- WISC-IV and clinical validation of the four and fi ve gence: A case study in Kenya. Intelligence, 29 , 401–418. factor interpretative approaches. Journal of Trimble, J. E. (2010). Cultural measurement equivalence. Psychoeducational Assessment, 31 , 14–131. In C. C. Clauss-Ehlers (Ed.), Encyclopedia of cross Wober, M. (1974). Towards an understanding of the cultural school psychology (pp. 316–318). New York: Kiganda concept of intelligence. In J. W. Berry & P. R. Springer. Dasen (Eds.), Culture and cognition: Readings in Tulsky, D., Saklofske, D. H., Chelune, G. J., Heaton, cross-cultural psychology (pp. 261–280). London: R. K., Ivnik, R. J., Bornstein, R., Prifi tera, A., & Methuen. Ledbetter, M. F. (2003). Clinical interpretation of the Wolf, T. (1973). Alfred Binet . Chicago: University of WAIS-III and WMS-III . San Diego: Academic. Chicago Press. Van de Vijver, F. J. R. (1997). Meta-analysis of cross- Yan, G., & Saklofske, D. H., (2004). Intelligence: Views cultural comparisons of cognitive test performance. of Chinese psychologists. Paper presented at 28th Journal of Cross-Cultural Psychology, 28 , 678–709. international congress of psychology, August. Beijing, Van de Vijver, F. J. R., & Leung, K. (1997). Methods and China. data analysis for cross-cultural research. Newbury Yan, G., Saklofske, D. H., & Oakland, T. (2009). Park, CA: Sage. Contrasting views of intelligence: Chinese and North Van de Vijver, F. J. R. (2001). The evolution of cross- American perspectives. School Psychology cultural research methods. In D. Matsumoto (Ed.), International, 30 , 456–473. Common and Variable Aspects of Intelligence 23

Arthur MacNeill Horton Jr. and Cecil R. Reynolds

Intelligence has been conceptualized as a product spectives. Level of intelligence is important to of the overall physiological effi ciency of the establish in multiple contexts as a selection and brain itself and crucial for adaptive problem solv- placement criterion as well as an aid in diagnosis ing. David Wechsler ( 1944 ) defi ned intelligence and treatment, including as a baseline of overall as “the aggregate or global capacity of the indi- mental function against which more specifi c cog- vidual to act purposefully, to think rationally, and nitive skills may be compared. Human intelli- to deal effectively with his environment” (p. 3). gence as a clinical biobehavioral concept was This chapter considers the assessment of human initially proposed by Alfred Binet. Essentially, intelligence from a broad perspective of brain- City of Paris, France, public school offi cials were behavior relationships as an aid to understanding concerned that children of impaired cognitive implications of normal as well as abnormal intel- ability were not being discriminated from chil- lectual ability through the discussion of biobe- dren of normal cognitive ability and that was dis- havioral paradigms related to human intelligence. ruptive to the education of all children because of All psychologists have a strong interest in intel- the need for differential education methods for ligence from theoretical, social, and clinical per- both groups of children. The education offi cials requested Binet to develop a method to discrimi- Note: This chapter is based in part on prior work of Drs. nate children of impaired cognitive ability from Reynolds and Horton in Reynolds, C. R., C. Castillo and children of normal cognitive ability to determine Horton, A. M. Jr. (2003). The Neuropsychological Basis of which children should be in special education, Intelligence: Forensic Implications. In A. Horton & and he subsequently created the fi rst successful L. Hartlage (Eds.), Handbook of Forensic Neuropsychology . Second Edition New York: Springer Publishing Company. standardized intelligence test (Binet and Simon 1905 , 1908 ). It is noteworthy that the task was Disclosure Statement : Dr. Reynolds is senior author of the Reynolds Intellectual Assessment Scale (RIAS) and one of practical signifi cance – selection and the Test of Verbal Conceptualization and Fluency (TVCF). appropriate placement of children in an academic Dr. Horton is second author of the TVCF. context, similar to what is done today through the A. M. Horton Jr. , Ed.D., ABPP, ABN (*) SAT Reasoning Test (previously the Scholastic Neuropsychology section , Psych Associates Aptitude Test and Scholastic Assessment Test) or of Maryland, LLC, 120 Sister Pierre Drive, Suite 403 , Graduate Record Exams (GRE). Towson , MD 21204 , USA e-mail: [email protected] Later, David Wechsler developed an intelligence test for adults to aid in clinical assessment C. R. Reynolds , Ph.D. Texas A&M University , 40 N 1H 35, 12B1 , (Wechsler 1939 ). Wechsler’s older brother (Israel Austin , TX 78701 , USA Wechsler) was a neurologist (Chief of Neurology e-mail: [email protected] at Bellevue Hospital in New York City), and the

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 367 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_23, © Springer Science+Business Media New York 2015 368 A.M. Horton Jr. and C.R. Reynolds need for an intelligence test for standardized rience with an individual of great intelligence – examination for neurological patients may have his own student who became the ruler of most of been an infl uence. Wechsler had been a research the then known world. The concept of g has been assistant to Robert Yerkes when Yerkes was in conceptualized as the average of an individual’s charge of assessing the 17 million men who were higher-level cognitive abilities as assessed by drafted to fi ght in World War I. The 17 million many different types of cognitive tasks. Put men were assessed with psychological tests com- another way, g can be conceptualized as a latent monly used at that time to select which men trait rather than an observable outcome. should be selected for offi cer training (Wechsler Research on g has been extremely important 1939). Two tests were developed, the Army in psychology, and g has been a very useful Alpha and the Army Beta. Army Alpha was a means of conceptualizing overall intellectual verbal test of mental abilities used to assess ability (Aluja-Fabregat et al. 2000 ; Kane 2000 ). recruits that were native English speakers. After Jensen (1998 ) after reviewing the empirical using Army Alpha, it was noted that many research for the presence of a general cognitive recruits were immigrants from Eastern Europe ability factor in intelligence concluded that if a and Italy who were not native English speakers very large number of tests were used to assess a but might be able to be good offi cers. To assess very wide spectrum of mental abilities, then a g these nonnative English speakers properly, Army factor would always be found (Jensen 1998 ). Beta, a nonverbal mental abilities test, was devel- Failures to fi nd a g factor in prior research studies oped. The psychological tests that had been most were attributed to failures to use a large enough successful for selecting offi cers during World number of tests and assess a large enough differ- War I were used by Wechsler when he developed ent types of abilities (Jensen 1998 ). his own fi rst intelligence test (Wechsler 1939 ). Moreover, researchers (Reynolds and French Wechsler combined tasks similar to those on 2003 ) have suggested the study of g and the study the Army Alpha and Beta tests (verbal and non- of cognitive processing styles are complementary verbal tests) to form the Wechsler-Bellevue areas in intelligence research investigation. Intelligence Scale (Wechsler 1939 ). The most Indeed, evidence for simultaneous and succes- important contribution of Wechsler ( 1939 ) was sive information processes in the human brain methodological as he standardized the adminis- may be complementary to the concept of a g fac- tration and scoring of the tests and normed the tor. The verbal and performance factors found in intelligence test in large national samples of research studies of intelligence testing are exam- adults (Wechsler 1952 ) and introduced age- ples of the related cognitive factors in diverse corrected deviation scaled scores for interpreta- populations (Reynolds 1981 ). The abstract con- tion of intelligence test performance. Clinical cept of g can be seen as possibly complementary psychology as a fi eld has thus had a long-term to differences in the level or effi ciency of infor- interest in evaluating intelligence (Horton and mation processing (Das et al. 1979 ; Detterman Wedding 1984 ). 1982 ). For example, Travers (1977 ) and Luborsky et al. ( 1971), in studies of psychotherapy outcome research, found that the best predictor of The Role of “g ” successful psychotherapy outcome was the intel- In Neuropsychological ligence level (i.e., g ) of the individual receiving Models of Intelligence psychotherapy. The researchers unfortunately seem to have overlooked the option of assessing The Greek philosopher Aristotle fi rst suggested the contribution of variable of the intelligence that intellect could be assessed by a single mental level of the person delivering the psychotherapy ability variable, nous (Detterman 1982 ). As to psychotherapy outcome but admittedly that Aristotle was also the tutor of Alexander the would be a diffi cult study to conduct for multiple Great, Aristotle clearly had some academic expe- reasons. In addition, rehabilitative success of 23 Common and Variable Aspects of Intelligence 369 brain-injured neurological patients is best pre- complex behaviors, but lower-level human brain dicted by the pre-morbid intelligence level of the functions (i.e., sensorimotor functions) are local- individual patients (Golden 1978 ). ized in a specifi c area of the brain (Reynolds It is noteworthy that the abstract concept of g 1981 ). Higher-level human brain functions are has been considered limited by the biological process specifi c, and processing of information integrity and physiological effi ciency of the requires coordination of diverse neuroanatomical human brain (Brand 1996; Vernon 1998). brain sections (Ashman and Das 1980 ). Higher- Harmony ( 1997) and Languis and Miller ( 1992 ) level human brain organization (Luria 1973 ) fur- have suggested that physiological measures, such ther was characterized as the brain’s higher-level as the EEG and/or auditory evoked potentials, processing being organized into three major could be utilized to assess aspects of cognitive human brain areas. The fi rst human brain area ability. Jensen’s research (1978 , 1998 ) on reaction included the brainstem and reticular formation, times and evoked potentials had suggested that g the midbrain, pons, and medulla. The second could be conceptualized as general physiological human brain area included the parietal, occipital, effi ciency of the central nervous system. Future and temporal lobes (Luria 1973 ). The third understanding of the concept of g will require elu- human brain area included all of the cerebral cor- cidation of the method and components of infor- tex anterior to the sensory-motor strip (i.e., mation processing in the brain. It is possible that Rolandic fi ssure). The three major higher-level further elucidation of brain-behavior models may brain areas (Luria 1973 ) all function in a dynamic contribute to understanding of g . Consideration of reciprocal interaction to subserve higher-level the relationship of g to contemporary models of cognitive processing, or in other words the brain-behavior may prove very helpful. higher-level processing depends on multiple diverse areas of the human brain (Reynolds 1981 ). As earlier noted, lower-level functions are Luria’s Brain-Behavior Model more hardwired to specifi c neuroanatomical brain areas. Alexander R. Luria, a Soviet neurologist and The notion of the brain as a dynamic func- neuropsychologist, had important insights into tional system, it should be acknowledged, was brain functioning (Horton 1987). Perhaps the fi rst proposed by Hughlings Jackson, an English most important insight was the concept of the physician who lived in the nineteenth century, complex functional system (i.e., multiple diverse and was further elucidated by Luria (Horton brain areas subserve particular behavioral abili- 1987 ). Essentially, higher mental processes are ties) (Horton 1987 ). Using the cultural-historical seen as based on multiple diverse human brain theory of brain-behavior relationship, Luria was areas communicating and working together, and able to perform an evaluation of an individual’s as a result higher-level functions may be dis- neurological status (Horton 1987 ). Grossly over- rupted by the destruction of a communication simplifi ed, it could be averred that he (Luria channel of the functional system (Luria 1964 ). 1973 ) described sensory and motor functions of Further, disturbances of higher-level mental the brain as having highly specifi c functional functions can be infl uenced based on the specifi c localizations, while higher-level mental pro- localization of the brain damage (Luria 1964 ). cesses required coordination of multiple areas of Therefore, rehabilitation of the human brain the brain. In other words, lower-level functions higher-level functional system, if there is spe- were hardwired in specifi c neuroanatomical areas cifi c brain damage, will require the brain to but higher-level functions were widely distrib- assemble an alternative sequence of human brain uted throughout the human brain. Put still another areas working together to perform specifi c way, higher-level human brain functions require behavioral tasks in a new way (Luria (1964 ). The multiple areas of the brain to accomplish localizing brain area responsible for behavioral 370 A.M. Horton Jr. and C.R. Reynolds disturbance can be determined by qualitatively lobes and organizes the information into analyzing the diffi culty experienced in perform- higher- level cognitive processes subserving coming a specifi c behavioral task (Luria 1964 ). plex human behavior.

Neuroanatomical Area One Neuroanatomical Area Three

The fi rst neuroanatomical area of the human The frontal lobes which involves the initiation, brain, the brain stem, subserves maintaining con- development, and monitoring of plans for behav- sistent arousal, attention, and concentration abili- ior are included in the third neuroanatomical area ties. The energy level and tone of the entire human of the human brain. In other words, frontal lobes, cerebral cortex allow a stable platform to organize the third neuroanatomical area, receive and eval- the various higher-level cognitive functions of the uate organized sensory input from the fi rst and human brain. The brain stem, fi rst neuroanatomi- second neuroanatomical areas of the human brain cal area, includes the reticular formation, the pos- and perform executive functions integrating the terior hypothalamic and brainstem portions of the information to subserve complex adaptive prob- brain. Damage to the fi rst neuroanatomical area of lem solving in a managerial role (Luria 1973 ; the human brain can cause lowering of the level of Obrzut and Obrzut 1982). The frontal lobes, in consciousness in the human cerebral cortex, dis- addition to direct connections to the second neu- rupting higher-level cognitive functioning thereby roanatomical areas, are also directly connected to giving rise to disorganized behavior. the reticular formation in the fi rst neuroanatomical area of the brain. This series of reciprocal communication neural networks mediates the Neuroanatomical Area Two activation and processing of higher-level cognitive processing throughout the human cerebral The area posterior to the central sulcus (i.e., pari- cortex. Performing an executive function, the etal, occipital, and temporal lobes) is included in frontal lobes direct attention and concentration the second neuroanatomical area of the brain. processes in the human brain. The direct connec- The second neuroanatomical area of the human tions among the fi rst, second, and third neuroana- brain is primarily receptive in nature, integrating tomical areas of the human brain facilitate diverse sensory inputs, storing, integrating, and reciprocal neural network communication sys- organizing sensory information. The second neu- tems that facilitate complex human decision roanatomical area of the brain allows perception, making and adaptive problem solving based on analysis, and synthesis of sensory stimuli (e.g., arousal, attention, and organized sensory input. auditory in the temporal lobes, visual in the The coordination of fi rst neuroanatomical area of occipital lobes, and tactile in the parietal lobes). the brain with the second and third neuroanatom- Within the second neuroanatomical area of the ical areas thereby facilitates the initiation, devel- brain, each lobe sensory stimuli processing (audi- opment, and monitoring of behavioral plans and tory in the temporal lobes, visual in the occipital their timely, effi cient, and effective evaluation. In lobes, and tactile in the parietal lobes) is orga- contrast with arousal role of the fi rst neuroana- nized into three hierarchical zones. The primary tomical area and the receptive role of the second zone perceives and retains incoming sensory neuroanatomical area, the third neuroanatomical stimuli. The secondary zone analyzes and orga- area of the brain has an expressive, generative nizes sensory information from the primary zone . role. In a nutshell, it is noted that human higher- The tertiary zone receives sensory information level cognitive functioning is facilitated by the (auditory in the temporal lobes, visual in the dynamic and reciprocal interplay of the three occipital lobes, and tactile in the parietal lobes) neuroanatomical area of the brain (Luria 1964 ; from the multiple secondary zones of the three Golden et al. 1979 ; Golden 1987 ; Horton 1987 ). 23 Common and Variable Aspects of Intelligence 371

Simultaneous and Successive Successive Processing Cognitive Processes In contrast, successive (or sequential) processing is linear accessing of information in a serial fash- Further elucidation of the functioning of the sec- ion (Das et al. 1979 ). In the second neuroanatom- ond neuroanatomical area (Luria 1964 ) involves ical area, the left temporal lobe of the brain appreciation of modes of information processing. subserves successive (or sequential) processing These can be characterized usefully as simultane- (Naglieri et al. 1983 ; Willis 1985 ). The succes- ous and successive (or sequential) cognitive pro- sive (or sequential) processing requires the main- cesses. Put another way, sensory information can tenance of the temporal order of input of be processed in sequence or one element at a time information (Naglieri et al. 1983 ). An example of in order or simultaneously where all of the infor- successive (or sequential) processing might mation is processed as a whole or as a gestalt, in include learning to read using a phonetic approach other words, describing a group of trees as oak, (Gunnison et al. 1982 ). That is not to say that pine, birch, etc., or as a forest. Simply put, sen- reading cannot be accomplished by simultaneous sory stimuli, in the second neuroanatomical area, processing such as the whole word approach, but can be processed through either simultaneous or rather with a phonetic approach, successive (or successive means (Kaufman 1979b ). Simultaneous sequential) processing is more effi cient. and successive processes can be used to process any specifi c sensory modality (i.e., auditory, visual, tactual, etc.) or stimulus elements (verbal, Hemispheric Specialization nonverbal) (Ashman and Das 1980 ). Which type and Simultaneous and Successive of processing, either simultaneous or successive Cognitive Processes cognitive processing, is most effi ciently effective will depend on the task demands, attention Different cerebral hemispheres are thought to be demands required by the task, and preferred more effi cient with either simultaneous or suc- means for completing the task (Hall et al. 1988 ; cessive processing (Naglieri et al. 1983). The left Watters and English 1995 ; Willis 1985 ). Verbal cerebral hemisphere may be more effi cient in communications may be processed effectively performing linguistic, serial, and analytic tasks. through linear successive methods such as dictat- The right hemisphere may be more effi cient in ing or writing a letter. Spatial tasks, such as map performing visual-spatial and gestalt-holistic reading, may be processed effectively through tasks (Bever 1975 ; Bogen 1969 ; Dean and simultaneous- processing strategies. Or to use Reynolds 1997; Gazzaniga 1970 ; Harnad et al. another example, a forest ranger might know each 1977 ; Kinsbourne 1978 , 1997 ; Naglieri et al. type of tree in a forest, but a hiker would be more 1983 ; Schwartz et al. 1975 ; Segalowitz and concerned with the concept of forest. Gruber 1977 ; Willis 1985 ). Modes of information processing appear likely related to hypothe- Simultaneous Processing sized differences in cerebral hemispheric This is the synthesis of separate elements into processing, and as earlier mentioned, the advan- spatially related groups with direct access to tage is that one cerebral hemisphere may be more any separate element (Das et al. 1979 ). Within effi cient in processing particular stimuli, but that the second neuroanatomical area, the right does not mean that the other cerebral hemisphere occipital and parietal lobes of the human brain cannot also process that same stimuli but rather a subserve simultaneous information processing relative degree of effi ciency may be lost. For (Naglieri et al. 1983 ; Willis 1985 ). Commonly example, there are persons who read using a considered measures of simultaneous process- whole word rather than a phonetic approach. ing can include visual-spatial ability tests Indeed, not all languages are phonetically based, (Kirby and Das 1977 ). so whole word approaches are essential in some 372 A.M. Horton Jr. and C.R. Reynolds languages. Utilization of specifi c cognitive pro- Sonnier 1992 ; Sonnier and Goldsmith 1985 ). cessing modes may optimize the effi ciency of Research on intelligence reviewed thus far has these hemispheric brain functions. focused on intelligence as a single factor g , a Cerebral hemispheric asymmetries of func- brain-based behavior model, and different modes tioning, as previously pointed out, are relative of cognitive information processing. preferences for process-specifi c strategies rather than stimulus-specifi c strategies . The mode of higher-level cognitive processing for task perfor- Halstead’s Theory of Biological mance depends on multiple factors which Intelligence include, but are not limited to, specifi c task demands, level of attention required for the task, Simply put, theoretical interest in the human individual cognitive abilities, genetics, and cul- mental abilities subserved by the human frontal tural traditions (Cumming and Rodda 1985 ; Hall lobes was the focus of Ward Halstead’s research et al. 1988 ; McCallum and Merritt 1983 ; Watters program (Halstead 1947). This research interest and English 1995 ; Willis 1985 ). The need for also included the concept of intelligence specifi c types of manipulation of stimuli can also (Halstead 1947 ). It might be noted that Boring be a reason for selection of a specifi c hemispheric (1930 ) has considered the concept of intelligence (e.g., Dean 1984 ; Grimshaw 1998 ; Mateer et al. as what intelligence tests measured which is, of 1984 ; Obrzut et al. 1985 ; Ornstein et al. 1980 ; course, tautological. Piccirilli et al. 1991 ; Tous et al. 1995 ). Halstead accepted Boring’s defi nition of intelligence as psychometric intelligence which was postulated to be what was measured by the intel- Hemisphericity and Cognitive ligence tests (Halstead (1947 ). In contrast, how- Processing ever, Halstead also conceptualized a type of intelligence that was different from psychometric Reynolds (1981 ) conceptualized hemisphericity intelligence (Halstead 1947 ). as preference for cognitive information- Biological intelligence as conceptualized by processing style independent of cerebral domi- Halstead (1947 ) was human adaptive abilities as nance. Hemisphericity can be defi ned as the subserved by an intact uninjured brain, in other tendency of an individual to rely differentially on words, human adaptive abilities that were signifi - the higher-level information-processing style of cantly impaired following brain damage. one cerebral hemisphere (Reynolds 1981 ). The concept of biological intelligence was Previous research appears to be essentially con- hypothesized in response to perceived limitations sistent with hemispheric specialization (Dean of intelligence tests. Halstead (1947 ) observed and Reynolds 1997). Optimal higher-level cogni- that in many cases, patients who had brain injuries tive functioning may require utilization of both were still able to score well on intelligence tests modes of information and also being able to shift despite clear brain damage and signifi cant adap- the cognitive information-processing mode in tive behavior problems in daily living. Halstead response to multiple factors (Gazzaniga 1974 , (1947 ) conceptualized that there was an addi- 1975 ). At the same time, dysfunctional hemi- tional brain-based latent construct that was sensi- sphericity may impede optimal higher-level cog- tive to human adaptive abilities but poorly nitive functioning (Newell and Rugel 1981 ; evaluated by intelligence tests. In other words, Roubinek et al. 1987 ). Research, over many Halstead agreed with David Wechsler’s defi nition years, has demonstrated that identifying the pre- (1944 ) of intelligence as “the aggregate or global ferred mode of cognitive information processing capacity of the individual to act purposefully, to (hemisphericity) may be advantageous in terms think rationally, and to deal effectively with his of addressing and remediating academic learning environment” (p. 3) but found then contemporary problems (Faust et al. 1993 ; Gunnison et al. intelligence tests inadequate to satisfy David 1982 ; Paquette et al. 1996 ; Roubinek et al. 1987 ; Wechsler’s defi nition ( 1944 ) of intelligence and 23 Common and Variable Aspects of Intelligence 373 sought to fi nd a brain-based latent construct that included the Halstead Tactual Performance would better satisfy Wechsler’s defi nition ( 1944). Test (speed component) and the Halstead In order to research the latent construct of bio- Dynamic Visual Field Test (peripheral com- logical intelligence, Halstead (1947 ) established an ponent) (Halstead 1947 ). The fi rst three fac- experimental brain-behavior research program at tors, C, A, and P, were interpreted as process the University of Chicago Medical School focused factors of biological intelligence, and D was on studying the biological integrity of the human interpreted as the factor through which expres- brain (Horton and Wedding 1984 ). Halstead (1947 ) sions of factors C, A, and P were directed. developed a number of sensitive measures to the It is noteworthy that both factors C and A had behavioral defi cits of brain-injured persons which had signifi cant loadings from intelligence tests but Halstead postulated represented the abstract con- the intelligence tests loaded on different factors. cept of biological intelligence and were distinct The tests measuring the four factors signifi cantly from psychometric intelligence. Of particular inter- differentiated between individuals with docu- est is Halstead’s factor analytic attempt to identify mented head injury and individuals with no docu- the aspects of higher cognitive functions that were mented history of head injury (Halstead 1947). involved in biological intelligence (Halstead 1947 ). Moreover, an average of the measures (the Halstead (1947 ) extracted four basic factors of Halstead Impairment Index) was the best mea- biological intelligence, and these factors are sure in differentiating these individuals (Halstead described below: 1947 ). Unfortunately, as the concept of biologi- C , the integrative fi eld factor. The ability to adapt cal intelligence was postulated to be related to the to new situations and to integrate new infor- integrity of the frontal lobes, subsequent experi- mation was postulated as the integrative fi eld mental research studies couldn’t cross-validate a factor (Reitan 1994 ). Tests that had loadings relationship between the frontal lobes and HII on factor C included the Halstead Category (Reitan 1975 ), thus failing to confi rm the concept Test, the Henmon-Nelson Tests of Mental of biological intelligence. In addition, as previ- Ability, the Speech-Sounds Perception Test, ously noted, tests of intelligence did load on fac- the Halstead Finger Oscillation Test, and the tors extracted from Halstead’s tests so the latent Halstead Time-Sense Test (Halstead 1947 ). construct biological intelligence appeared to A , the abstraction factor. The ability to draw overlap rather than be orthogonal to psychomet- meaning from a series of events or to hold in ric intelligence. Interestingly, Halstead’s tests mind abstract nonverbal ideas without the use were better able to differentiate brain damaged of past experience. Tests that had loadings on from normal subjects than intelligence tests alone factor A included the Carlo Hollow-Square but exactly why remains elusive. Performance Test for Intelligence, the Reitan (1994 ) had validated a modifi ed and Halstead Category Test, the Halstead Tactual augmented neuropsychological test battery based Performance Test (memory component), and on Halstead’s tests as a core to improve diagnos- the Halstead Tactual Performance Test (local- tic accuracy (Hevern 1980 ; Reed 1985 ; ization component) (Halstead 1947 ). Swiercinsky 1979 ). It is noteworthy that formal P , the power factor . The reserve power available intelligence testing has always included as an to an amplifi er not already functioning at peak integral portion of Reitan’s comprehensive clini- ability was postulated to be the power factor. cal neuropsychological test battery, in addition to Tests that had loadings on factor P included Halstead’s core tests and a number of additional the Halstead Flicker-Fusion Test, the Halstead test procedures added to assess brain areas not Tactual Performance Test (recall component), related to intelligence tests. It is noteworthy that the Halstead Dynamic Visual Field Test (central Halstead’s factor structure (Horton and Wedding form), and the Halstead Dynamic Visual Field 1984 ) was very similar to the factor structure Test (central color) (Halstead 1947 ). found for the age-appropriate Wechsler scales D , the directional factor . An attentional compo- (Kamphaus 2001 ; Kaufman 1994 ). Basically, the nent. Tests that had loadings on factor D contemporary Wechsler scales have a factor 374 A.M. Horton Jr. and C.R. Reynolds structure which consists of verbal comprehen- primary grades. They include the Wechsler Adult sion, perceptual organization, working memory, Intelligence Scale – Fourth Edition (WAIS-IV) and processing speed factors. In other words, (Wechsler 2008 ), the Wechsler Intelligence Scale very similar to the four factors identifi ed by for Children – Fourth Edition (WISC-IV) Halstead (1947 ). (Wechsler 2003 ), and the Wechsler Preschool Subsequent factor analysis studies with neuro- and Primary Scale of Intelligence – Third Edition psychological tests have produced comparable (WPPSI-III) (Wechsler 2002 ) and all allow results. A few examples are cited. A study with examiners to compute full-scale IQs. The adult neuropsychiatric patients (Fowler et al. WAIS-IV is designed to assess from ages 16 to 1988 ) extracted fi ve factors (verbal comprehen- 90 and 11 months, the WISC-IV is designed to sion, perceptual organization, sensory attention, assess from ages 6 to 16 and 11 months, and the primary motor, and tactual-spatial abilities). WPPSI-III is designed to assess from age 2 years Also, a study of children aged 9–14 (Brooks et al. and 6 months to age 7 years and 3 months. In 1989 ) extracted 4 factors (simple motor, tactile terms of factor structures as recommended for kinesthesis, memory/attention, and nonverbal clinical interpretation in the technical and inter- visual-spatial memory). In addition, a study with pretation manuals, there is a great deal of similar- children aged 9–12 (Francis et al. 1992 ) extracted ity. For the WAIS-IV (Wechsler 2008 ), the 5 factors (simple motor skill, complex visual- recommended factor structure forms the bases spatial relations, simple spatial motor operations, for the Verbal Comprehension Index, the motor steadiness, and speeded motor sequenc- Perceptual Reasoning Index, the Working ing). Moreover, a study with younger children Memory Index, and the Processing Speed Index. aged 5–7 (Foxcroft 1989 ) extracted six factors Similarly for the WISC-IV (Wechsler 2003 ), the (analytic-synthetic visual motor ability, percep- recommended factor structure includes again the tual organization, cross-modality motoric effi - Verbal Comprehension Index, the Perceptual ciency, directed motor speed, patterned critical Reasoning Index, the Working Memory Index, discrimination, and strength). Briefl y put, multi- and the Processing Speed Index. Essentially both ple factor analysis studies appear to extract fac- the WAIS-IV and WISC-IV are reported to have tors which are relatively similar to factors found the same factor structure. For the WPPSI-III from the age-appropriate Wechsler scales (Wechsler 2002 ), however, there are differences (Kamphaus 2001 ; Kaufman 1994 ). depending on the age of the child. Essentially Therefore, conceptualizations of intelligence from ages 2 years and 6 months to 3 years and 11 that are consistent with David Wechsler’s (1944 ) months, a two-factor model is recommended defi nition of intelligence as “the aggregate or with a verbal intelligence quotient (VIQ) and per- global capacity of the individual to act purpose- formance intelligence quotient (PIQ). From age 4 fully, to think rationally, and to deal effectively to 7 years and 3 months, a three-factor model is with his environment” might be seen as loading recommended with a verbal intelligence quotient on the above factors. The relationship with g and (VIQ) and performance intelligence quotient the above factors needs to be reconciled. A pos- (PIC) and a processing speed quotient (PSQ). In sible answer, however, will be addressed in the summary, all of the Wechsler scales of intelli- next sections. gence show verbal and performance factors at every age level. For the WPPSI-III, a two-factor model of verbal and performance is preferred Contemporary Wechsler Scales from ages 2 years and 6 months to 3 years and of Intelligence 11 months; in a three-factor model of verbal performance and processing, speed is preferred from There are three contemporary Wechsler intelli- ages 4 to 7 years and 3 months. For the WISC-IV gence scales designed to assess adults, school- and WAIS-IV, four-factor models (verbal compre- aged children, and children in preschool and hension, perceptual reasoning, working memory, 23 Common and Variable Aspects of Intelligence 375 and processing speed) are preferred. As earlier Horn-Carroll (CHC) theory (McGrew 2009 ). In mentioned, all of the Wechsler scales allow the Carroll’s (1993 ) theory, the third stratum is uni- computation of a full-scale IQ. Recent research tary or, put another way, is composed of one con- studies (Canivez and Watkins 2010 ) supported struct only, g as previously described. Multiple the WAIS-IV as a measure of general intelligence studies of human intelligence have found that g but noted the remaining factor structure accounted accounts for the major portion of variance for small portions of total and common variance. assessed by intelligence test batteries. Similarly, Benson et al. ( 2010 ) suggested that a Cattell- intelligence tests are strong and consistent pre- Horn- Carroll (CHC) structure provides a better dictors of very important social outcomes, such description of test performance with abilities that as academic achievement (Binet and children in include crystallized ability (Gc), fl uid reasoning the Paris Public Schools) and occupational per- (Gf), visual processing (Gv), short-term memory formance (Wechsler and offi cer candidates in the (Gsm), and processing speed (Gs). Moreover, US Army in World War I and in addition the Weiss et al. ( 2013a) found that either a four- or SATs, GREs, etc.). The predictive ability is fi ve-factor structure fi ts the data, but a fi ve-factor directly related to the amount of g measured by structure was a better fi t with a quantitative rea- the intelligence test. Simply put, intelligence soning (RQ) factor included. tests with greater amounts of g are signifi cantly For the WISC-IV, Keith et al. (2006 ) sug- better predictors of important outcomes in soci- gested the scoring structure was not supported ety than are intelligence tests with lower amounts and the Cattell-Horn-Carroll (CHC) theory was a of g . Clearly intelligence tests with large amounts better fi t, but Watkins (2010 ) found the four fi rst- of g have important purposes in society, espe- order factors as suggested by the WISC-IV test cially in terms of prediction of success in aca- manual (Wechsler 2003 ). More recently, Weiss demic and occupational settings. It is a conundrum et al. (2013) found that either a four- or fi ve- that while the psychometric concept of g has factor structure fi ts the data and both were suit- proven useful in society for over a century, the able, and the fi ve-factor model included inductive full understanding of the latent concept of g reasoning (IR). remains elusive and is not yet completely under- A common concern was that the g factor was stood by psychologists even after a century of the majority of the variance and other factors research and clinical application. The CHC the- were quite small. ory posits various types of g such as Gc and Gf among others. Carroll’s (1993 ) second stratum of traits is Carroll’s Theory of Intelligence composed of combinations of stratum one measures and second-stratum measures that combine Carroll’s (1993 ) three-stratum theory of intelli- to form the third stratum. Typically, stratum one gence has averred that the latent traits tapped by measures are more specifi c traits of interest. intelligence tests are independent of the specifi c Stratum one measures are combined to become test battery. Carroll (1993 ) has postulated that stratum two measures and result in enhanced numerous mental ability tests measured the same measurement of complex higher-level cognitive abilities which Carroll labeled crystallized, traits such as verbal and nonverbal intelligence. visual-perceptual, and memory abilities. In inter- Similarly, stratum two measures are then in a preting extant research fi ndings, Carroll ( 1993 ) hieratical fashion combined to allow for the mea- has proposed there are three strata of intelligence. surement of a complex stratum three trait, such as An important feature is the reconciliation of pre- the latent construct of intelligence or g . Concepts vious research results related to the assessment of such as fl uid intelligence, crystallized intelli- human intelligence by combining the Cattell- gence, general memory and learning, broad Horn notion of crystallized G (Gc) and fl uid G visual perception, broad auditory perception, and (gf) with the Carroll paradigm into the Cattell- processing speed are examples of second-stratum 376 A.M. Horton Jr. and C.R. Reynolds traits (Carroll 1993 ). Multiple research studies the CHC model has found that latent mental traits appear to suggest second-stratum traits can be are test battery independent and numerous tests ranked in terms of their abilities to assess g measured the same latent mental traits tapped by (Kamphaus 2001 ). Second-stratum traits which intelligence tests. Multiple research studies have involve reasoning abilities are better measures of found g accounts for the major portion of vari- g. Examples of second-stratum traits that involve ance assessed by intelligence test batteries. Also, abstraction abilities might be seen as general Carroll’s (1993 ) second stratum consists of sequential reasoning, induction, deduction, syl- higher-level traits such as verbal and nonverbal logisms, series tasks, matrix reasoning, analogies intelligence (Reynolds and Kamphaus 2003 ) that and quantitative reasoning, etc. (Carroll 1993 ). are assessed by combinations of stratum one An example of a contemporary intelligence test measures. Stratum one measures are typically that uses the CHC theory as a basis is the single subtests that measure a trait of interest and Reynolds Intellectual Assessment Scale (RIAS) can be combined to form stratum two measures which will be described in the next section. and measure higher-level cognitive abilities such as fl uid intelligence, crystallized intelligence, general memory and learning, broad visual per- Reynolds Intellectual Assessment ception, broad auditory perception, and process- Scale (RIAS) ing speed. In turn, stratum two measures are combined into a complex stratum three trait, such The Reynolds Intellectual Assessment Scale as general intelligence as conceptualized as g . (RIAS) (Reynolds and Kamphaus 2003 ) follows Therefore, intelligence can be conceptualized the more contemporary Carroll (1993 ) theoretical on multiple theoretical levels. Intelligence can be model of intelligence model. The RIAS has dem- seen as represented by a single score that has onstrated impressive evidence for its interpreta- impressive predictive abilities, different cognition as a measure of intelligence (i. e., validity) tive processing modes that have implications for (Beaujean et al. 2010 ) as well as being time effi - higher cognitive functioning and multiple more cient, user-friendly for administration and scoring, specifi c higher-level cognitive ability factors that and not having a disparate impact when used to represent less comprehensive important cognitive assess members of minority groups, different gen- skills (Carroll 1993 ). Relative to the most appro- ders, or groups of clinical patients. The RIAS pro- priate conceptualization of the latent trait of posed two-factor structure (verbal intelligence and intelligence, the two-factor model exemplifi ed by nonverbal intelligence) has been cross-validated a the RIAS appears the right choice. As previously number of times (Nelson et al. 2007 ; Dombrowski mentioned, it appears the greater amount of g et al. 2009 ; Nelson and Ganivez 2012 ). accounted for the theoretically more appropriate measure of intelligence. The contemporary intelligence test that maximizes the utilization of g is Discussion: Common and Variable the RIAS. It should be recalled that for the Aspects of Intelligence WPPSI-III, the younger age has only two factors and for the older age of the WPPSI-III and for the The concept of intelligence appears clearly WISC-IV and WAIS-IV, the later-appearing fac- related to the biological integrity of the brain tors such as working memory and processing (Luria 1973 ). Perhaps not solely to the frontal speed generally account for lesser amounts of g . lobes alone (Reynolds and Horton 2006 ), but In the earlier discussion of Halstead’s factor anal- clearly intelligence is related to optimal human ysis of Halstead’s neuropsychological tests, it brain functioning (Reitan 1994 ). This chapter has was noted that the fi rst two factors included intel- demonstrated that the concept of intelligence can ligence tests of the day. Indeed, the WISC-IV and be conceptualized in multiple ways. Carroll’s WAIS-IV (Wechsler 2008 ) now have a measure (1993 ) three-stratum theory of intelligence and known as the General Ability Index (GAI) which 23 Common and Variable Aspects of Intelligence 377 is a composite score of the three subtests that References make up the VCI and the three subtests that make up the PRI. The GAI is proposed to be used when Aluja-Fabregat, A., Colom, R., Abad, F., & Juan-Espinosa, working memory and processing speed measures M. (2000). Sex differences in general intelligence may have been impaired due to neuropsychologi- defi ned as g among young adolescents. Personality cal problems (Wechsler 2008 ). and Individual Differences, 28 , 813–820. Ashman, A. F., & Das, J. P. (1980). Relation between In other words, the two-factor solution is a planning and simultaneous-successive processing. superior measure than FSIQ with these clinical Perceptual and Motor Skills, 51 , 371–382. groups (Wechsler 2008). Thinking back to the Beaujean, A. A., Firmim, M. W., Michonski, J. D., & earlier mentioned studies of factor analyses of Berry, T. (2010). A multitrait-multimethod examination of the Reynolds Intellectual Assessment Scales in neuropsychological test batteries, it would seem a college sample. Assessment, 17(3), 347–360. that human mental abilities in excess of a two- Benson, N., Hulac, D. M., & Kranzler, J. H. (2010). factor solution such as the RIAS and GAI might Independent examination of the Wechsler Adult be better characterized as neuropsychological Intelligence Scale-Fourth Edition (WAIS-IV): what does the WAIS-IV measure? Psychological abilities rather than intelligence (Reynolds and Assessment, 22 (1), 121–130. Kamphaus 2003 ). Simply put, the common struc- Bever, T. G. (1975). Cerebral asymmetries in humans are ture of intelligence is composed of g , and the due to the differentiation of two incompatible pro- most g loaded two factors (verbal intelligence cesses: Holistic and analytic. In D. Aaronson & R. Reiber (Eds.), Developmental psycholinguistics and nonverbal intelligence) are the best approxi- and communication disorders . New York: New York mation of the latent construct of intelligence and Academy of Sciences. also are the most appropriate basis for a contem- Binet, A., & Simon, T. (1905). New methods for the diag- porary comprehensive intelligence test. nosis of the intellectual level of subnormals. L'Année Psychologique, 11 , 191–244. Development of the various and common Binet, A., & Simon, T. (1908). The development of intel- aspects of intelligence involve brain mechanisms ligence in the child. L'Année Psychologique, 14 , 1–90. assisted by cultural-historical experiences, as Bogen, J. E. (1969). The other side of the brain: Parts I, II, suggested by Luria (1966 , 1973 ). Intelligence has and III. Bulletin of the Los Angeles Neurological Society , 34 , 73–105, 135–162, 191–203. been conceptualized as certainly infl uenced by Boring, E. G. (1929). History of experimental psychology. the person’s environmental history (i.e., for a dis- New York: Genesis Publishing. cussion of Luria’s Cultural-Historical Theory, see Brand, C. (1996). Doing something about g. Intelligence, Horton 1987 , Reynolds 1981 ) but also with 22 , 311–326. Brooks, D. A., Dean, R. S., & Gray, J. W. (1989). HRNB genetic infl uences mediating the functional factors with simultaneous-successive marker vari- development of the various anatomical structures ables. International Journal of Neuroscience, 46 (3–4), of the brain. Intelligence appears related to an 157–160. individual’s ability to adapt to various life cir- Canivez, G. L., & Watkins, M. W. (2010). Investigation of the factor structure of the Wechsler Adult intelligence cumstances (Pallier et al. 2000 ). Further develop- Scale-Fourth Edition (WAIS-IV): Exploratory and ments of the theoretical foundations of higher order factor analysis. Psychological intelligence appear likely to continue to elucidate Assessment, 22 (4), 827–836. how the human brain carries out higher-order Carroll, J. B. (1993). Human cognitive abilities: A survey of factor analytic studies. New York: Cambridge cognitive functioning. A number of excellent new University Press. measures of intelligence have been developed in Cumming, C. E., & Rodda, M. (1985). The effects of audi- the past two decades (e.g., Kaufman and Kaufman tory deprivation on successive processing. Canadian 1983; Naglieri et al. 2013 ; Reynolds and Journal of Behavioural Science, 17 , 232–245. Das, J. P., Kirby, J. R., & Jarman, R. F. (1979). Kamphaus 2003 ), but additional research related Simultaneous and successive cognitive processes . to the elaboration of the latent concept of intelli- New York: Academic. gence is needed. The pace of new knowledge is Dean, R. S. (1984). Functional lateralization of the brain. expected to increase and more differentiated and Journal of Special Education, 18(3), 239–256. Dean, R. S., & Reynolds, C. R. (1997). Cognitive process- complex understanding of the latent concept of ing and self-report of lateral preference. human intelligence is expected. Neuropsychology Review, 7 , 127–142. 378 A.M. Horton Jr. and C.R. Reynolds

Detterman, D. K. (1982). Does “g” exist? Intelligence, 6 , Harnad, S., Doty, R. W., Goldstein, L., Jaynes, J., & 99–108. Krauthamer, G. (Eds.). (1977). Lateralization in the Dombrowski, S. C., Watkins, M. W., & Brogan, M. J. nervous system . New York: Academic. (2009). An exploratory investigation of the factor Hevern, V. W. (1980). Recent validity studies of the structure of the Reynolds Intellectual Assessment Halstead-Reitan approach to clinical neuropsycholog- Scales (RIAS). Journal of Psychoeducational ical assessment: A critical review. Clinical Assessment, 27 (6), 494–507. Neuropsychology, 2 , 49–61. Faust, M., Kravetz, S., & Babkoff, H. (1993). Horton, A. M., Jr. (1987). Luria’s contributions to clinical Hemisphericity and top-down processing of language. and behavioral neuropsychology. Neuropsychology, Brain and Language, 44 , 1–18. 1 (2), 39–44. Fowler, P. C., Zillmer, E., & Newman, A. C. (1988). A Horton, A. M., Jr., & Wedding, D. (1984). Clinical and multifactor model of the Halstead-Reitan neuropsy- behavioral neuropsychology . New York: Praeger. chological test battery and its relationship to cognitive Jensen, A. R. (1978, September). “g”:Outmoded concept status and psychiatric diagnosis. Journal of Clinical or unconquered frontier? Invited address at the annual Psychology, 44 (6), 898–906. meeting of the American Psychological Association, Foxcroft, C. D. (1989). Factor analysis of the Reitan- New York. Indiana neuropsychological test battery. Perceptual Jensen, A. R. (1998). The g factor: The science of mental and Motor Skills, 69 (3 pt. 2), 1303–1313. ability . Westport: Praeger. Francis, D. J., Fletcher, J. M., Rourke, B. P., & York, M. J. Kamphaus, R. W. (2001). Clinical assessment of child and (1992). A fi ve-factor model for the motor, psychomo- adolescent intelligence (2nd ed.). Boston: Allyn and tor and visual-spatial tests used in the neuropsycho- Bacon. logical assessment of children. Journal of Clinical and Kane, H. D. (2000). A secular decline in Spearman’s g : Experimental Neuropsychology, 14 (4), 625–637. Evidence from the WAIS, WAIS-R and WAIS-III. Gazzaniga, M. S. (1970). The bisected brain. New York: Personality and Individual Differences, 29 , 561–566. Appleton. Kaufman, A. S. (1979b). Cerebral specialization and Gazzaniga, M. S. (1974). Cerebral dominance viewed as a intelligence testing. Journal of Research and decision system. In S. Dimond & J. Beaumont (Eds.), Development in Education, 12 , 96–107. Hemisphere functions in the human brain . London: Kaufman, A. S. (1994). Intelligent testing with the WISC- Halstead Press. III . New York: Wiley. Gazzaniga, M. S. (1975). Recent research on hemispheric Kaufman, A. S., & Kaufman, N. L. (1983). Administration lateralization of the human brain: Review of the split- and scoring manual for the Kaufman assessment battery brain. UCLA Educator , May , 9–12. for children . Circle Pines: American Guidance Service. Golden, C. J. (1978). Diagnosis and rehabilitation in clin- Keith, T. Z., Fine, J. G., Taub, G. E., Reynolds, M. R., & ical neuropsychology . Springfi eld: Charles C. Thomas. Kranzler, J. H. (2006). Higher order multi-sampled Golden, C. J. (1987). Luria-Nebraska neuropsychological confi rmatory factor analysis of the Wechsler intelli- battery children’s revision . Los Angeles: Western gence scale children-fourth edition: What does it mea- Psychological Services. sure? School Psychology Review, 35 (1), 108–127. Golden, C. J., Purisch, A. D., & Hammeke, T. A. (1979). Kinsbourne, M. (1978). Biological determinants of func- The Luria-Nebraska neuropsychological test battery: tional bisymmetry an asymmetry. In M. Kinsbourne A manual for clinical and experimental uses . Lincoln: (Ed.), Asymmetrical function of the brain . Cambridge: The University of Nebraska Press. Cambridge University Press. Grimshaw, G. M. (1998). Integration and interference in Kinsbourne, M. (1997). Mechanisms and development of the cerebral hemispheres: Relations with hemispheric cerebral lateralization in children. In C. R. Reynolds & specialization. Brain and Cognition, 36 , 108–127. E. Fletcher-Janzen (Eds.), Handbook of clinical child Gunnison, J., Kaufman, N. L., & Kaufman, A. S. (1982). neuropsychology (2nd ed., pp. 102–119). New York: Reading remediation based on sequential and simul- Plenum Press. taneous processing. Academic Therapy, 17 , Kirby, J. R., & Das, J. P. (1977). Reading achievement, 297–306. IQ, and simultaneous-successive processing. Journal Hall, C. W., Gregory, G., Billinger, E., & Fisher, T. (1988). of Educational Psychology, 69 , 564–570. Field independence and simultaneous processing in Languis, M. L., & Miller, D. C. (1992). Luria’s theory of preschool children. Perceptual and Motor Skills, 66 , brain functioning: A model for research in cognitive 891–897. psychophysiology. Educational Psychologist, 27 , Halstead, W. C. (1947). Brain and intelligence . Chicago: 493–511. University of Chicago Press. Luborsky, L., Auerbach, A. H., Chandler, M., Cohen, J., Harmony, T. (1997). Psychophysiological evaluation of & Bachrach, H. M. (1971). Factors infl uencing the neuropsychological disorders in children. In C. R. outcome of psychotherapy: A review of quantitative Reynolds & E. Fletcher-Janzen (Eds.), Handbook of research. Psychological Bulletin, 75 , 145–185. clinical child neuropsychology (2nd ed., pp. 356–370). Luria, A. R. (1964). Neuropsychology in the local diagno- New York: Plenum Press. sis of brain damage. Cortex, 1 , 3–18. 23 Common and Variable Aspects of Intelligence 379

Luria, A. R. (1966). Higher cortical functions in man . chological assessment: Vol. III. San Francisco: New York: Basic Books. Jossey-Bass. Luria, A. R. (1973). The working brain . London: Penguin. Reitan, R. M. (1994). Ward Halstead’s contributions to Mateer, C. A., Rapport, R. L., & Kettrick, C. (1984). neuropsychology and the Halstead-Reitan neuropsy- Cerebral organization of oral and signed language chological test battery. Journal of Clinical Psychology, responses: Case study evidence from amytal and corti- 50 , 47–69. cal stimulation studies. Brain and Language, 21 , Reynolds, C. R. (1981). The neuropsychological basis of 123–135. intelligence. In G. Hynd & J. Obrzut (Eds.), McCallum, R. S., & Merritt, F. M. (1983). Simultaneous- Neuropsychological assessment and the school-aged successive processing among college students. Journal child . New York: Grune & Stratton. of Psychoeducational Assessment, 1 , 85–93. Reynolds, C. R., & French, C. (2003). Neuropsychological McGrew, K. S. (2009). CHC theory and the human cogni- basis of intelligence: Revisited. In A. M. Horton Jr. & tive abilities project: Standing on the shoulders of the L. C. Hartlage (Eds.), Handbook of forensic neuropsy- giants of psychometric intelligence research. chology . New York: Springer. Intelligence, 37 (1), 1–10. Reynolds, C. R., & Horton, A. M., Jr. (2006). Test of ver- Naglieri, J. A., Kamphaus, R. W., & Kaufman, A. S. bal conceptualization and fl uency: Examiner’s man- (1983). The Luria-Das simultaneous-successive model ual . Austin: PRO-ED. applied to the WISC-R. Journal of Psychoeducational Reynolds, C. R., & Kamphaus, R. W. (2003). Reynolds Assessment, 1 , 25–34. intellectual assessment scales . Odessa: PAR. Naglieri, J. A., Das, J. P., & Goldstein, S. (2013). Cognitive Roubinek, D. L., Bell, M. L., & Cates, L. A. (1987). Brain assessment system(2nd ed.). Austin, TX: ProEd. hemispheric preference of intellectually gifted chil- Nelson, J. M., & Canivez, G. L. (2012). Examination of dren. Roeper Review, 10 , 120–122. the structural, convergent, and incremental validity of Schwartz, G. E., Davidson, R. J., & Maer, F. (1975). Right the Reynolds Intellectual Assessment scale (RIAS) hemisphere lateralization for emotion in the human brain: with a clinical simple. Psychological Assessment, Interactions with cognition. Science, 190 , 286–288. 24 (1), 129–140. Segalowitz, S. J., & Gruber, F. A. (Eds.). (1977). Language Nelson, J. M., Canivez, G. L., Lindstrom, W., & Hatt, development and neurological theory . New York: C. V. (2007). Higher-order exploratory factor analysis Academic. of the Reynolds intellectual assessment scales with a Shure, G. H., & Halstead, W. C. (1958). Cerebral localiza- referred sample. Journal of School Psychology, 45 (4), tion of Intellectual processes. Psychological 439–456. Monographs: General and Applied, 72 (12), 1–40. Newell, D., & Rugel, R. P. (1981). Hemispheric special- Sonnier, I. L. (1992). Hemisphericity as a key to under- ization in normal and disabled readers. Journal of standing individual differences. In I. L. Sonnier (Ed.), Learning Disabilities, 14 , 296–298. Hemisphericity as a key to understanding individual Obrzut, J. E., & Obrzut, A. (1982). Neuropsychological differences (pp. 6–8). Springfi eld: Charles C. Thomas. perspectives in pupil services: Practical application of Sonnier, I. L., & Goldsmith, J. (1985). The nature of Luria’s model. Journal of Research and Development human brain hemispheres: The basis for some indi- in Education, 15 , 38–47. vidual differences. In I. L. Sonnier (Ed.), Methods and Obrzut, J. E., Obrzut, A., Bryden, M. P., & Bartels, S. G. techniques of holistic education (pp. 17–25). (1985). Information processing and speech lateraliza- Springfi eld: Charles C. Thomas. tion in learning-disabled children. Brain and Swiercinsky, D. P. (1979). Factorial pattern description Language, 25 , 87–101. and comparison of functional abilities in neuropsycho- Ornstein, R., Johnstone, J., Herron, J., & Swencionis, C. logical assessment. Perceptual and Motor Skills, 48 , (1980). Differential right hemisphere engagement in 231–241. visuospatial tasks. Neuropsychologia, 18 , 49–64. Tous, J. M., Fusté, A., & Vidal, J. (1995). Hemispheric Pallier, G., Roberts, R. D., & Stankov, L. (2000). specialization and individual differences in cognitive Biological versus psychometric intelligence: processing. Personality and Individual Differences, Halstead’s (1947) distinction revisted. Archives of 19 , 463–470. Clinical Neuropsychology, 15 , 205–226. Travers, R. M. W. (1977). Essentials of learning (4th ed.). Paquette, C., Tosoni, C., Lassonde, M., & Peretz, I. New York: Macmillan. (1996). Atypical hemispheric specialization in intel- Vernon, P. A. (1998). From the cognitive to the biological: lectual defi ciency. Brain and Language, 52 , 474–483. A sketch of Arthur Jensen's contributions to the study Piccirilli, M., D'Alessandro, P., Mazzi, P., Sciarma, T., & of g. Intelligence, 26 , 267–271. Testa, A. (1991). Cerebral organization for language Watkins, M. W. (2010). Structure of the Wechsler intelli- in Down’s syndrome patients. Cortex, 27 , 41–47. gence scale for children-fourth edition among national Reed, J. (1985). The contributions of Ward Halstead, sample of referred students. Psychological Assessment, Ralph Reitan and their associates. International 22 (4), 782–787. Journal of Neuroscience, 25 , 289–293. Watters, J. J., & English, L. D. (1995). Children’s applica- Reitan, R. M. (1975). Assessment of brain-behavior relation of simultaneous and successive processing in tionships. In P. McReynolds (Ed.), Advances in psy- inductive and deductive reasoning problems: 380 A.M. Horton Jr. and C.R. Reynolds

Implications for developing scientifi c reasoning skills. Wechsler, D. (2008). Wechsler adult intelligence scale- Journal of Research in Science Teaching, 32 , 699–714. fourth edition: Technical interpretive manual . San Wechsler, D. (1939). The measurement of adult intelli- Antonio: NCS Pearson. gence . Baltimore: Williams & Wilkins. Weiss, L. G., Keith, T. Z., Zhu, J., & Chen, H. (2013a). Wechsler, D. (1944). The measurement of adult intelli- WAIS-IV and clinical validation of the four- and fi ve gence . Baltimore: Williams & Wilkins. factor interpretive approaches. Journal of Wechsler, D. (1952). The range of human capacities. Psychoeducational Assessment, 31 (2), 94–113. Baltimore: Williams & Wilkins. Weiss, L. G., Keith, T. Z., Zhu, J., & Chen, H. (2013b). Wechsler, D. (2002). Wechsler preschool and primary scale WISC-IV and clinical validation of the four- and fi ve of intelligence-third edition: Technical interpretive factor interpretive approaches. Journal of manual . San Antonio: The Psychological Corporation. Psychoeducational Assessment, 31 (2), 114–131. Wechsler, D. (2003). Wechsler intelligence scale for Willis, W. G. (1985). Successive and simultaneous pro- children-fourth edition: Technical and interpretive cessing: A note on interpretation. Journal of manual . San Antonio: The Psychological Corporation. Psychoeducational Assessment, 4 , 343–346. Current Concepts in the Assessment of Emotional 24 Intelligence

Steven J. Stein and Justin M. Deonarine

All learning has an emotional base. Plato

Emotional intelligence (EI) refers to the ability to “common sense” and ability to get along in the perceive, manage, and evaluate emotions. Some world. researchers suggest that emotional intelligence Ever since the fi rst peer-reviewed publication can be learned and strengthened, while others on emotional intelligence appeared (Salovey and claim it is an inborn characteristic. Mayer 1990 ) and its later popularization in the The concept of emotional intelligence has trade publications (e.g., Goleman 1995 ), there brought new depth to the understanding of human has been controversy over the concept. A number intelligence; it enhances the ability to evaluate of academics have attacked the models presented one’s general or overall intelligence. Unlike for their defi nitional fuzziness, lack of empirical cognitive intelligence, emotional intelligence has data, inconsistencies of the various measures been diffi cult to defi ne. Broadly speaking, emo- developed, and so on. Many of the same criti- tional intelligence addresses the emotional, per- cisms applied to emotional intelligence could sonal, social, and survival dimensions of easily be applied to cognitive intelligence, such intelligence, which are often more important for as the fact that there is still no uniformly accepted daily functioning than the more traditional cogni- single theory after over 100 years of research. tive aspects of intelligence. Emotional intelli- Regardless of their claims, the popularity, use, gence is concerned with understanding oneself and acceptance of the term “emotional intelli- and others, relating to people, and adapting to gence” have now been fi rmly established. and coping with the immediate surroundings in A Google search on “emotional intelligence” order to be more successful in dealing with envi- currently brings up over 25,000,000 entries. ronmental demands. Emotional intelligence is In addition, the emerging fi eld of emotional tactical (immediate functioning), while cognitive intelligence has been attacked for its dispropor- intelligence is strategic (long- term capacity). tionate emphasis on application before valida- Emotional intelligence helps to predict success, tion. This has created several gaps in the fi eld that because it refl ects how a person applies knowl- have been pointed out in a number of critical edge to the immediate situation. In a way, to mea- commentaries (e.g., Landy 2005 ; Zeidner et al. sure emotional intelligence is to measure one’s 2004 ; Murphy 2006 ). Much of the skepticism surrounding the applications of the EI construct comes from issues that are directly related to its * S. J. Stein , Ph.D. ( ) • J. M. Deonarine , M.A. assessment strategies. One issue concerns the Multi-Health Systems , 3770 Victoria Park Ave. , Toronto ON M2H 3M6 , ON , Canada methodological treatment of the construct’s con- e-mail: [email protected]; [email protected] ceptual multidimensionality.

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 381 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_24, © Springer Science+Business Media New York 2015 382 S.J. Stein and J.M. Deonarine

With roots going back to different theoretical rise to the call is one’s way of successfully coping traditions, the nature of EI has been elaborated in with one’s environment—a component of most a number of comprehensive models comprising defi nitions of intelligence. different dimensions and spanning multiple psy- In addition, one’s emotional expression might chological systems, domains, and processes serve as a warning to others, such as to stay away (Bar-On and Parker 2000). This theoretical com- when angry, or as a signal to the herd if an plexity, however, cannot be found in the applied impending danger is sensed. An expression of arena, where the practice of combining various sadness or hurt might signal others to help. One’s EI dimensions into a single composite score and ability to use these emotional signals, either for treating different models and measures of EI as self-motivation, effi ciently alerting others, or mutually interchangeable or otherwise incompat- reading another’s emotions, is identifi ed as an ible has been quite common (Zeidner et al. 2008 ). intelligence as it varies across individuals and, Failure to account for the non-homogeneity of once again, infl uences one’s ability to adapt to complex constructs is not only likely to compli- the environment. cate their true nature and functions but also can However, to gain a practical perspective, we lead to inaccurate decisions and treatment strate- will focus on the development of the concept of gies for individual test takers (McGrath 2005 ; emotional intelligence or one’s emotional quo- Smith et al. 2009 ). Adopting a multidimensional tient (EQ) as developed in the twentieth and approach toward the study and measurement of twenty-fi rst centuries. Back in the 1920s, the EI is clearly needed to remove the ambiguity of American psychologist Edward Thorndike wrote past research and move the EI fi eld forward. about a concept he called “social intelligence.” Another major obstacle to progress in the fi eld He attempted to measure one’s social skills and has been the proliferation and use of EI measures abilities through a paper and pencil test. with inadequate or under researched psychomet- Later, the importance of “emotional factors” ric properties (Zeidner et al. 2008 ). was recognized by David Wechsler, one of the fathers of IQ testing. In a rarely cited paper written in the 1940s, Wechsler urged that the Brief History of Emotional “nonintellective aspects of general intelli- Intelligence gence” be included in any “complete” measurement of intelligence. This paper also discussed Many people, upon fi rst hearing the term emo- what he called “affective” and “conative” abili- tional intelligence, presume that it is an oxymoron. ties—basically, emotional and social intelli- After all, how could one be emotional while at the gence—which he thought would prove critical same time be intelligent? Most likely, emotional to an overall view. Unfortunately, these factors intelligence is as old as time. In the 1870s, Charles were not included in Wechsler’s IQ tests, and Darwin published the fi rst modern book on the little attention was paid to them at the time role of emotional expression in survival and adap- (Wechsler 1940 , 1943 ). tation. The heavily illustrated book presents the In 1948, another American researcher, R. W. thesis that emotions are universal—not only Leeper, promoted the idea of “emotional thought,” among cultures but across species. Additionally, which he believed contributed to “logical emotions serve a function that aided in the survival thought.” But few psychologists or educators pur- of species over time. They served as an early warn- sued this line of questioning until more than 30 ing system, alerting the individual (and possibly years later. (One notable exception was Albert others in fl ocks) that danger was near and that a Ellis, who, in 1955, began to explore what would fi ght or fl ight response was required. Another become known as rational emotive behavior ther- example is the warning of a threat to an offspring apy—a process that involved teaching people to where protection was required. Basically, emo- examine their emotions in a logical, thoughtful tions are a call to action. The ability to successfully way.) In his 1976 book “The Shattered Mind” and 24 Current Concepts in the Assessment of Emotional Intelligence 383 his 1983 book “Frames of Mind: The Theory of 1990 ; Salovey and Rodin 1985 ). In 1990, Mayer Multiple Intelligences,” Howard Gardner wrote and Salovey would coin and formally defi ne the about the possibility of “multiple intelligences,” term “emotional intelligence” as “the subset of including what he called “intra-psychic capaci- social intelligence that involves the ability to ties”—in essence, an aptitude for introspection— monitor one’s own and others’ feelings and emo- and “personal intelligence” Gardner (1976 , 1983 ). tions, to discriminate among them and to use this By this time, Reuven Bar-On was completing information to guide one's thinking and actions.” his dissertation in which he had contributed the Together with David Caruso they expanded upon phrase “emotional quotient” (EQ) as a measure of Gardner’s approach and looked primarily at six emotional intelligence based on his developing components of emotional intelligence that are model and measure of this construct. The evolu- very similar to what Bar-On (1997 ) refers to as tion of the Emotional Quotient Inventory (EQ-i) emotional self-awareness, assertiveness, empa- as an assessment of emotional intelligence began thy, interpersonal relationship, stress tolerance, in 1980 with the independent development of a and impulse control. theoretically eclectic and multifactorial approach The release of Goleman’s (1995 ) “Emotional to operationally defi ning and measuring emo- Intelligence: Why It Can Matter More Than IQ” tional intelligence. The seminal work of Bar-On served to popularize the construct of emotional (1988 ) was inspired by his work as a clinical psy- intelligence, rendering it a hot topic among aca- chologist, with the goal of answering the ques- demics, coaches, consultants, psychologists, and tion, “Why do some people have better the layperson alike, thereby paving the way for— psychological well-being than others?” This and in some cases creating a demand for—a valid question ultimately expanded into, “Why are and reliable measure of EI. However, Goleman’s some individuals better able to succeed in life writing and “theories” of emotional intelligence than others?” These questions commanded a thor- have been heavily criticized by Mayer for going ough review of the factors (emotional skills) far beyond their original work, which he drew thought to determine general success, in addition from (Mayer et al. 2004 ). to maintaining positive emotional health. It soon Finally, the extensive work of Bar-On culmi- became clear that the key to determining and pre- nated in 1997 with the release of the fi rst psycho- dicting success is not cognitive intelligence alone, metrically valid and reliable measure of emotional as many cognitively intelligent people fl ounder in intelligence, the EQ-i (Bar-On 1997 ). Interest in life, while many less cognitively intelligent indi- the EQ-i steadily increased since its release in viduals succeed and prosper. During the 3-year 1997. The energy created and sustained by it is period between 1983 and 1986 while he com- evidenced by the number of publications apply- pleted his doctoral studies in South Africa, Bar-On ing or investigating the EQ-i, including books had directed his efforts at identifying the most (13), trade publications (countless), dissertations important factors involved in coping with envi- (77), and peer-reviewed articles (71). The EQ-i ronmental demands, at which point his research also garnered attention in several peer-reviewed revealed a nonsignifi cant relationship between journal special issues including Emotion , cognitive intelligence and emotional intelligence. Psychological Inquiry , and the Journal of Psychologists continued to challenge the con- Organizational Behavior. The EQ-i has captured ventional view of intelligence. John Mayer at the the attention of researchers and practitioners University of New Hampshire and Peter Salovey worldwide. It can be employed in many ways and at Yale University concentrated their research in a variety of settings. efforts on the “emotional” aspect of intelligence The term emotional intelligence (EI) has sig- (Mayer 1986 ; Mayer et al. 1988 , 1990 , 1991 ; nifi cantly evolved since the fi rst release of the Mayer and Salovey 1988 , 1993 , 1995 ; Mayer and EQ-i, enduring rigorous debate over how to Volanth 1985 ; Salovey and Birnbaum 1989 ; defi ne it, how to measure it, whether it can be Salovey et al. 1991 , 1993 ; Salovey and Mayer developed, and whether or not it adds incremental 384 S.J. Stein and J.M. Deonarine value over and above personality and IQ. This subsections or scales. Based on updated research intense scrutiny of EI helped to refi ne the con- and the latest theories on emotional intelligence, struct; not only has EI endured, its utility is more the MHS team has created an updated model of deeply understood, and its use more widespread. emotional intelligence that is measured by the As a result, the operational defi nition of emo- Emotional Quotient Inventory 2.0 (EQ-i 2.0). It tional intelligence as it relates to the EQ-i 2.0 is now includes 16 subsections or scales as illus- “…a set of emotional and social skills that infl u- trated in Fig. 24.1 . ence the way we perceive and express ourselves, The fi ve realms and their scales include the develop and maintain social relationships, cope following: with challenges, and use emotional information The self-perception realm concerns one’s abil- in an effective and meaningful way.” ity to know and manage oneself. It embraces Here is a brief history of events leading up to emotional self-awareness, the ability to recognize the current concept of emotional intelligence: feelings and why one feels that way and the • 1930s – Edward Thorndike describes the con- impact one’s emotions have on thoughts and cept of “social intelligence” as the ability to actions of oneself and others; self-regard, the get along with other people. ability to recognize strengths and weaknesses • 1940s – David Wechsler suggests that affective and to feel good about oneself despite any weak- components of intelligence may be essential to nesses; and self-actualization, the ability persis- success in life. tently try to improve and pursue meaningful • 1950s – Humanistic psychologists such as goals that lead to a richer life. Abraham Maslow describe how people can The self-expression realm deals with the way build emotional strength. one faces the world. Emotional expression is the • 1976 – Howard Gardner publishes The Shattered ability to express feelings both in words and non- Mind , which introduces the concept of multiple verbally. Independence is the ability to be self- intelligences. This book was followed by directed and self-controlled, to stand on one’s own Frames of Mind: The Theory of Multiple two feet; assertiveness is the ability to clearly Intelligences in 1983. express thoughts and beliefs, stand one’s ground, • 1985 – Reuven Bar-On introduced the term and defend a position in a constructive way. EQ (emotional quotient) in his doctoral dis- The interpersonal realm concerns “people sertation in which he created a test to measure skills”—the ability to interact and get along with one’s emotional quotient. others. It is composed of three scales. Empathy is • 1990 – Psychologists Peter Salovey and John the ability to recognize, understand, and appreci- Mayer publish their landmark article, “Emotional ate what others may be feeling and thinking. It is Intelligence,” in the journal Imagination, the ability to view the world through another per- Cognition, and Personality . son’s eyes. Social responsibility is the ability to • 1995 – The concept of emotional intelligence is be a cooperative and contributing member of popularized after publication of psychologist one’s social group and to society at large. and New York Times science writer Daniel Interpersonal relationships refer to the ability to Goleman’s book Emotional Intelligence: Why forge and maintain relationships that are mutu- It Can Matter More Than IQ . ally benefi cial and marked by give and take and a sense of trust and compassion. The decision-making realm involves the abil- Models of Emotional Intelligence ity to use emotions in the best way to help solve problems and make optimal choices. Its three Emotional Quotient Inventory (2.0) scales are impulse control, the ability to resist or delay a temptation to act rashly; reality testing, Basically, Bar-On arrived at a way to capture the ability to see things as they actually are, rather emotional and social intelligence by dividing it than the way one wishes or fears they might be; into fi ve general areas or realms and 15 and problem solving, the ability to fi nd solutions 24 Current Concepts in the Assessment of Emotional Intelligence 385

Fig. 24.1 Model of emotional intelligence (Reprinted with permission of Multi-Health Systems, Inc., Toronto, Canada. www.mhs.com ) to problems where emotions are involved using rights reserved, Reproduced with permission of the right emotion at an optimum value. Multi-Health Systems, www.mhs.com . The stress management realm concerns the ability to be fl exible, tolerate stress, and control impulses. Its three scales are fl exibility, the abil- The Four Branches of Emotional ity to adjust feelings, thoughts, and actions to Intelligence changing, challenging, or unfamiliar conditions; stress tolerance, the ability to remain calm and Mayer and Salovey proposed a model that identi- focused, to constructively withstand adverse fi ed four different factors of emotional intelli- events and confl icting emotions without caving gence: the perception of emotion, the ability to in; and optimism, the ability to maintain a realis- reason using emotions, the ability to understand tically positive attitude, particularly in the face of emotion, and the ability to manage emotions. adversity. 1 . Perceiving emotions : The fi rst step in under- There is also an independent indicator of hap- standing emotions is to accurately perceive piness. Happiness is the ability to feel satisfi ed them. In many cases, this might involve under- with life, to enjoy oneself and others, and to standing nonverbal signals such as body lan- experience zest and enthusiasm in a range of guage and facial expressions. The ability to activities (Table 24.1 ). detect and decipher emotions in faces, pic- Modifi ed defi nitions from MHS Staff, tures, voices, and cultural artifacts—including Emotional Quotient Inventory 2.0 Manual the ability to identify one’s own emotions—is (2011), Multi-Health Systems: Toronto. All included in this factor. Perceiving emotions 386 S.J. Stein and J.M. Deonarine

Table 24.1 The EQ-i 2.0 scales and what they assess EQ-i 2.0 scales The EI skill assessed by each scale Self-perception Self-regard Ability to respect and accept one’s strengths and weaknesses Self-actualization Ability to improve oneself and pursue meaningful objectives Emotional self-awareness Ability to be aware of and understand one’s feelings and their impact Self-expression Emotional expression Ability to express one’s feeling verbally and nonverbally Assertiveness Ability to express feelings, beliefs, and thoughts in a nondestructive way Independence Ability to be self-directed and free of emotional dependency on others Interpersonal Interpersonal Relationship Ability to develop and maintain mutually satisfying relationships Empathy Ability to recognize, understand, and appreciate the feelings of others Social responsibility Ability to contribute to society, one’s social group, to the welfare of others Decision making Problem solving Ability to solve problems where emotions are involved using emotions Reality testing Ability to remain objective by seeing things as they really are Impulse control Ability to resist or delay an impulse, drive, or temptation to act Stress management Flexibility Ability to adapt one’s feeling, thinking, and behavior to change Stress tolerance Ability to effectively cope with stressful or difﬁ cult situations Optimism Ability to be remain hopeful and resilient, despite setbacks Additional scale Happiness Ability to feel satisﬁ ed with oneself, others, and life in general

represents a basic aspect of emotional intelli- been fi ghting with his wife. The ability to gence, as it makes all other processing of comprehend emotion language and to appre- emotional information possible. ciate complicated relationships among 2 . Facilitating thought: The next step involves emotions is included in this factor. For exam- using emotions to promote thinking and cog- ple, understanding emotions encompasses the nitive activity. Emotions help prioritize what ability to be sensitive to slight variations we pay attention and react to; we respond between emotions and the ability to recognize emotionally to things that garner our atten- and describe how emotions evolve over time. tion. The ability to harness emotions to facili- 4 . Managing emotions: The ability to manage tate various cognitive activities, such as emotions effectively is a key part of emotional thinking and problem solving, is included in intelligence. Regulating emotions, responding this factor. The emotionally intelligent person appropriately, and responding to the emotions can capitalize fully upon his or her changing of others are all important aspects of emo- moods in order to best fi t the task at hand. tional management. The ability to regulate 3 . Understanding emotions : The emotions that emotions in both ourselves and in others is we perceive can carry a wide variety of mean- included in this factor. Therefore, the emo- ings. If someone is expressing angry emotionally intelligent person can harness emotions, the observer must interpret the cause of tions, even negative ones, and manage them to their anger and what it might mean. For exam- achieve intended goals. ple, if your boss is acting angry, it might mean According to Salovey and Mayer, the four that he is dissatisfi ed with your work, or it branches of their model are “arranged from more could be because he got a speeding ticket on basic psychological processes to higher, more psy- his way to work that morning or that he has chologically integrated processes.” For example, 24 Current Concepts in the Assessment of Emotional Intelligence 387 the lowest level branch concerns the (relatively) of job performance ratings. The model borrows simple abilities of perceiving and expressing emo- concepts from the “Basic Models of Emotion” tion. In contrast, understanding and managing paradigm (Gross and Thompson 2007 ), as well as emotions are more complex tasks. from Mayer and Salovey (1997 ). The model is also built on ability-based measures of emotional intelligence (as opposed to self-report measures). Other Theories and Models The authors offer criticisms of self-report measures (to justify using only ability-based mea- Other theories, mostly taken from one or both of sures), claiming that self-report measures lack the leading models of emotional intelligence in scientifi c rigor and treat the term emotional intel- the fi eld, have been proposed. Goleman, the ligence as “…an umbrella term for a broad array New York Times writer, proposed his own theory of constructs that are connected only by their of emotional intelligence, sometimes referred to nonredundancy with cognitive intelligence…” as a “mixed model.” He proposes the following (Joseph and Newman 2010a ). aspects of emotional intelligence: The new theory proposes that emotion percep- 1. Self-awareness – the ability to know one’s tion, emotion understanding, and emotion regula- emotions, strengths, weaknesses, drives, val- tion fi t a progressive structure (where perception ues, and goals and recognize their impact on precedes understanding and understanding leads others while using gut feelings to guide to conscious regulation). The perception of emo- decisions tion includes perceiving the emotions of others 2. Self-regulation – involves controlling or redi- and perceiving self-emotion. These two abilities recting one’s disruptive emotions and impulses may overlap greatly (Joseph and Newman and adapting to changing circumstances 2010b). The understanding of emotion allows 3. Social skill – managing relationships to move mediation between emotion perception and emo- people in the desired direction tion regulation. However, it only applies to emo- 4. Empathy – considering other people's feelings tions which are consciously regulated, not especially when making decisions emotions which are processed automatically 5. Motivation – being driven to achieve for the (such as fear). The model also includes cognitive sake of achievement ability and both conscientiousness and emotional The model introduced by Goleman focuses stability (from the Big Five personality traits) as on emotional intelligence as a wide array of factors which affect performance. Emotion per- competencies and skills that drive leadership ception, understanding, and regulation are all performance. He includes a set of “emotional partial mediators for cognitive ability, conscien- competencies” within each construct of emo- tiousness, and emotional stability. tional intelligence. Emotional competencies are not innate talents, but rather learned capabilities that must be worked on and can be developed to Measuring Emotional Intelligence achieve outstanding performance. Goleman posits that individuals are born with a general emo- Methods of Measuring Emotional tional intelligence that determines their potential Intelligence for learning emotional competencies. Goleman’s model of EI has been criticized in the research Generally, there are three methods used for mea- literature as mere “pop psychology” (Mayer suring emotional intelligence. et al. 2008 ). The fi rst is self-report inventories. Well- In a 2010 meta-analysis, Dana Joseph and constructed self-report inventories compare an Daniel Newman proposed a new theory of emo- individual’s responses to a database of thousands tional intelligence, called the “Cascading Model.” of others, preferably matched for age and gen- It was originally created to address the prediction der, stratifi ed based on census data, and include 388 S.J. Stein and J.M. Deonarine items refl ecting how one sees oneself dealing Questionnaire (TEIQue) (Petrides and Furnham with various situations, interacting with others, 2001 ; Petrides et al. 2007a ) under the broad fac- managing stress, perceiving the future, and using tors of emotionality, self-control, sociability, emotions in various ways—such as dealing with and well-being. However, this model lacks suf- stress and making decisions. While many self- fi cient data for general use. report inventories have appeared following the The third method of assessing emotional intel- popularization of the concept, few have met the ligence is through performance or ability mea- standards of test development as prescribed by sures. These instruments are designed much like the American Psychological Association traditional cognitive intelligence tests. They do (American Psychological Association 1999 ). not rely on one’s opinions, thoughts, or feelings The second is 360° assessments. These add about oneself (or someone else’s thoughts, feel- to the self-report inventory parallel report forms ings, or opinions) but rather on performance on completed by others, such as one’s work super- well-normed tasks with known levels of diffi - visor, subordinates, peers, clients, and even culty. This method is known as an “ability-based spouse. They all report on how they view the approach.” target individual in the same domains (e.g., According to John Mayer, “In regard to mea- emotional self- awareness, empathy, interper- suring emotional intelligence – I am a great sonal skills, etc.). The idea behind 360 reports believer that criterion-report (that is, ability test- is that others are more likely to be candid about ing) is the only adequate method to employ. the target individual’s true emotional skills. On Intelligence is an ability, and is directly measured the other hand, it has been argued that others only by having people answer questions and eval- may not intimately know what really goes on in uating the correctness of those answers.” the target’s mind. In their 2010 meta-analysis, Joseph and These methods are sometimes referred to as Newman (2010) attempted to build their theory “trait-based approaches” or “mixed models.” on ability-based measures of emotional intelli- The trait-based/mixed model approach concep- gence. In their conclusion, the authors ulti- tualizes emotional intelligence as a set of mately concede that self-report models of emotion-related dispositions, attitudes, and self- emotional intelligence are stronger indicators perceptions located at the lower levels of the of performance (such as job performance) than hierarchical personality taxonomy (Petrides and ability-based measures. This admission is sup- Furnham 2001 ; Petrides et al. 2007b ). Like ported by Bachman et al. (2000 ), who used a other personality variables, trait emotional intel- self-report measure (the Bar-On EQ-i) to com- ligence is measured through self-report ques- pare more successful debt collectors to their tionnaires, where respondents are asked to less successful peers. They concluded that spe- report on their typical beliefs, feelings, and cifi c levels of emotional intelligence lead to behaviors. The most infl uential trait-based the- enhanced job performance. However, in a large ory of EI has been Bar-On’s (1997 ) model of meta-analysis of the emotional intelligence lit- emotional–social intelligence, operationalized erature, O’Boyle et al. (2010 ) suggest that each with the Emotional Quotient Inventory (EQ-i; of the three methods are (a) signifi cantly and Bar-On 1997 ; EQ-i 2.0, MHS Staff 2011 ). The positively correlated with job performance as a EQ-i assessed individual differences on a vari- collective measure, (b) signifi cantly and posi- ety of traits and self-concepts organized into tively correlated with job performance individ- four broad EI dimensions: intrapersonal, inter- ually, (c) not signifi cantly different from each personal, adaptability, and stress management. other in their ability to independently predict The revised EQ-i 2.0 is organized into self-per- job performance, (d) correlated positively with ception, self-expression, interpersonal, decision extroversion, openness, agreeableness, consci- making, and stress management. Similar traits entiousness, and cognitive ability, and (e) nega- are measured by the Trait Emotional Intelligence tively related to neuroticism. 24 Current Concepts in the Assessment of Emotional Intelligence 389

Table 24.2 Fifteen factors of the original EQ-i was correlated with self-regard and optimism, Intrapersonal components Interpersonal components while total program effectiveness (a component of Self-regard (SR) Empathy (EM) intellectual capital) was correlated with total EI Emotional self-awareness Social responsibility (RE) score, an intrapersonal composite (consisting of (ES) self-regard, self-actualization, emotional self- Assertiveness (AS) Interpersonal relationship awareness, assertiveness, and independence), as (IR) well as with fl exibility, self- regard, and indepen- Independence (IN) dence. Independence and reality testing were also Self-actualization (SA) Adaptability components Stress management found to be correlated with structural capital. components Independence was also found to be related to total Reality testing (RT) Stress tolerance (ST) fi nancial capital. Flexibility (FL) Impulse control (IC) Enhelder ( 2011 ) also found a positive rela- Problem solving (PS) tionship (via correlation and linear regression) General mood components between total EI score and sales performance of Optimism (OP) fi nancial advisors. They also found a correlation Happiness (HA) between sales performance and the assertiveness, Copyright © 2011 Multi-Health Systems Inc. All rights independence, self-actualization, interpersonal reserved. Based on the original BarOn EQ-i authored by relationships, optimism, and stress tolerance Reuven Bar-On, copyright 1997 measures of the EQ-i. Additionally, in a multiple regression of the previously mentioned subscales, Scales for EI Assessment it was found that self-actualization predicts variability in sales performance. Bar-On’s EQ-i. A self-report test designed to Johnson (2011 ) investigated occupational measure competencies including awareness, well-being in law enforcement. Using the stress tolerance, problem solving, and happiness. EQ-360, Johnson found that differences between According to Bar-On, “Emotional intelligence is managerial self-perceptions and employee per- an array of noncognitive capabilities, competen- ceptions of the manager’s emotional intelligence cies, and skills that infl uence one’s ability to suc- were predictive of the employee’s occupational ceed in coping with environmental demands and well-being. The author further suggests that EI pressures.” Of the 15 factors underlying the EQ-i, training can be used to encourage positive inter- 11 factors had been at the forefront of Bar-On’s action between managers and employees gener- research for nearly two decades. The 15 factors ally, as well as between law enforcement of the original EQ-i (Bar-On 1997 ) and their personnel and the population they serve. Roth abbreviations are in Table 24.2 . (2011 ) examined the connection between level of The EQ-i has been used extensively to success- emotional intelligence in a pastor (EQ-i) and fully assess the effect of emotional intelligence on their ability to draw more attendees to their various measures of performance and quality of life. church (measured by increasing or decreasing Emotional intelligence has often been linked to attendance rates), fi nding that that fi ve competen- employee selection, job satisfaction, fi nancial cies were the best predictors of a pastor’s success performance, and sales performance (Porras and (emotional self-awareness, independence, fl exi- Anderson 1981 ; Spencer and Spencer 1993 ; bility, assertiveness and optimism). Pesuric and Byham 1996 ; McClelland 1999 ; Farnham (2012 ) found a positive relation- McEnrue and Groves 2006 ). Some newer research ship between emotional intelligence and sales applies this knowledge to niche areas. Brown performance in hospice sales professionals, (2011 ) examined the relationship between non- fi nding that a 1-point increase in EQ-i total profi t human service organization intellectual and score leads to 0.316-unit increase in sales. fi nancial capital and CEO emotional intelligence Gender and tenure possessed no explanatory (measured using the EQ-i). Intellectual capital power in this study. Additionally, Mendelson 390 S.J. Stein and J.M. Deonarine

(2012 ) found that elite-level athletes (who are from public schools were found to have a higher currently in the workplace) possess higher lev- rating of EI than those in private schools (despite els of emotional intelligence than their nonath- having an overall lower level of academic perfor- lete coworkers and that number of years of mance) (Malik and Shujja 2013 ). experience is a predicting factor of the athlete’s Emotional intelligence has also been impli- level of emotional intelligence. cated as a factor in leadership abilities. The measure of EI (via the EQ-i) has also been Jones (2012 ) examined the relationship involved in studies of education and academic between emotional intelligence (measured using performance. When comparing a high school stu- the EQ-i) and leadership effectiveness (mea- dent’s EQ-i: Youth Version score to their GPA for sured using the Leadership Practices Inventory) that year, it was found that academic success was in senior level university sponsored administra- strong associated with the total EI score, the tion professionals. Results indicated a signifi cant intrapersonal composite, the interpersonal com- correlation between EI and leadership effective- posite, the adaptability, and the stress manage- ness practices. Additionally, the “Enabling ment (Parker et al. 2004a ). These results are Others to Act” component of the Leadership consistent with the fi ndings of Parker et al. Practices Inventory was correlated with the total (2004b ), who, using the EQ-i Short Form, found EI score, the interpersonal composite score, that these same EI dimensions (except for inter- empathy, social responsibility, impulse control, personal abilities) were predictive of the success- adaptability, problem solving, general mood, ful transition from high school to fi rst year and optimism. university (measured by each student’s academic Osborne (2012 ) found a similar result in phy- performance in the fi rst year). sicians, suggesting that physicians with higher In a study examining emotional intelligence emotional intelligence (EQ-i Short Form) showed (EQ-i), the teacher’s sense of humor (Multi- better leadership practices (Leadership Practices dimensional Sense of Humor Scale), and each Inventory). In regard to the “Enable Others to student’s pretest versus posttest improvement on a Act” component of the Leadership Practices standardized reading test, it was found that both Inventory, Osborne found that only the interper- the total EI and humor score of the teacher were sonal composite was signifi cantly correlated in related to improvements in academic performance, physicians. There were no differences between though EI held a greater explanatory power. Total males and females in these results. EI was also related to the teacher’s humor score (Fernandez 2011 ). MSCEIT. The Mayer-Salovey-Caruso Emotional A study examining the relationship between Intelligence Test is an ability-based scale which EI (using the EQ-i Short Form) and the under- measures how well individuals perform tasks graduate clinical teaching effectiveness in a nurs- (related to emotions) and solve emotional probing faculty (using the Nursing Clinical Teacher lems. The goal of the scale is to gather responses Effectiveness Inventory) found that there is a sig- which represent the ability to solve emotional nifi cant positive relationship between total EI problems. Scores are unaffected by many con- score and total teaching effectiveness score, as founding factors, such as self-concept, response well as between many subscales of the two tools set, and emotional state. (Allen et al. 2012 ). This scale is roughly based on the Multifactor The EQ-i was also used to determine that the Emotional Intelligence Scale (MEIS; Mayer level of a teacher’s emotional intelligence con- et al. 2000). The MEIS included 402 items and tributed to determining their students’ success in four subscales: perception, assimilation, under- a standardized test of mathematics (Shank 2012 ). standing, and managing emotion. The test itself The Urdu EQ-i: Youth Version found a positive was much like an IQ test, including possessing correlation between academic achievement and correct answers. The test also possessed two emotional intelligence. Additionally, children forms of scoring, including consensus scoring 24 Current Concepts in the Assessment of Emotional Intelligence 391

(which assesses the respondent’s choice against internal consistency for ratings from others ranges the majority opinion based on hundreds of from 0.54 to 0.90 (Boyatzis and Sala 2004 ). observations) and expert scoring (which com- While many have used the ECI or ECI-2.0 in pares the respondent’s choice against the major- their work, there have been very few peer- ity opinion of researchers who are experts in reviewed articles published on the reliability and emotion-based research). validity of the study (Cherniss 2000; Spencer The MSCEIT provides 15 scores in total. The 2001 ; Weinburger 2002 ; Conte and Dean 2006 ). total emotional IQ score is an overall index of the The majority of these examinations are not test taker’s overall emotional intelligence. Two favorable towards the ECI. The developers of the “area scores” are also provided, consisting of an ECI suggest that validity evidence from the Self- “experiential emotional intelligence score” (based Assessment Questionnaire acts as the evidence on the respondent’s ability to perceive emotional for the ECI’s validity, though this is met with the information, relate it to other sensations and use it criticism that these tests no longer hold merit, as to facilitate thought) and a “strategic emotional the scale has been revised numerous times. For intelligence score” (based on the respondent’s example, alpha coeffi cients reported for the fi rst ability to understand emotional information and version of the ECI (Boyatzis and Burckle 1999 ) use it for planning and self- management pur- will not apply to the ECI-U (which has a differ- poses). There are four “branch scores,” which ent underlying structure) (Lewis et al. 2005 ). correspond to the elements of the “Four Branches Those who have managed to examine the con- of Emotional Intelligence” theory. Finally, there tent of the ECI have noticed an overlap with four are eight “task scores,” which correspond to each of the Big Five personality dimensions (consci- of the either tasks found in the MSCEIT. entiousness, emotional stability, extraversion, The MSCEIT has a strong reliability rating and openness), as well as with other psychologi- (Mayer, Salovey, Caruso, and Sitarenios, 2003). cal concepts, such as self-awareness and self- Bracket and Mayer ( 2001 ) found a test–retest confi dence (Conte and Dean 2006 ). This has led reliability of r = 0.86 for the full-scale MSCEIT. to the suggestion that “…the ECI does not Split-half estimates of the original test show a deserve serious consideration until peer- reliability of 0.91 for the full-scale test, 0.90 for reviewed empirical studies using this measure the experiential area score, and 0.85 for the stra- are conducted” (Conte and Dean 2006). Since tegic area score. The MSCEIT also possesses this 2006 quote, peer-reviewed papers have been strong validity ratings. Interrater reliability for published, though it provides (at best) modest face validity was r = 0.83. It also possesses strong support towards the validity of the ECI. structural validity (Mayer et al. 2001 ) and dis- Byrne et al. (2007 ) argue that, while the ECI is criminant validity (Daus 2006 ). moderately correlated with factors in the Big Five personality dimensions (r = 0.22–0.57), con- Emotional Competence Inventory (ECI). fi rmatory factor analysis suggests that the factor Based on an older instrument known as the Self- structure of the ECI is different from that of the Assessment Questionnaire, the ECI involves hav- Big Five personality dimensions (which provides ing people who know the individual offer ratings some evidence for discriminant validity for the of that person’s abilities on a number of different ECI self-ratings). They also provide minimal evi- emotional competencies. The test contains 72 dence for convergent validity, by fi nding signifi - items, 4 dimensions (self-awareness, social cant correlations (r = 0.17–0.25) between ECI awareness, self-management, and social skills), self-ratings and the judges’ ratings of emotional and 18 competencies (present only in ECI-2). It competency behaviors displayed by the individ- also features 360° assessment techniques, involv- ual during a “leaderless group discussion.” ing assessments of the individual from their peers However, due to the magnitude of the correla- and supervisor. Internal consistency for self- tions found, they admit that the evidence is weak ratings ranges from 0.45 to 0.77, while the at best. Byrne and colleagues presented small 392 S.J. Stein and J.M. Deonarine correlations (0.11–0.29) between ECI scores and measure) and ability-based measures (such as the measures of work-related ratings (leaderless MSCEIT) (Mayer et al. 2000; Brackett and group discussion peer nominations, age-adjusted Mayer 2003 ). The ECI literature lacks these promotions, and coworker ratings of managerial comparisons. skills), though they admit that ECI self-ratings McEnrue and Groves (2006 ) criticize the lack predicted only one of the three criteria (coworker of validity in the ECI-2. The authors question the ratings of managerial skills) after controlling for content validity of the test, claiming that several personality and age. Finally, they fi nd that there competencies which Goleman identifi ed in his is no relationship between ECI self-ratings and theory seem to be products (not dimensions) of general mental abilities, while other tests hold a emotional intelligence. They also state that the relationship with general mental abilities, such as ECI-2 seems to assess competencies which are MSCEIT (Van Rooy and Viswesvaran 2004 ). not emotional (such as serving as a catalyst for Lewis et al. (2005 ) used a confi rmatory factor change) and does not assess competencies which analysis to validate the ECI-U and determine are related to emotionally intelligent individuals whether four factors (self-awareness, self- (such as knowledge of when to express emo- management, social awareness, and relationship tions). McEnrue and Groves also question the management) were identifi able in their medical construct validity of the ECI-2, stating that the student sample. They chose these four factors as measured competencies are signifi cantly related Boyatzis and Goleman (2001 ) reported that the to many existing personality indexes (thus, the ECI-U to test 21 competencies as part of four test offers little value beyond already existing clusters (consisting of these four factors). tests). They also claim that face validity is also Statistical analyses testing this four-factor model only supported by anecdotal information. The suggested that the fi t of this model to the data was authors criticize the external validity of the ECI- very poor. Additionally, the internal consisten- 2, drawing attention to the lack of normative data cies of three of the four factors (measure by and compiled demographic information of Cronbach’s alpha) did not reach the minimum respondents. Additionally, the ECI-1 and ECI-2 threshold for reliability. Also, there were large are very different from each other, thus support- correlations between factors (r = 0.999 between ing evidence for the external validity of the ECI-1 self-awareness and self-management, and cannot apply to the ECI-2. r = 0.973 between social awareness and relation- Finally, unlike the EQ-i, the ECI lacks fi ndings ship management), suggesting that the ECI-U’s which allow the prediction of performance and clusters were not entirely distinct from each satisfaction in various areas of life. The ECI was other. The authors of this paper conclude that “… found to have signifi cant overlap in most of its what this exercise has shown us is that we cannot dimensions and contributed only a small amount be certain that we are measuring what this scale of variance above the NEO-PI in predicting orga- purports to measure i.e. EI” (p. 347). nizational performance (Murensky 2000 ). Results Matthews et al. (2003 ) noted that ECI proof the ECI were compared before and after train- vides some utility, as it assesses many dissimilar ing school teachers with a program designed to concepts simultaneously. However, they addi- increase awareness and understanding of cogni- tionally state that there are more advanced tech- tive messages which create feelings and behav- niques for assessing many of the factors found in iors. There were no signifi cant differences the ECI. Leslie and Fleenor (1998 ) also note that pre- and posttest, for both the experimental and other 360° rating instruments exist, which mea- control group (Walker 2001 ). In a study involving sure similar skills to the ECI. Proponents of the emotional intelligence and an individual’s toler- ECI have yet to show that the ECI 360° ratings ance levels towards workplace bullying, it was are different from other well-known 360° rating found that none of the four competency clusters instruments. Finally, there have been compari- nor the total score from the ECI was related sons between the Bar-On EQ-i (a self-report to an individual’s tolerance (Roundy 2007 ). 24 Current Concepts in the Assessment of Emotional Intelligence 393

In a study examining the connection between an construct measured by each of the scales and ESCI score and leadership style (via the subscales. Test–retest reliability was calculated Multifactor Leadership Questionnaire), no signif- ( n = 203), producing correlations for total and icant relationship was found. However, it was subscale scores ranging from r = 0.76 to 0.89. found that those who score higher on the ESCI Test–retest stability values were also calculated, believe that they practice a form of leadership with the results suggesting that 90 % or more of closer to a transformational leadership style (as the individual’s scores (for both the total score compared to the alternatives of transactional lead- and each subscale score) did not change by more ership and passive avoidance) (Woods 2010). than one standard deviation between the two test- Self-awareness factors of the ECI-2.0 do not pre- ing periods. dict the scores on a Burnout scale (Maslach The EQ 360 2.0 also possesses high validity Burnout Inventory—GS) in chefs, while the con- ratings, as determined by multiple tests. The fi rst scientiousness factors from the NEO were able to set of tests examined the correlations between predict the Burnout scale score (Hintertoisser the composite scales and the subscales within 2011 ). Finally, only the infl uence competency the normative sample, in an attempt to determine component of the ECI is signifi cantly related to if the pattern of results found in the EQ-i 2.0 general job satisfaction (Minnesota Satisfaction would be replicated. The resulting correlations Questionnaire), leading the authors to conclude ranged from r = 0.64 (self-expression/interper- that there is a lack of suffi cient evidence to con- sonal) to 0.86 (decision making/stress manage- clude a meaningful relationship between emo- ment). In most cases, the correlations were tional intelligence competencies (as tested by the stronger than the corresponding correlation for ECI) and general job satisfaction (Agbolou 2011 ). the EQ-i 2.0 normative sample. Additionally, subscales within the same composite scale pos- EQ 360 2.0. The EQ 360 2.0 is a 360° assess- sessed strong correlations. Every value exceeded ment (also known as a multirater or multisource a medium effect size, and most exceeded a large feedback assessment), based on the EQ-i 2.0, effect size. Correlations ranged from r = 0.37 which provides the individual with feedback (emotional express/independence) to 0.81 from their peers and supervisors, allowing them a (empathy/interpersonal relationships). These deeper understanding of their performance from results suggest that the composite scales and multiple perspectives. This method of assessment subscales share a relevant underling factor. Also, provides the means for comparing internal per- these results are similar to the results of this ceptions with external perceptions and acts as a analysis for the EQ-i 2.0. means for enhanced self-awareness and subse- The second set of tests examined the associa- quent behavioral change. tions between 108 pairs of self-ratings and the rat- The normative group consists of 3,200 ratees ings of their peer (often family members, spouses, (59.2 % female). Ratees are demographically or friends who interacted with the participants similar to the 2008 US and 2006 Canadian census frequently and knew the participant very well). results. Data from both countries are included in The correlation between the total scores was the normative sample, as there were no statisti- r = 0.60 (p < 0.01) with correlations ranging from cally signifi cant differences between the ratings r = 0.44 (stress tolerance) to 0.72 (happiness). provided by the US and Canadian participants. Correlations for the composite scales and sub- The EQ 360 2.0 possesses high reliability rat- scales were all signifi cant at p < 0.01. Almost ings. In regard to the internal consistency of the every correlation reached the criterion for a large normative sample, the Cronbach’s alpha is at effect size. The observed pattern suggests that least 0.85 for all but two subscales. The excep- there is strong agreement between self-report and tions are emotional expression ( α = 0.82) and the report of others and that EI (as measured by assertiveness (α = 0.79). This suggests that the the EQ-i 2.0 and the EQ 360 2.0) is a robust trait items cohesively measure total EI, as well as the that is evaluated similarly via both self-report and 394 S.J. Stein and J.M. Deonarine report from external observers. However, these score for either the rater or the ratee. Analysis of correlations are not high enough to suggest redun- covariance techniques were used to examine dancy between the two forms of report (especially potential effects on the EQ 360 2.0 total score, as self-reports will not always align with external using rater and ratee ethnicity as independent observer ratings). Each measure provides unique variables, as well as ratee gender and age group and important information about the individual. as covariates. Multivariate analyses of covari- In order to supplement the above correla- ance were also used to examine the effects of the tional results, the standard scores of the EQ-i above independent variables and covariates on 2.0 and the EQ 360 2.0 were compared. By sub- the composite scales and subscales of the EQ tracting the EQ 360 2.0 standard score from the 360 2.0. None of the analyses produced signifi - EQ-i 2.0 score, a difference score was deter- cant results, suggesting that the raters did not mined. It was previously established that a dif- show differences in their ratings based on the ference of 10 points would result in a meaningful ethnicity of the ratee. Additionally, the Wilks’ difference between the two standard scores Lambda values suggested that a negligible (Multi-Health Systems 2011 , see EQ-i 2.0 amount of variance could be explained by the User’s Handbook, Chapter 6 ). Over half of the interaction between rater and ratee ethnicity, difference scores (for total EI score, composite while none of the effect sizes met the require- scales, and subscales) had a value of less than ments to be classifi ed as a small effect size. 10, suggesting a good degree of consistency between the EQ-i 2.0 and EQ 360 2.0 scores. Trait Meta-Mood Scale (TMMS). The Trait However, given that almost half of the differ- Meta-Mood Scale (Salovey et al. 1995 ) was the ence scores were meaningful, both scores are very fi rst self-report measure to be applied in still important to collect in order to build a com- EI-related research. The 30-item scale is based plete assessment of the individual. on the earlier work on the cognitions accompany- The third set of tests examined the relation- ing subjective mood experiences (Mayer and ship between emotional intelligence and general/ Gaschke 1988 ) and measures three core meta- social adjustment. Social adjustment was mea- mood processes: attention (attending to and mon- sured by the Social Adjustment Scale – Self- itoring changes in one’s mood states), clarity Report (SAS-R from Weissman 1999 ) and (discerning the nature and causes of one’s feel- compared to the EQ-i 2.0 and EQ 360 2.0. ings), and repair (regulating one’s emotions in Correlations between the SAS-R and the EQ-i adaptive ways). The TMMS possesses sound reli- 2.0/EQ 360 2.0 were mostly strong. Stepwise ability (Cronbach’s α = 0.70) and good validity, multiple regression revealed that the SAS-R total supported by a sizeable body of evidence sup- score is independently related to the total EI porting the three-factor structure. It also pos- score for both the EQ-i 2.0 and the EQ 360 2.0, as sesses strong predictive utility for a range of well as most of the composite scales and sub- performance and health-related outcomes (e.g., scales (p < 0.05). In other words, the ratings in the Gignac et al. 2003 ; Salovey et al. 2002 ; Thompson self-report data and the external observer data are et al. 2007). Its main shortcoming, however, is both uniquely informative about the individual’s that the TMMS was never designed to measure SAS-R total score. Additionally, by examining the EI construct specifi cally (Salovey et al. 1995 ). the R2 change values in the stepwise regression, As a result, several components of major EI mod- it can be observed that each of the subscales and els (such as empathy, understanding emotions of composite scales from the EQ-i 2.0 and the EQ others, interpersonal skills) are overlooked, 360 2.0 provided unique and incremental contri- which limits the scope of its utility in theoretical butions towards social adjustment. and applied research (Parker et al. 2011 ). Finally, the potential of an ethnicity bias effect in the rater–ratee relationship was exam- Assessing Emotions Scale (AES). This is ined. Ideally, ethnicity should not affect the another brief self-report measure that has been 24 Current Concepts in the Assessment of Emotional Intelligence 395 widely used in EI-related research (Schutte The four-factor structure of the EQ-i:S scores et al. 1998 ). The AES is based on Salovey and has been confi rmed in a large sample of adults Mayer’s (1990 ) original defi nition of EI as com- (Bar-On 2002 ), with each subscale possessing prising four ability domains (appraising emo- internal consistency (Cronbach’s α = 0.70) and tions in self and others, expressing emotions, congruence with scores on the corresponding managing emotions of self and others, and uti- long-form scale ( r = 0.73–0.93). Parker et al. lizing emotions in problem solving). This scale (2011 ) provided evidence that EQ-i:S scores are has become the choice for many applied EI more strongly associated with measures of con- researchers, due to its solid theoretical founda- ceptually similar constructs (such as the MSCEIT tion and the concise nature of the scale (Schutte and the 20-item Toronto Alexithymia Scale) than et al. 2009 ). However, the AES has several psy- they are with more general dimensions of per- chometric problems, including an unreliable sonality. Additionally, at the latent-variable measurement structure. In some factor analytic level, the EQ-i:S also shares more conceptual investigations (Brackett and Mayer 2003 ; Schutte overlap with the MSCEIT (65 % shared vari- et al. 1998 ), AES responses were found to be rep- ance) than alexithymia (41 % shared variance). resented adequately by a single common factor, They concluded that “…the EQ-i:S produced whereas in others (Keele and Bell 2008 ; Petrides internally consistent, temporally reliable and and Furnham 2000 ; Saklofske et al. 2003 ) four theoretically meaningful responses that also fol- separate factors were identifi ed, although their lowed a stable, gender-invariant, multidimen- item composition varied from study to study and sional measurement structure” (p. 773). On the did not fully correspond to the domains in other hand, the TEIQue–SF shows low levels of Salovey and Mayer’s conceptual model (Parker internal consistency at the individual factor level et al. 2011 ). Given these issues, most applications (Cronbach’s α = 0.70). The developers of the use only the total score, as this measure is a reli- scale have recommend the use of the total scale able (Cronbach’s α = 0.80) but nonspecifi c index score (Cronbach’s α = 0.80) instead of the indi- of global EI (Schutte et al. 2009 ). vidual factors (Petrides 2009 ; Petrides and Furnham 2006 ). Like the AES, the TEIQue–SF EQ-i and TEIQue–Short Forms. The 35-item is useful only as a general measure of global EI, Emotional Quotient Inventory–Short Form while the EQ-i:S offers a unique advantage over (EQ-i:S; Bar-On 2002 ) and the 30-item Trait other brief EI scales and can be used for a wider Emotional Intelligence Questionnaire–Short range of research questions and applications Form (TEIQue–SF; Petrides and Furnham 2006 ) (Parker et al. 2011 ). are short-form alternatives to their respective full length assessments. Neither of the two short forms has been researched as extensively as the Neurological Correlates TMMS or the AES, their well-known theoretical of Emotional Intelligence bases make them worthwhile measures to consider. Both short forms were designed to match A wealth of information has been generated over the higher-order measurement structure of their recent years concerning the emotional implica- respective parent scales, with items selected for tions for performance success and the role emo- inclusion on the basis of item-total correlations tions play in survival and everyday decision and representative coverage of the facet level making. Recent technological advances in the content (Parker et al. 2011 ). The EQ-i:S yields a study of emotion (e.g., fMRI, EEG, etc.) have total EI score and four subscale scores (intraper- shed light on the cortical and subcortical struc- sonal, interpersonal, adaptability, and stress man- tures of the brain linked to the emotion network agement), and the TEIQue–SF provides a total that drives how we think, feel, and act. Of par- trait EI score and four-factor scores ( emotionality, ticular interest are the subcortical structures of self-control, sociability, and well-being). the thalamus, cingulate cortex, amygdala, and the 396 S.J. Stein and J.M. Deonarine cortical regions, such as the temporal lobe and lower EI despite showing no difference in the prefrontal cortex. In the event one of these IQ. Given the implications of vmPFC in the areas should become impaired, so too does the somatic marker hypothesis for decision making individual’s ability to process or use information and subsequent behavior, Killgore and Yurgelun- effectively, often affecting thoughts, feelings, and Todd (2007 ) set out to assess the link between actions. This structure–function relationship vmPFC activation (measured using fMRI) and maintains implications for stress tolerance, well- levels of emotional intelligence. Similar to the being, decision making, and success. Recently, results of Bar-On et al. (2003 ), Killgore and some researchers have attempted to link EI to the Yurgelun-Todd reported that adolescents with very neural mechanisms that have been demon- relatively low emotional intelligence respond to strated to have a direct impact on what we per- emotionally provocative pictures with greater ceive, how we react, the decisions we make, and and more extensive brain activation than do those ultimately the quality of life we lead. with well-developed emotional intelligence. In According to the somatic marker hypothesis other words, emotional intelligence can moderate (Damasio et al. 1991 ), when we make a decision, the impact of stressful stimuli, allowing the brain we fi rst weigh the pros and cons or the benefi ts/ to operate more effi ciently under stressful condi- consequences of the response options, a process tions (Haier et al. 1992 ). that requires both emotional and cognitive pro- Damage to temporal lobe functioning has cessing. However, when faced with a simple been reliably linked to increased agitation, diffi - choice, we often resort to cognitive rules to assist culty managing emotions, heightened irritability, in the decision process; as the decision becomes and, more recently, impaired social cognition more complex or we place greater value on the (Walpole et al. 2008 ). If damage or functional outcome, our cognitive rules may not be suffi - impairment of the temporal lobe presents emo- cient to render a decision and we get stuck. Enter tional challenges, then perhaps the temporal emotions and the somatic marker hypothesis. region of the brain is linked to EI. Walpole et al. Somatic markers are simply connections ( 2008 ) in a controlled experiment measured the between a stimulus (i.e., choice) and a resulting emotional intelligence of patients with temporal physiological sensation we experience when pre- lobe epilepsy and a healthy cohort matched for sented with such a stimulus. When presented age and IQ. These fi ndings suggest that impair- with a given stimulus, we experience certain sen- ment to the medial temporal lobe is related to sations which in turn bias or infl uence our deci- lower emotional intelligence and impaired facial sions. In most cases, the somatic marker directs recognition and to greater psychological distress attention to the most meaningful information to as compared to healthy individuals. enhance decision making (Damasio et al. 1991 ); To this point, emotional intelligence has been however, in the event a defi cit in emotional pro- linked to vmPFC and the temporal lobe by means cessing is evident, decision making and judgment of comparing relatively healthy individuals to become impaired. those with a structural defi cit. In a series of stud- Bar-On et al. (2003 ) set out to determine ies conducted by Killgore and colleagues (Kahn- whether individuals with impaired somatic mark- Greene et al. 2006 ; Killgore et al. 2007 , 2008 ), ers (i.e., lesions to the ventromedial prefrontal healthy participants were used in a repeated mea- cortex; vmPFC) would refl ect that impairment sures design to isolate the simulated effects of through abnormal emotional intelligence. In this vmPFC impairment on emotion functioning. case, damage to the vmPFC is often associated Killgore et al. (2008 ) were able to demonstrate with impaired judgment and decision making the cortical connection of emotional intelligence and, in turn, should be refl ected by lower scores to the prefrontal cortex via sleep deprivation. It in EI. Comparing patients with lesions to various has been shown that sleep deprivation can result brain regions, Bar-On et al. (2003 ) demonstrated in temporary impairment of the prefrontal cortex, that those with lesions to the vmPFC reported resulting in diffi culties regulating higher-order 24 Current Concepts in the Assessment of Emotional Intelligence 397 executive functions such as impulse control, inhi- anterior insula allows the integration of bition of aggression, willingness to act in a cognition and emotion (Gu et al. 2012 ), and the socially acceptable way (Kahn-Greene et al. vmPFC is associated with understanding and 2006 ), and moral judgment (Killgore et al. 2007 ). managing emotional information (via the When comparing sleep-deprived EQ-i results to MSCEIT strategic emotional intelligence corre- baseline results, sleep-deprived participants lation), as well as emotional control (via the reported decreased total EQ, intrapersonal, inter- EQ-i stress management correlation). personal, and stress management composite Aron Barbey and his colleagues (2012) inves- scores (Killgore et al. 2008), and those scoring tigated the idea that the neural architecture of lower at baseline were more susceptible to decre- emotional and social intelligence overlaps that of ments in moral judgment performance (Killgore cognitive intelligence. Their sample consisted of et al. 2007 ). 152 males who experienced penetrating head More recently, regional brain volumes have injuries (leading to brain lesions in a variety of been linked to emotional intelligence. Killgore locations) during the Vietnam War. Each partici- and colleagues (2012 ) examined the correlation pant was administered the MSCEIT as a measure between gray matter volume in the somatic of emotional intelligence, the WAIS-III (Wechsler marker circuitry (Bar-On et al. 2003 ) and the 1997 ) as a measure of cognitive intelligence, and ability-based model of emotional intelligence the NEO-PI-R (Costa and McCrae 1997 ) as a (via the MSCEIT), as well as a self-report mea- measure of personality traits. sure of emotional intelligence (the Bar-On Emotional intelligence defi cits (lower scores EQ-i). Their results suggest that the complete on the MSCEIT) were associated with damage to ability- based model of emotional intelligence is the social cognitive network (review in Saxe positively correlated with the volume of gray 2006 ), suggesting that social and emotional pro- matter in the left posterior insula, while the cesses are integrated at the neural level. Specifi c complete self-report model was not correlated areas of damage included the left posterior tem- with gray matter volume within the somatic poral cortex (associated with recognition of the market circuitry. Additionally, individual com- form of human bodies), the left posterior superior ponents of each model correlated with gray mat- temporal sulcus (associated with interpreting the ter volume. The strategic emotional intelligence motions of the human body in relation to the indi- subscale of the MSCEIT was correlated with the vidual’s goals), the left temporoparietal junction gray matter volume of the bilateral medial pre- (associated with the ability to reason about the frontal cortex, the left posterior and anterior contents of mental states), and the left orbitofron- insula, and the ventrolateral prefrontal cortex tal cortex (associated with the support of emo- area (including the inferior frontal gyrus). The tional empathy and triadic relations between two stress management subscale of the EQ-i was minds and an object, as well as supporting shared positively correlated with the gray matter vol- attention and collaborative goals). Additionally, ume of both the right and left vmPFC. It is inter- there were damaged white matter fi ber tracts esting to note that the volume of the amygdala is which are associated with the social cognitive not related to either the ability or self-report network, including the super longitudinal/arcuate models of emotional intelligence, despite its fasciculus (which connects the temporal lobe, importance within the somatic marker circuitry. parietal lobe, and inferior area of the frontal However, the authors also note that the posterior lobe), the superior fronto-occipital fasciculus insula, the anterior insula, and the vmPFC are (which connects the dorsolateral prefrontal cor- also important to the somatic market circuitry tex and the frontal area of the superior parietal and are positively correlated with the tasks used cortex), and the uncinate fasciculus (which con- in this study. Specifi cally, the posterior insula nects the anterior temporal cortex and the amyg- infl uences somatic processing (Craig 2003 ) and dala with the orbitofrontal and frontopolar emotional processing (Xue et al. 2010 ), the temporal cortex). 398 S.J. Stein and J.M. Deonarine

Emotional intelligence defi cits were also (including the superior longitudinal/arcuate fas- associated with impaired behavioral perfor- ciculus, the superior fronto-occipital fasciculus, mance in the verbal comprehension component and the uncinate fasciculus). of the WAIS-III, suggesting that emotional intel- Overall, Barbey and colleagues found that the ligence shares neural systems which are essen- network involved in emotional intelligence was tial for cognitive or crystallized intelligence. distributed in nature, though the core compo- Selective damage was observed in the left hemi- nents consisted in either white matter areas of sphere perisylvian language network (review in the brain (suggesting the importance of commu- Hickok and Poeppel 2007). Impaired perfor- nication between the employed brain regions for mance was associated with both the ventral and greater emotional intelligence) or regions associ- the dorsal pathways of this network. The ventral ated with social information processing (sug- pathway (consisting of the anterior and posterior gesting the importance in the regulation and middle temporal gyrus, as well as the middle control of social behavior in emotional intelli- posterior superior temporal sulcus) is known for gence). They also suggest that the orbitofrontal its role in language comprehension, including cortex is a central component of the network for the support of mapping sensory or phonological emotional intelligence, especially as it has been representations onto lexical or conceptual repre- suggested as an important region for emotional sentations. The dorsal pathway (consisting of the and social cognition (review in Kringelbach anterior and posterior insula, as well as a section 2005 ). Researchers have proposed that the along the parietotemporal boundary) is known medial orbitofrontal cortex is crucial for emo- for its role in language production, including the tional intelligence processes, as it is involved in support of converting sensory or phonological monitoring emotional properties of social and representations into motor articulations. environmental stimuli. Emotional intelligence defi cits were also The results of recent neurological investiga- found in relation to defi cits in processing speed tions show promise for enhancing our under- (via the WAIS-III). Defi cits in both processing standing of the role emotional intelligence plays speed and emotional intelligence were related to in our capacity to meet our daily demands. damage to the left dorsolateral prefrontal cortex Emotional intelligence appears to have a broad (which is involved in the regulation and control neurological representation which, in turn, serves of both emotional and social behavior). to moderate the effects of emotional stimuli Defi cits in emotional intelligence were found affecting both our decisions and behaviors. to be linked to only the conscientiousness personality trait of the Big Five personality traits. Brain regions associated with the impairments are References involved in the regulation and control of behavior and are also implicated in social information pro- Agbolou, K. (2011). Relationship between emotional cessing networks (review in Saxe 2006 ). These intelligence and job satisfaction: A correlational analysis of a retail organization (Doctoral Dissertation). regions include the right dorsolateral prefrontal Available from ProQuest Dissertations & Theses: Full cortex, the left orbitofrontal cortex, and the left Text (3471686). temporoparietal junction. Allen, D. E., Ploeg, J., & Kaasalainen, S. (2012). The rela- Finally, when residual scores (which removed tionship between emotional intelligence and clinical teaching effectiveness in nursing faculty. Journal of variance shared with signifi cant cognitive and Professional Nursing, 28 (4), 231–240. personality predictors) were analyzed, additional American Psychological Association. (1999). Standards brain structures were shown to be involved in for educational and psychological testing. Developed emotional intelligence. These areas showing jointly by the American Educational Research Association (AERA), American Psychological additional activation include the right orbitofron- Association (APA), and National Council on tal cortex, the left inferior and superior parietal Measurement in Education (NCME), American cortex, and the major white matter fi ber tracts Psychological Association, Washington, DC. 24 Current Concepts in the Assessment of Emotional Intelligence 399

Bachman, J., Stein, S., Campbell, K., & Sitarenios, G. Costa, P. T., Jr., & McCrae, R. R. (1997). Stability and (2000). Emotional intelligence in the collection of change in personality assessment: The revised NEO debt. International Journal of Selection and personality inventory in the year 2000. Journal of Assessment, 8 (3), 176–182. Personality Assessment, 68 , 86–94. Barbey, A. K., Colom, R., & Grafman, J. (2012). Craig, A. D. (2003). Interoception: The sense of the physi- Distributed neural system for emotional intelligence ological condition of the body. Current Opinion in revealed by lesion mapping. Social Cognitive and Neurobiology, 13 (4), 500–505. Affective Neuroscience, 9, 265–272 Damasio, A. R., Tranel, D., & Damasio, H. (1991). Bar-On, R. (1988). The development of a concept of psy- Somatic markers and the guidance of behaviour: chological well-being. Unpublished doctoral disserta- Theory and preliminary testing. In H. S. Levin, H. M. tion, Rhodes University, South Africa. Eisenberg, & A. L. Benton (Eds.), Frontal lobe func- Bar-On, R. (1997). The Bar-On Emotional Quotient tion and dysfunction (pp. 217–229). New York: Oxford Inventory (EQ-i): Technical manual. Toronto: Multi- University Press. Health Systems. Daus, C. S. (2006). The case for an ability-based model of Bar-On, R. (2002). Bar-On Emotional Quotient Inventory: emotional intelligence. In K. R. Murphy (Ed.), A cri- Short (EQ-i:S): Technical manual . Toronto: Multi- tique of emotional intelligence: What are the problems Health Systems. and how can they be fi xed? (pp. 301–324). Mahwah: Bar-On, R., & Parker, J. D. A. (2000). The Bar-On Lawrence Erlbaum Associates. Emotional Quotient Inventory: Youth Version (EQ-i:YV) Enhelder, M. (2011). Emotional intelligence and its rela- Technical manual. Toronto: Multi-Health Systems. tionship to fi nancial advisor sales performance Bar-On, R., Tranel, D., Denburg, N. L., & Bechara, A. (Doctoral Dissertation). Available from ProQuest (2003). Exploring the neurological substrate of emo- Dissertations & Theses: Full Text (3465842). tional and social intelligence. Brain, 126 , 1790–1800. Farnham, B. (2012). Exploring impacts of emotional Boyatzis, R. E., & Burckle, M. (1999). Psychometric intelligence, gender and tenure on sales performance properties of the ECI . Boston: Hay/McBer Group. among hospice sales professionals (Doctoral Boyatzis, R. E., & Goleman, D. (2001). Emotional Dissertation). Available from ProQuest Dissertations competence inventory – University version. Self- & Theses: Full Text (3501884). assessment questionnaire, profi le and interpretive Fernandez, S. (2011). The relationship between teachers’ notes. London: Hay Resources Direct. emotional intelligence and sense of humor, and stu- Boyatzis, R. E., & Sala, F. (2004). Assessing emotional intel- dent achievement (Doctoral Dissertation). Available ligence competencies. In G. Geher (Ed.), The measure- from ProQuest Dissertations & Theses: Full Text ment of emotional intelligence: Common ground and (3467199). controversy (pp. 147–180). Hauppauge: Nova Science. Gardner, H. (1976). The shattered mind. London: Routledge Brackett, M., & Mayer, J. D. (2001). Comparing measures & Kegan Paul. of emotional intelligence. Paper presented at the Third Gardner, H. (1983). Frames of mind: The theory of multi- Positive Psychology Summit, Washington, DC. ple intelligences . New York: Basic Books. Brackett, M. A., & Mayer, J. D. (2003). Convergent, dis- Gignac, G. E., Palmer, B., Bates, T., & Stough, C. (2003). criminant, and incremental validity of competing mea- Differences in confi rmatory factor analysis model sures of emotional intelligence. Personality and Social close-fi t index estimates obtained from AMOS 4.0 and Psychology Bulletin, 29 , 1147–1158. AMOS 5.0 via full information maximum likelihood– Brown, J. S. (2011). Emotional intelligence exploration: no imputation: Corrections and extension to Palmer A potential relationship between CEO EI and non- et al. (2003). Australian Journal of Psychology, 58 (3), profi t intelligence and fi nancial capital (Doctoral 144–150. Dissertation). Available from ProQuest Dissertations Goleman, D. (1995). Emotional intelligence . New York: & Theses: Full Text (3468350). Bantam Books. Byrne, J. C., Dominick, P. G., Smither, J. W., & Reilly, Gross, J. J., & Thompson, R. A. (2007). Emotion regula- R. R. (2007). Examination of the discriminant, conver- tion: Conceptual foundations. In J. J. Gross (Ed.), gent and criterion-related validity of self-ratings on the Handbook of emotion regulation . New York: Guilford emotional competence inventory. International Press. Journal of Selection and Assessment, 15 (3), 341–353. Gu, X., Liu, X., Van Dam, N. T., Hof, P. R., & Fan, J. Cherniss, C. (2000, April 15). Emotional intelligence: (2012). Cognition-emotion integration in the anterior What it is and why it matters. Paper presented at the insular cortex. Cerebral Cortex , 23(1), 20–27. Annual Meeting of the Society for Industrial and Haier, R. J., Siegel, B., Tang, C., Abel, L., & Buchsbaum, Organizational Psychology, New Orleans. M. S. (1992). Intelligence and changes in regional Conte, J. M., & Dean, M. A. (2006). Can emotional intel- cerebral glucose metabolic rate following learning. ligence be measured? In K. R. Murphy (Ed.), A critique Intelligence, 16 , 415–426. of emotional intelligence: What are the problems and Hickok, G., & Poeppel, D. (2007). The cortical organiza- how can they be fi xed? (pp. 59–77). Mahwah: Lawrence tion of speech processing. Nature Reviews Neuroscience, Erlbaum Associates. 8 , 393–402. 400 S.J. Stein and J.M. Deonarine

Hinterstoisser, T. M. (2011). An investigation into the ments for management development . Greensboro: relationship between conscientiousness, self- Center for Creative Leadership. awareness and occupational stress outcomes in culi- Lewis, N. J., Rees, C. E., Hudson, J. N., & Bleakley, A. nary chefs (Doctoral Dissertation). Available from (2005). Emotional intelligence in medical education: ProQuest Dissertations & Theses: Full Text (3478296). Measuring the unmeasurable? Advances in Health Johnson, A. M. (2011). The impact of managerial emo- Sciences Education, 10 , 339–355. tional intelligence perceptions on the occupational Malik, F., & Shujja, S. (2013). Emotional intelligence and well-being of employees in a police department academic achievement: Implications for children’s (Doctoral Dissertation). Available from ProQuest performance in schools. Journal of the Indian Dissertations & Theses: Full Text (3482199). Academy of Applied Psychology, 39 (1), 51–59. Jones, V. D. (2012). The relationship between emotional Matthews, G., Zeidner, M., & Roberts, R. (2003). intelligence and leadership effectiveness among spon- Emotional intelligence: Science and myth . Cambridge, sored research administrators (Doctoral Dissertation). MA: The MIT Press. Available from ProQuest Dissertations & Theses: Full Mayer, J. D. (1986). How mood infl uences cognition. In Text (3514683). N. E. Sharkey (Ed.), Advances in cognitive science Joseph, D. L., & Newman, D. A. (2010a). Emotional intel- (pp. 290–314). Chichester: Ellis Horwood Limited. ligence: An integrative meta-analysis and cascading Mayer, J. D., & Gaschke, Y. N. (1988). The experience model. Journal of Applied Psychology, 95 (1), 54–78. and meta-experience of mood. Journal of Personality Joseph, D. L., & Newman, D. A. (2010b). Discriminant and Social Psychology, 55 , 102–111. validity of self-reported emotional intelligence: A Mayer, J. D., & Salovey, P. (1988). Personality moderates multitrait–multisource study. Educational and the interaction of mood and cognition. In K. Fiedler & Psychological Measurement, 70 (4), 672–694. J. Forgas (Eds.), Affect, cognition and social behavior Kahn-Greene, E. T., Lipizzi, E. L., Conrad, A. K., (pp. 87–99). Toronto: Hogrefe. Kamimori, G. H., & Killgore, W. D. (2006). Sleep Mayer, J. D., & Salovey, P. (1993). The intelligence of deprivation adversely affects interpersonal responses emotional intelligence. Intelligence, 17 (4), 433–442. to frustration. Personality and Individual Differences, Mayer, J. D., & Salovey, P. (1995). Emotional intelligence 41 (8), 1433–1443. and the construction and regulation of feelings. Keele, S. M., & Bell, R. C. (2008). The factorial validity Applied and Preventive Psychology, 4 , 197–208. of emotional intelligence: An unresolved issue. Mayer, J. D., & Salovey, P. (1997). What is emotional Personality and Individual Differences, 44 , 487–500. intelligence? In P. Salovey & D. Sluyter (Eds.), Killgore, W. D., & Yurgelun-Todd, D. A. (2007). Neural Emotional development and emotional intelligence: correlates of emotional intelligence in adolescent chil- Implications for educators . New York: Basic Books. dren. Cognitive, Affective, & Behavioral Neuroscience, Mayer, J. D., & Volanth, A. J. (1985). Cognitive involve- 7 (2), 140–151. ment in the mood response system. Motivation and Killgore, W. D., Killgore, D. B., Day, L. M., Li, C., Emotion, 9 , 261–275. Kamimori, G. H., & Balkin, T. J. (2007). The effects Mayer, J. D., Mamberg, M. H., & Volanth, A. J. (1988). of 53 hours of sleep deprivation on moral judgment. Cognitive domains of the mood system. Journal of Sleep, 30 (3), 345–352. Personality, 56 , 453–486. Killgore, W. D., Kahn-Greene, E. T., Lipizzi, E. L., Mayer, J. D., DiPaolo, M. T., & Salovey, P. (1990). Newman, R. A., Kamimori, G. H., & Balkin, T. J. Perceiving affective content in ambiguous visual stim- (2008). Sleep deprivation reduces perceived emotional uli: A component of emotional intelligence. Journal of intelligence and constructive thinking skills. Sleep Personality Assessment, 54 , 772–781. Medicine, 9 (5), 517–526. Mayer, J. D., Salovey, P., Gomberg-Kaufman, S., & Killgore, W. D., Weber, M., Schwab, Z. J., DelDonno, Blainey, K. (1991). A broader conception of mood S. R., Kipman, M., Weiner, M. R., & Rauch, S. L. experience. Journal of Personality and Social (2012). Gray matter correlates of trait and ability mod- Psychology, 60 , 100–111. els of emotional intelligence. NeuroReport, 23 (9), Mayer, J. D., Caruso, D. R., & Salovey, P. (2000). Selecting 551–555. a measure of emotional intelligence: The case for abil- Kringelbach, M. L. (2005). The human orbitofrontal cor- ity testing. In R. Bar-On & J. D. A. Parker (Eds.), tex: Linking reward to hedonic experience. Nature Handbook of emotional intelligence (pp. 320–342). Reviews Neuroscience, 6 , 691–702. New York: Jossey-Bass. Landy, F. L. (2005). Some historical and scientifi c issues Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G. related to research on emotional intelligence. Journal (2001). Emotional intelligence as a standard intelli- of Organizational Behaviour, 26 (4), 411–424. gence. Emotion, 1(3), 232–242. Leeper, R. W. (1948). A motivational theory of emotions Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G. to replace “emotions as disorganized responses”. (2003). Measuring emotional intelligence with the Psychological Review, 55 , 5–21. MSCEIT V2.0. Emotion, 3 (1), 97–105. Leslie, J. B., & Fleenor, J. W. (1998). Feedback to manag- Mayer, J. D., Salovey, P., & Caruso, D. R. (2004). ers: A review and comparison of multi-rater instru- Emotional intelligence: Theory, fi ndings, and implications. Psychological Inquiry, 15 (3), 197–215. 24 Current Concepts in the Assessment of Emotional Intelligence 401

Mayer, J. D., Barsade, S. G., & Roberts, R. D. (2008). Petrides, K. V. (2009). Technical manual for the Trait Human abilities: Emotional intelligence. Annual Emotional Intelligence Questionnaires (TEIQue) . Review of Psychology, 59 , 507–536. London: London Psychometric Laboratory. McClelland, J. L. (1999). Cognitive modeling: Petrides, K. V., & Furnham, A. (2000). On the dimen- Connectionist. In R. A. Wilson & F. C. Keil (Eds.), sional structure of emotional intelligence. Personality The MIT encyclopedia of the cognitive sciences and Individual Differences, 29 , 313–320. (pp. 137–141). Cambridge, MA: MIT Press. Petrides, K. V., & Furnham, A. (2001). Trait emotional McEnrue, M. P., & Groves, K. (2006). Choosing among intelligence: Psychometric investigation with refer- tests of emotional intelligence: What is the evidence? ence to established trait taxonomies. European Journal Human Resource Development Quarterly, 17 (1), of Personality, 15 , 425–448. 9–42. Petrides, K. V., & Furnham, A. (2006). The role of trait McGrath, R. E. (2005). Conceptual complexity and con- emotional intelligence in a gender-specifi c model of struct validity. Journal of Personality Assessment, 85 , organizational variables. Journal of Applied Social 112–124. Psychology, 36 , 552–569. Mendelson, R. A. (2012). From athlete to employee: Petrides, K. V., Furnham, A., & Mavroveli, S. (2007a). A quantitative study of the impact of emotional intel- Trait emotional intelligence: Moving forward in the ligence in the workplace – Elite athletes versus fi eld of EI. In G. Matthews, M. Zeidner, & R. Roberts nonathletes (Doctoral Dissertation). Available from (Eds.), Emotional intelligence: Knowns and unknowns ProQuest Dissertations & Theses: Full Text (Series in affective science). Oxford: Oxford (3524312). University Press. Multi-Health Systems Inc. (2011). Emotional Quotient Petrides, K. V., Pita, R., & Kokkinaki, F. (2007b). The loca- Inventory 2.0: User’s handbook. Toronto: Multi- tion of trait emotional intelligence in personality factor Health Systems. space. British Journal of Psychology, 98 , 273–289. Murensky, C. L. (2000). The relationships between emo- Porras, J. I., & Anderson, B. (1981). Improving manage- tional intelligence, personality, critical thinking abil- rial effectiveness through modeling-based training. ity and organizational leadership performance at Organizational Dynamics, 9 , 60–77. upper levels of management (Doctoral Dissertation). Roth, J. (2011). The relationship between emotional intel- Available from ProQuest Dissertations & Theses: Full ligence and pastor leadership in turnaround churches Text (9962991). (Doctoral Dissertation). Available from ProQuest Murphy, K. R. (2006). A critique of emotional intelli- Dissertations & Theses: Full Text (3487845). gence: What are the problems and how can they be Roundy, C. A. (2007). Workplace bullying: Investigating fi xed? Mahwah: Lawrence Erlbaum Associates. the potential link with emotional intelligence (Doctoral O’Boyle, E. H., Jr., Humphrey, R. H., Pollack, J. M., Dissertation). Available from ProQuest Dissertations Hawver, T. H., & Story, P. A. (2010). The relation & Theses: Full Text (3302625). between emotional intelligence and job performance: Saklofske, D. H., Austin, E. J., & Minski, P. S. (2003). A meta-analysis. Journal of Organizational Behavior, Factor structure and validity of a trait emotional intel- 32 (5), 788–818. ligence measure. Personality and Individual Osborne, A. M. (2012). The relationship between emo- Differences, 34 , 707–721. tional intelligence and leadership practices among Salovey, P., & Birnbaum, D. (1989). The infl uence of physicians (Doctoral Dissertation). Available from mood on health-relevant cognitions. Journal of ProQuest Dissertations & Theses: Full Text (3523050). Personality and Social Psychology, 57 , 539–551. Parker, J. D. A., Creque, R. E., Sr., Barnhart, D. L., Harris, Salovey, P., & Mayer, J. D. (1990). Emotional intelli- J. I., Majeski, S., Wood, L. M., Bond, B. J., & Hogan, gence. Imagination, Cognition, and Personality, 9 , M. J. (2004a). Academic achievement in high school: 185–211. Does emotional intelligence matter? Personality and Salovey, P., & Rodin, J. (1985). Cognitions about the self: Individual Differences, 37 , 1321–1330. Connecting feeling states to social behavior. Review of Parker, J. D. A., Summerfeldt, L. J., Hogan, M. J., & Personality and Social Psychology, 6 , 143–166. Majeski, S. (2004b). Emotional intelligence and aca- Salovey, P., Mayer, J. D., & Rosenhan, D. L. (1991). demic success: Examining the transition from high Mood and helping: Mood as a motivator of helping school to university. Personality and Individual and helping as a regulator of mood. Review of Differences, 36 , 163–172. Personality and Social Psychology, 12 , 215–237. Parker, J. D. A., Keefer, K. V., & Wood, L. M. (2011). Salovey, P., Hsee, C., & Mayer, J. D. (1993). Emotional Toward a brief multidimensional assessment of emo- intelligence and the self-regulation of affect. In D. M. tional intelligence: Psychometric properties of the Wegner & J. W. Pennebaker (Eds.), Handbook of mental emotional quotient inventory—Short form. control (pp. 258–277). Englewood Cliffs: Prentice-Hall. Psychological Assessment, 23 (3), 762–777. Salovey, P., Mayer, J. D., Goldman, S., Turvey, C., & Pesuric, A., & Byham, W. (1996). The new look in behav- Palfai, T. (1995). Emotional attention, clarity, and ior modeling. Training and Development, 50 (7), repair: Exploring emotional intelligence using the trait 24–30. meta-mood scale. In J. Pennebaker (Ed.), Emotion, 402 S.J. Stein and J.M. Deonarine

disclosure, and health (pp. 125–154). Washington, Van Rooy, D. L., & Viswesvaran, C. (2004). Emotional DC: American Psychological Association. intelligence: A meta-analytic investigation of predic- Salovey, P., Stroud, L. R., Woolery, A., & Epel, E. S. tive validity and nomological net. Journal of (2002). Perceived emotional intelligence, stress reac- Vocational Behavior, 65 , 71–95. tivity, and symptom reports: Further explorations Walker, A. E. (2001). Emotional intelligence of the using the trait meta-mood scale. Psychology and classroom teacher (Doctoral Dissertation). Available Health, 17 , 611–627. from ProQuest Dissertations & Theses: Full Text Saxe, R. (2006). Uniquely human social cognition. (9996964). Current Opinion in Neurobiology, 16 , 235–239. Walpole, P., Isaac, C. L., & Reynders, H. (2008). Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D. J., A comparison of emotional and cognitive intelligences Cooper, J. T., Golden, C. J., & Dornheim, L. (1998). in people with and without temporal lobe epilepsy. Development and validation of a measure of emo- Epilepsia, 49 (8), 1470–1474. tional intelligence. Personality and Individual Wechsler, D. (1940). Nonintellective factors in general Differences, 25 , 167–177. intelligence. Psychological Bulletin, 37 , 444–445. Schutte, N. S., Malouff, J. M., & Bhullar, N. (2009). The Wechsler, D. (1943). Nonintellective factors in general assessing emotions scale. In C. Stough, D. H. intelligence. Journal of Abnormal Psychology, 38 , Saklofske, & J. D. A. Parker (Eds.), Assessing emo- 100–104. tional intelligence: Theory, research, and applications Wechsler, D. (1997). Wechsler adult intelligence test (pp. 119–134). New York: Springer. administration and scoring manual . San Antonio: The Shank, J. C. (2012). Emotional intelligence and educa- Psychology Corporation. tional leadership: Measuring the emotional intelli- Weinberger, L. A. (2002). Emotional intelligence: Its con- gence of educational leaders and their corresponding nection to HRD theory and practice. Human Resource student achievement (Doctoral Dissertation). Available Development, 1 , 215–243. from ProQuest Dissertations & Theses: Full Text Weissman, M. (1999). Social adjustment scale – Self- (3510549). report (SAS-SR) user’s manual . North Tonawanda: Smith, G. T., McCarthy, D. M., & Zapolski, T. C. B. Multi-Health Systems. (2009). On the value of homogenous construct for Woods, B. R. (2010). The relationship between a manag- construct validation, theory testing and the description er’s emotional intelligence and perceived leadership of psychopathology. Psychological Assessment, 21 , style (Doctoral Dissertation). Available from ProQuest 272–284. Dissertations & Theses: Full Text (3409369). Spencer, L. (2001). The economic value of emotional Xue, G., Lu, Z., Levin, I. P., & Bechara, A. (2010). The intelligence competencies and EIC-based HR pro- impact of prior risk experiences on subsequent risky grammes. In C. Cherniss & D. Goleman (Eds.), The decision-making: The role of the insula. NeuroImage, emotionally intelligent workplace (pp. 45–83). San 50 (2), 709–716. Francisco: Jossey-Bass. Zeidner, M., Matthews, G., & Roberts, R. (2004). Spencer, L. M. J., & Spencer, S. (1993). Competence at Emotional intelligence in the workplace: A critical work . New York: John Wiley. review. Applied Psychology: An International Review, Thompson, B. L., Waltz, J., Croyle, K., & Pepper, A. C. 53 , 371–399. (2007). Trait meta-mood and affect as predictors of Zeidner, M., Roberts, R., & Matthews, G. (2008). The somatic symptoms and life satisfaction. Personality science of emotional intelligence. European Psychologist, and Individual Differences, 43 , 1786–1795. 13 (1), 64–78. P a r t V

Applications of Intellectual Theory Intelligence and Success 25 Tarmo Strenze

This chapter is about the relationship between intelligence and success. The central message of Intelligence and Success: the chapter is simple: intelligent people are more An Overview successful than less intelligent people. The chapter will review the scientifi c evidence for that So what is the evidence for the relationship message and discuss some theoretical problems, between intelligence and success? Many readers in an attempt to show that the message is really would probably be convinced by specifi c exam- not that simple. But at fi rst we should probably ples of people who are known to be highly intel- ask: What is success? ligent and who have achieved great success in Success can be defi ned as doing or achieving some fi eld. For instance, Bill Gates is rumored to something that is generally considered desirable have received an extremely high score on his col- in the society. Naturally, there are many ways to lege SAT,1 which would mean that he also must be successful, and some of these ways may be in have a very high IQ score. One can guess that Bill confl ict with one another, so that achieving suc- Gates’ rise to one of the richest and most power- cess in one fi eld may restrict you from having ful men on earth must have something to do with success in another fi eld. Some readers may be his IQ. However, such cases never prove anything tempted to say that success is a purely subjective conclusively because we can always fi nd some phenomenon, which each individual defi nes for contrary examples. For instance, a legendary oneself. That is certainly true, but it seems that punk groupie Nancy Spungen had an IQ of 134, there is usually a high degree of consensus in and still she became a drug addict and was society as to what is desirable and what is not. expelled from school and, ultimately, from her Even if there are individuals who reject some own home by her own parents (Spungen 1983 ). form of success (for instance, claim that they do Instead of well-selected examples, we should not care about money), that form of success still look at the statistical relationships between intel- remains socially important and worthy of ligence and various forms of success. Such rela- research. tionships have been examined in numerous studies, and it is not possible to review all of them here. This review will concentrate on meta- analyses because results from meta-analyses are T. Strenze , Ph.D. (*) Institute of Social Studies , University of Tartu, Estonia , Lossi 36-405 , Tartu 51003 , Estonia 1 The Biography Channel website. Retrieved Sep 01, 2013, e-mail: [email protected] from http://www.biography.com/people/bill-gates-9307520

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 405 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_25, © Springer Science+Business Media New York 2015 406 T. Strenze

Table 25.1 Relationship between intelligence and measures of success (Results from meta-analyses) Measure of success r k N Source Academic performance in primary education .58 4 1,791 Poropat (2009 ) Educational attainment .56 59 84,828 Strenze (2007 ) Job performance (supervisory rating) .53 425 32,124 Hunter and Hunter (1984 ) Occupational attainment .43 45 72,290 Strenze (2007 ) Job performance (work sample) .38 36 16,480 Roth et al. (2005 ) Skill acquisition in work training .38 17 6,713 Colquitt et al. (2000 ) Degree attainment speed in graduate school .35 5 1,700 Kuncel et al. (2004 ) Group leadership success (group productivity) .33 14 Judge et al. (2004 ) Promotions at work .28 9 21,290 Schmitt et al. (1984 ) Interview success (interviewer rating of applicant) .27 40 11,317 Berry et al. (2007 ) Reading performance among problem children .26 8 944 Nelson et al. (2003 ) Becoming a leader in group .25 65 Judge et al. (2004 ) Academic performance in secondary education .24 17 12,606 Poropat (2009 ) Academic performance in tertiary education .23 26 17,588 Poropat (2009 ) Income .20 31 58,758 Strenze (2007 ) Having anorexia nervosa .20 16 484 Lopez et al. (2010 ) Research productivity in graduate school .19 4 314 Kuncel et al. (2004 ) Participation in group activities .18 36 Mann (1959 ) Group leadership success (group member rating) .17 64 Judge et al. (2004 ) Creativity .17 447 Kim (2005 ) Popularity among group members .10 38 Mann (1959 ) Happiness .05 19 2,546 DeNeve & Cooper (1998 ) Procrastination (needless delay of action) .03 14 2,151 Steel (2007 ) Changing jobs .01 7 6,062 Griffeth et al. (2000 ) Physical attractiveness −.04 31 3,497 Feingold (1992 ) Recidivism (repeated criminal behavior) −.07 32 21,369 Gendreau et al. (1996 ) Number of children −.11 3 Lynn (1996 ) Traffi c accident involvement −.12 10 1,020 Arthur et al. (1991 ) Conformity to persuasion −.12 7 Rhodes and Wood (1992 ) Communication anxiety −.13 8 2,548 Bourhis and Allen (1992 ) Having schizophrenia −.26 18 Woodberry et al. (2008 ) r correlation between intelligence and the measure of success, k number of studies included in the meta-analysis, N number of individuals included in the meta-analysis more reliable than results from single studies. desirable outcomes and negatively correlated Table 25.1 presents a list of meta-analytic corre- with undesirable outcomes. Let us take a closer lations between IQ scores and a number of outlook at some of the outcomes in Table 25.1 . One comes that can reasonably be designated as of the most classical outcomes of intelligence is “success” (or lack of success). Of course, several academic performance, typically measured by important forms of success have never been sub- grade point average or a specifi c academic test. jected to meta-analysis and are, consequently, It is classical because IQ testing was originally absent from Table 25.1 ; on the other hand, some invented to predict academic success of students, forms of success have been meta-analyzed more so it is no surprise that its correlation with intel- than once, in which case I used the largest ligence is positive. However, the correlation is meta-analysis. perhaps not as strong as one might have expected. Overall, it is evident from Table 25.1 that Only among elementary school students is the intelligence tends to be positively correlated with correlation really noteworthy (.58); among high 25 Intelligence and Success 407 school and college students, it is much lower his or her work tasks. That is obviously of great (0.24 and 0.23). This result goes against the claim relevance to organizations, and much research of some critics (e.g., McClelland 1973 ) that IQ has been devoted to fi nding good predictors of test is nothing else but a test of school learning. job performance. Positive correlations with But why is the correlation weaker on higher edu- supervisory ratings of job performance (0.53) cational levels? The answer probably has to do and work sample tests (0.38) in Table 25.1 dem- with decreasing variance: as people move from onstrate that intelligent people are good workers, elementary education to secondary and tertiary and IQ tests are, therefore, good personnel selec- education, less intelligent students are excluded tion devices. Indeed, some researchers believe with each transition, reducing the variance of that IQ tests are the best personnel selection intelligence and thereby also its correlation with devices available (Schmidt and Hunter 1998 ). An academic performance. interesting fi nding is that IQ tests are better pre- A highly desirable form of success in the dictors of performance among cognitively com- modern world is career success (or socioeco- plex jobs, compared to less complex jobs (Ones nomic success). In Table 25.1 , it is represented by et al. 2005 ). This means that IQ tests are very education, occupation, income, and promotions. useful in selecting good engineers, architects, or All these things are positively correlated with dentists (cognitively complex jobs according to intelligence – correlation with educational attain- Roos and Treiman 1980 ); IQ tests are less useful ment is among the strongest correlations in for selecting good dishwashers, weavers, or gar- Table 25.1 (0.56), correlation with occupational bage collectors, although, even among dishwash- attainment is also strong (0.43), and income and ers, it is obvious that an intelligent worker is promotions have somewhat weaker correlations better than a less intelligent one. (0.20 and 0.28). These results mean that intelli- Some correlations in Table 25.1 are not quite gent people generally occupy higher positions in as expected. For instance, the correlation with society. A society with such IQ-based stratifi ca- happiness is only 0.05. One might wonder, if tion is called meritocracy (Young 1958) and is intelligent people are so successful in achieving often considered to be a fair and effi cient form of desirable goals, then how come they are not sig- society, because people are allowed to achieve nifi cantly happier than less intelligent people? positions that correspond to their talents, as The answer to this is simple. According to some opposed to being allocated to positions according prominent theories of happiness (see Diener et al. to their social origin (position of parents), race, or 1999 ), the personal level of subjective well-being gender. There has been quite a lot of dispute on is actually rather stable and not much dependent how meritocratic contemporary western society on life events. Happiness is like a personality really is. In 1994, Herrnstein and Murray pub- trait; you either have it or not, and things you lished a book called The Bell Curve that became achieve in life (or fail to achieve) will not affect it notorious for claiming that, in the United States, very strongly. intelligence has a considerably stronger effect on Another surprising result in Table 25.1 is the various forms of success than social origin and positive correlation with anorexia nervosa (0.20). that American society is moving toward IQ-based Most studies have found that intelligent people class system. Saunders (1997 ) found that the are healthier and live longer than less intelligent same might be true for Great Britain. Such results people (Calvin et al. 2010 ), so why are they more imply that society is rather meritocratic. However, likely to contract a serious disorder like anorexia critics have argued that these studies overesti- nervosa? To offer a speculative answer, we can mated the effect of intelligence and underesti- use the evolutionary theory of intelligence devel- mated the effect of social origin (Breen and oped by Kanazawa (2004 ). According to this Goldthorpe 1999 ; Fischer et al. 1996 ). theory, general intelligence is a brain function Another important form of success is job per- that has evolved in human evolution to deal with formance, a measure of how well a worker performs evolutionarily novel tasks. Take, for instance, 408 T. Strenze activities like fi nding food, having children, and have come up with lower estimates, somewhere collaborating with other humans – these are all around 0.50 (Devlin et al. 1997), which is still tasks that our ancestors have been solving for quite high. Given the substantial genetic basis of millions of years, and for these tasks, it is likely intelligence and the robust relationship between that specifi c hereditary brain mechanisms have intelligence and social success, one can conclude developed that promote the successful perfor- that the difference between successful and less mance of that task. But activities like getting successful people is also genetic, to some degree. good grades at school, making a lot of money, or This was indeed the conclusion made by being thin have just recently been invented by our Herrnstein and Murray (1994 ). They argued that society, and they do not (yet) have their own brain contemporary western society allows people to mechanisms. For these novel tasks, people use fulfi ll their genetic potential, which means that intelligence, which is a generic ability to solve people can achieve as much success as their any type of (unexpected) problems. Kanazawa genetic IQ enables them to achieve; the social notes that intelligence correlates positively with position of each individual is, thus, ultimately evolutionarily novel activities, but the correlation determined by genes, and western society evolves with ancient activities is zero or even negative. toward genetic hierarchy where people with That is also evident in Table 25.1 , which mostly “good genes” live in luxury and people with “bad lists novel school- or job-related forms of success genes” struggle to survive. That system is further that have the expected positive correlation with solidifi ed by assortative mating, the tendency for intelligence, but one of the most ancient forms of people to marry and have children with partners success, number of children, has a negative cor- of similar IQ. Children of intelligent and rich par- relation (−0.11). As for anorexia nervosa, the ents have, thus, a double advantage – they inherit desire for thinness (the basis for anorexia) is their parents’ IQ as well as their resources – chil- clearly a novel goal that probably takes some dren of less intelligent poor parents, however, are intelligence to achieve it. handicapped on both accounts (Herrnstein and Overall, the results in Table 25.1 present a Murray 1994 ). kind of a portrait of an intelligent person with Is there any reason to believe that contempo- positive correlations depicting the characteristics rary society could be such a genetic caste soci- that an intelligent person is likely to have and ety? Behavioral genetic research has found that negative correlations depicting the characteristics almost all human characteristics and behaviors he or she is not likely to have. To make sense of have some genetic basis. In addition to intelli- these correlations, it is good to have theories like gence, it has been found that criminality, alcohol- the one by Kanazawa (described above) that do ism, and smoking also have a genetic component not concentrate on just one specifi c form of suc- (Malouff et al. 2008). The same is true about the cess, but strive to explain the whole pattern of main social status indicators – education, occupa- correlations. tion, and income. The heritability coeffi cient of education is about 0.50 (Rowe et al. 1999 ; Silventoinen et al. 2000 ), heritability of occupa- Genes, Intelligence, and Success tion is about 0.40 (Tambs et al. 1989 ; Plomin and Bergeman 1991 ), and heritability of income is It is diffi cult to discuss intelligence without going about 0.30 (Taubman 1976 ; Rowe et al. 1999 ). into discussion about genes. Intelligence cer- This means that the similarity of parents and chil- tainly would not be such a controversial subject if dren in terms of social status is partly due to the there was no reason to believe that IQ differences genetic transmission of characteristics that foster between people are, to a considerable degree, (or hinder) status attainment. These numbers are caused by genetic differences. Some researchers interesting, but it should be noted that the herita- have suggested that the heritability of intelligence bility values of status characteristics are still much could be as high as 0.80 (Jensen 1969), but most lower than the heritability values of physical 25 Intelligence and Success 409 characteristics. The heritability of height, for irrespective of their social background. At the instance, is about 0.75 (Silventoinen et al. 2000 ; same time, the occupational system became more Benyamin et al. 2005 ). So, the diagnosis of complex with a lot of new cognitively demanding “genetic caste society” must be an overstatement. jobs requiring intelligent workers. These two his- There is probably some correspondence between torical developments – increasing openness and the genetic structure of human population and the complexity – are the main social factors that cre- social structure of human society, but not any ated the positive correlation between intelligence genetic IQ castes with impenetrable borders. and career success, according to Herrnstein and And what is the role of intelligence in the Murray (1994 ). genetic transmission of social status? Given that This scenario sounds convincing, but it has intelligence is heritable, Herrnstein and Murray been criticized on several grounds. Some authors made an automatic conclusion that intelligence have presented evidence showing that intelligent must be the characteristic that plays the central people were, in fact, able to be successful in ear- role in the intergenerational transmission of lier historical periods. Weiss describes the towns genetic advantages. But that conclusion may of the sixteenth- to seventeenth-century Germany have been premature. Bowles and Gintis (2002 ) where a lot of young men from modest social have calculated that the role of “IQ genes” in the background were able to work themselves into parent–child similarity of social status is actually higher positions (Weiss 1995 ). Weiss speculates quite small. They do not deny that social status is that these men probably had higher than average heritable, but they claim that it is so mostly due to IQ, given that they had no other advantage that other genetic characteristics, like race, health, or would explain their rise. Botton describes how personality. This conclusion by Bowles and Napoleon changed the nineteenth-century French Gintis rests on sophisticated calculations, which I army so that new offi cers were recruited and pro- have never seen anybody else make. I would treat moted on the basis of their talent, rather than their conclusion with some caution until their social background (Botton 2004 ). Again, we can method has found more acceptance. speculate that this new system increased the correlation between intelligence and rank in the army (and perhaps in society, more generally). History, Intelligence, and Success Adkins and Guo go much further back in time to claim that the positive effect of genetic character- The evidence discussed so far has come exclu- istics (such as intelligence) on success was prob- sively from contemporary western societies. But ably the strongest in the archaic society of what about earlier historical periods and less hunter-gatherers, before the emergence of private developed societies? Do these societies also have property and desire to pass it on to children intelligent people on top? There is, of course, no (Adkins and Guo 2008 ). Of course, “success” direct evidence on the intelligence of people from had a completely different meaning back then, earlier than the twentieth century. But the general but it depended entirely on individual ability. opinion seems to be that earlier historical periods Another line of criticism against The Bell mostly did not allow intelligent people to get Curve concerns the claim of increasing meritoc- ahead in society. These societies had a rigid class racy during the twentieth century. A number of system, and a person born to lower ranks had no studies have tried to test this claim, and most opportunity to raise to upper ranks, no matter have failed to find the strengthening of the how intelligent he or she was. According to The IQ–success correlation, predicted by The Bell Bell Curve, western societies really started to Curve (Hauser and Huang 1997 ; Bowles et al. become more meritocratic only in the middle of 2001 ; Strenze 2007 ). All these studies have used the twentieth century. Around that time, the edu- data collected over several decades (mostly start- cational system became more democratic, and ing with the 1960s), and they have not found colleges were opened up to intelligent youth, any signs of the IQ–success relationship getting 410 T. Strenze

Fig. 25.1 The relationship between IQ–success relationship in a country (vertical axis) and economic development of a country (horizontal axis)

stronger during that time. Does it mean that the cross-national analysis of that kind has not yet thesis of increasing meritocracy is wrong? been conducted because of lack of data. But as a Perhaps, but it must be said that these studies preliminary gauge, take a look at Fig. 25.1 that have almost exclusively relied on data from the presents a simple scatterplot based on data from second half of the twentieth century, beginning the International Adult Literacy Survey (IALS). with the 1960s; there is little usable data from the IALS is a cross-national survey, conducted in fi rst half of the century. According to The Bell 1994–1998, that measured the literacy ability of Curve, however, the most radical change took adults in 20 countries; it also included data on the place somewhere in the middle of the century (at career success of these adults. I calculated for least in the Unites States) – so it is possible that each country the effect (regression coeffi cient) of the available data are simply too late for the literacy ability on income (see Strenze 2013 , for change we are looking for. But at least it seems details) – that effect is presented on the vertical relatively safe to say that after 1960–1970 there axis of Fig. 25.1 . The horizontal axis of Fig. 25.1 has been little change in the correlation between is the 1995 per capita gross national product IQ and success. (GDP), a measure of societal development. Based An alternative way to address the same issue on the reasoning offered above, one would expect is to compare data from different countries to see to fi nd a positive relationship between GDP and if more developed countries have a stronger rela- ability–income correlation, but in fact the relationship between intelligence and success – that tionship in Fig. 25.1 is negative. The more devel- would support the idea that societies become oped countries with higher GDP, like Norway or more meritocratic as they evolve from traditional Switzerland, tend to exhibit lower correlations into industrial and postindustrial. A sophisticated between people’s ability and income than the less 25 Intelligence and Success 411 developed countries like Chile or Hungary. Of outcomes and to keep away from undesirable ones course, the data used in Fig. 25.1 is far from per- (but there are also some interesting exceptions to fect, and the number of countries is too small to that). The relationship between intelligence and draw any ironclad conclusions. However, a simi- success is partly based on genetics, as intelligence lar result was obtained by Psacharopoulos and is itself partly genetic characteristic. But on the Patrinos (2004 ) as they compared the relationship other hand, the relationship is partly based on soci- between education and income in nearly 100 etal context, as only certain social conditions allow countries and found that the relationship is stron- intelligent people to fulfi ll their potential. ger in less developed countries. That supports the As we think about intelligence and success, we impression that, among the societies that exist must remember that the scientifi c question about today, less developed societies are the ones where the relationship between intelligence and success people with higher ability (and education) get is closely connected to other scientifi c questions better fi nancial rewards. about intelligence and, most importantly, to the Based on these results, we can piece together following question: What is it that IQ tests really a speculative scenario of the history of the measure? This chapter was based on the implicit IQ–success relationship. The primitive society of assumption that IQ tests are reasonably good mea- our ancestors was presumably rather meritocratic sures of general cognitive ability, but not all social as each person had to earn one’s place in the tribe scientists would agree with that. To make sense of using one’s own abilities and nobody got any the correlations between intelligence and success, help from their “rich daddy.” As human society one must have a view on IQ testing and on the grew more complex, large inequalities between nature of intelligence, in general, which is why I social groups emerged (think of slaves and citi- now direct the reader to other chapters of this book zens in ancient Rome or peasants and aristocrats where these related topics are discussed. in medieval Europe), and most people were des- tined to live in the social class of their parents – intelligence probably had little effect on people’s References life in these societies. These rigid class boundaries started to break down with the advent of Adkins, D. E., & Guo, G. (2008). Societal development industrial society, as democratic values became and the shifting infl uence of the genome on status attainment. Research in Social Stratifi cation and prevalent and there was increased demand for Mobility, 26 , 235–255. able workers – that created an opportunity for Arthur, W., Barret, G. V., & Alexander, R. A. (1991). intelligent people to move up in the social ladder. Prediction of vehicular accident involvement: A meta- This process apparently reached its apex in the analysis. Human Performance, 4 , 89–105. Benyamin, B., Wilson, V., Whalley, L. J., Visscher, P. M., middle of the twentieth century (in western soci- & Deary, I. J. (2005). Large, consistent estimates of eties) when the fi nal push toward liberalization of the heritability of cognitive ability in two entire popu- educational and occupational market took place. lations of 11-year-old twins from Scottish mental sur- But after that, it seems, the relationship between veys of 1932 and 1947. Behavior Genetics, 35 , 525–534. intelligence and career success has stayed the Berry, C. M., Sackett, P. R., & Landers, R. N. (2007). same or even declined, possibly due to the ten- Revisiting interview-cognitive ability relationships: dencies in postindustrial welfare society to Attending to specifi c range restriction mechanisms in reduce inequality and competition. meta-analysis. Personnel Psychology, 60 , 837–874. Botton, A. (2004). Status anxiety. Hamish Hamilton: Penguin. Bourhis, J., & Allen, M. (1992). Meta-analysis of the rela- Conclusion tionship between communication apprehension and cognitive performance. Communication Education, 41 , 68–76. This short chapter was about the social conse- Bowles, S., & Gintis, H. (2002). The inheritance of quences of intelligence. We saw that high IQ gen- inequality. Journal of Economic Perspectives, 16 , erally helps people to achieve numerous desirable 3–30. 412 T. Strenze

Bowles, S., Gintis, H., & Osborne, M. (2001). The deter- and job performance: Can one construct predict them minants of earnings: A behavioral approach. Journal all? Journal of Personality and Social Psychology, 86 , of Economic Literature, 39 , 1137–1176. 148–161. Breen, R., & Goldthorpe, J. H. (1999). Class inequality Lopez, C., Stahl, D., & Tchanturia, K. (2010). Estimated and meritocracy: A critique of Saunders and an alter- intelligence quotient in anorexia nervosa: A system- native analysis. British Journal of Sociology, 50 , 1–27. atic review and meta-analysis of the literature. Annals Calvin, C. M., Deary, I. J., Fenton, C., Roberts, B. A., Der, of General Psychiatry, 9 , 40. G., Leckenby, N., & Batty, G. D. (2010). Intelligence in Lynn, R. (1996). Dysgenics. Genetic deterioration in youth and all-cause-mortality: Systematic review with modern populations . London: Praeger. meta-analysis. International Journal of Epidemiology , Malouff, J. M., Rooke, S., & Schutte, N. S. (2008). The 1–19. heritability of human behavior: Results of aggregating Colquitt, J. A., LePine, J. A., & Noe, R. (2000). Toward an meta-analyses. Current Psychology, 27 , 153–161. integrative theory of training motivation: A meta- Mann, R. D. (1959). A review of the relationships between analytic path analysis of 20 years of research. Journal personality and performance in small groups. of Applied Psychology, 85 , 678–707. Psychological Bulletin, 56 , 241–270. Devlin, B., Daniels, M., & Roeder, K. (1997). The herita- McClelland, D. C. (1973). Testing for competence rather bility of IQ. Nature, 388 , 468–471. than for intelligence. American Psychologist, 28, 1–14. Diener, E., Suh, E. M., Lucas, R. E., & Smith, H. L. Nelson, J. R., Benner, G. J., & Gonzalez, J. (2003). (1999). Subjective well being: Three decades of prog- Learner characteristics that infl uence the treatment ress. Psychological Bulletin, 125 , 276–302. effectiveness of early literacy interventions: A meta- DeNeve, K. M. & Cooper, H. (1998). The happy person- analytic review. Learning Disabilities Research & ality: A meta-analysis of 137 personality traits and Practice, 18 , 255–267. subjective well-being. Psychological Bulletin, 124, Ones, D. S., Viswesvaran, C., & Dilchert, S. (2005). 197–229. Cognitive ability in personnel selection decisions. In Feingold, A. (1992). Good-looking people are not what A. Evers, N. Anderson, & O. Voskuijl (Eds.), The we think. Psychological Bulletin, 111 , 304–341. Blackwell handbook of personnel selection (pp. 143– Fischer, C. S., Hout, M., Jankowski, M. S., Lucas, S. R., 173). Malden, MA: Blackwell. Swidler, A., & Voss, K. (1996). Inequality by design. Plomin, R., & Bergeman, C. S. (1991). The nature of nur- Cracking the Bell Curve myth. Princeton, NJ: ture: Genetic infl uence on environmental measures. Princeton University Press. Behavioral and Brain Sciences, 14 , 373–427. Gendreau, P., Little, T., & Goggin, C. (1996). A meta- Poropat, A. E. (2009). A meta-analysis of the fi ve-factor analysis of the predictors of adult offender recidivism: model of personality and academic performance. What works! Criminology, 34 , 575–607. Psychological Bulletin, 135 , 322–338. Griffeth, R. W., Hom, P. W., & Gaertner, S. (2000). A Psacharopoulos, G., & Patrinos, H. A. (2004). Returns to meta-analysis of antecedents and correlates of investment in education: A further update. Education employee turnover: Update, moderator tests, and Economics, 12 , 111–134. research implications for the next millennium. Journal Rhodes, N., & Wood, W. (1992). Self-esteem and intelli- of Management, 26 , 463–488. gence affect infl uenceability: The mediating role of Hauser, R. M., & Huang, M.-H. (1997). Verbal ability and message reception. Psychological Bulletin, 111 , socioeconomic success: A trend analysis. Social 156–171. Science Research, 26 , 331–376. Roos, P. A., & Treiman, D. J. (1980). DOT scales for the Herrnstein, R. J., & Murray, C. (1994). The bell curve. 1970 census classifi cation. In A. R. Miller, D. J. Intelligence and class structure in American life . Treiman, P. S. Cain, & P. A. Roos (Eds.), Work, jobs, New York: Free Press. and occupations: A critical review of the “Dictionary Hunter, J. E., & Hunter, R. F. (1984). Validity and utility of Occupational Titles” (pp. 336–389). Washington, of alternative predictors of job performance. D.C.: National Academy Press. Psychological Bulletin, 96 , 72–98. Roth, P. L., Bobko, P., & Farland, L. A. (2005). A Meta- Jensen, A. R. (1969). How much can we boost IQ and analysis of work sample test validity: Updating and scholastic achievement? Harvard Educational Review, integrating some classic literature. Personnel 39 , 1–123. Psychology, 58 , 1009–1037. Judge, T. A., Colbert, A. E., & Ilies, R. (2004). Intelligence Rowe, D. C., Vesterdal, W. J., & Rodgers, J. L. (1999). and leadership: A quantitative review and test of theo- Herrnstein's syllogism: Genetic and shared environ- retical propositions. Journal of Applied Psychology, mental infl uences on IQ, education, and income. 89 , 542–552. Intelligence, 26 , 405–423. Kanazawa, S. (2004). General intelligence as a domain- Saunders, P. (1997). Social mobility in Britain: An empiri- specifi c adaptation. Psychological Review, 111 , 512–523. cal evaluation of two competing explanations. Kim, K. H. (2005). Can only intelligent people be creative? Sociology, 31 , 261–288. Journal of Secondary Gifted Education, 16 , 57–66. Schmidt, F. L., & Hunter, J. E. (1998). The validity and Kuncel, N. R., Hezlett, S. A., & Ones, D. S. (2004). utility of selection methods in personnel psychology: Academic performance, career potential, creativity, Practical and theoretical fi ndings of 85 years of 25 Intelligence and Success 413

research ﬁ ndings. Psychological Bulletin, 124 , Strenze, T. (2013). Allocation of talent in society and its 262–274. effect on economic development. Intelligence, 41 , Schmitt, N., Gooding, R. Z., Noe, R. A., & Kirsch, M. 193–202. (1984). Meta-analyses of validity studies published Tambs, K., Sundet, J. M., Magnus, P., & Berg, K. (1989). between 1964 and 1982 and the investigation of study Genetic and environmental contributions to the covari- characteristics. Personnel Psychology, 37 , 407–422. ance between occupational status, educational attain- Silventoinen, K., Kaprio, J., & Lahelma, E. (2000). ment, and IQ: A study of twins. Behavior Genetics, 19 , Genetic and environmental contributions to the asso- 209–222. ciation between body height and educational attain- Taubman, P. (1976). Earnings, education, genetics, and ment: A study of adult Finnish twins. Behavior environment. Journal of Human Resources, 11 , Genetics, 30 , 477–485. 447–461. Spungen, D. (1983). And I don't want to live this life . Weiss, V. (1995). The emergence of a cognitive elite. New York: Fawcett Crest. Comment on The Bell Curve. Mankind Quarterly, 35 , Steel, P. (2007). The nature of procrastination: A meta- 373–390. analytic and theoretical review of quintessential Woodberry, K. A., Giuliano, A. J., & Seidman, L. J. self- regulatory failure. Psychological Bulletin, 133 , (2008). Premorbid IQ in schizophrenia: A meta- 65–94. analytic review. The American Journal of Psychiatry, Strenze, T. (2007). Intelligence and socioeconomic suc- 165 , 579–587. cess: A meta-analytic review of longitudinal research. Young, M. (1958). The rise of meritocracy 1870–2033 . Intelligence, 35 , 401–426. London: Penguin. The Use of Intelligence Tests in the Diagnosis of Speciﬁ c 26 Reading Disability

Nancy Mather and Deborah Schneider

dyslexia. Although we view these terms as being The Use of Intelligence Tests synonymous, throughout this chapter, we will most in the Diagnosis of Specifi c Reading often use the term specifi c reading disability, as Disability some object to the term “dyslexia,” viewing it as a medical diagnosis, or even doubting its very exis- Specifi c reading disability has been the subject tence. As Siegel and Mazabel (2013 ) lamented, of formal academic inquiry for over a century. “We do not understand why the term “dyslexia” is Throughout this period, intelligence tests have often viewed as if it were a four-letter word, not to played a central but constantly evolving role in the be uttered in polite company” (p. 187). evaluation and diagnosis of this disorder. Within this chapter, we do not examine specifi c intelligence tests per se but rather their historic and con- D e fi nitions temporary uses in the identifi cation and diagnosis of specifi c reading disability. We discuss (a) the Although a universally accepted defi nition of spe- current defi nition of reading disability, (b) a brief cifi c reading disability (SRD) has not been estab- historical perspective on the use of intelligence lished, most contemporary defi nitions emphasize tests to identify and diagnose specifi c reading the following characteristics: (a) it is a neurobio- disability, (c) present-day methods of diagnosing logical disorder that affects the mastery of sound- specifi c reading disability, (d) specifi c cognitive symbol correspondences and the fl uency and constructs and their relevance to the accurate diag- automaticity of reading, writing, and spelling; nosis of reading disability, and (e) the future use of (b) it is often accompanied by specifi c weak- intelligence tests in the identifi cation and diagnosis nesses in cognitive factors that predict poor read- of a specifi c reading disability, often referred to as ing and spelling performance, such as phonological awareness or processing speed; (c) it is a lifelong N. Mather , Ph.D. (*) condition, but effective interventions can reduce University of Arizona , its impact; (d) it can occur in individuals of any 7161 East Kiva Way , Tucson , AZ 85750 , USA level of intelligence; and (e) other abilities are e-mail: [email protected] often intact and can even be advanced, thus mak- D. Schneider ing this type of reading disability a specifi c disor- Doctoral Candidate and Department Fellow der. In the next sections, we discuss some of the in Special Education, University of Arizona , 10814 E. Limberlost Rd , Tucson , AZ 85749 , USA most commonly used defi nitions of specifi c read- e-mail: [email protected] ing disability in educational and clinical practice.

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 415 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_26, © Springer Science+Business Media New York 2015 416 N. Mather and D. Schneider

Diagnostic and Statistical Manual International Dyslexia Association of Mental Disorders The most widely accepted defi nition of specifi c The fourth edition of Diagnostic and Statistical reading disability in the United States is that of the Manual of Mental Disorders (DSM-IV-TR) of the International Dyslexia Association (IDA) (Lyon American Psychological Association (APA) et al. 2003 ). Despite changes in federal education (1994 ) defi ned specifi c reading disorder (else- law, the IDA continues to employ a defi nition of where characterized as dyslexia, specifi c reading specifi c reading disability (termed dyslexia ) in disability, or reading-related learning disorder) as which an unexpected discrepancy between read- reading achievement “…that falls substantially ing achievement and cognitive ability is a charac- below that expected given the individual’s chron- teristic feature: “[Dyslexia] is characterized by ological age, measured intelligence, and age- diffi culties with accurate and/or fl uent word recog- appropriate education…[and] signifi cantly nition and by poor spelling and decoding abilities. interferes with academic achievement or with These diffi culties typically result from a defi cit in activities of daily living that require reading the phonological component of language that is skills” (p. 48). The fi fth edition of the Manual often unexpected in relation to other cognitive (DSM-5) (American Psychological Association abilities and the provision of effective classroom 2013), however, has abandoned reading disorder instruction” (International Dyslexia Association as an independent diagnosis altogether, subsum- 2007 ). This defi nition, though employed by the ing it under the diagnostic umbrella of specifi c National Institute of Child Health and Human learning disorder. Under this new defi nition, read- Development (NICHD), is controversial. The ing-related learning disorders may be present major reason that it has been criticized is for pur- when reading and/or spelling abilities “…are sub- portedly undue emphasis on the phonological stantially and quantifi ably below those expected component of dyslexia. A number of authors have for the individual’s chronological age and cause argued that the phonological defi cit model is inad- signifi cant interference with academic or occupa- equate for explaining all cases of reading disorder tional performance, or with activities of daily (Peterson and Pennington 2012 ; Snowling and living…” (American Psychological Association Hulme 2012 ) and that its importance has been 2013, p. 67). overstated (Swanson et al. 2003 ). Consequently, Mirroring the language of the Individuals with other advocacy and professional organizations Disabilities Education Improvement Act (IDEA include in their defi nitions additional cognitive 2004), the APA has removed the requirement of a factors that are correlates of dyslexia. discrepancy between ability and achievement for the diagnosis of specifi c learning disorders, while notably citing “Intellectual Disability, Global British Dyslexia Association Developmental Delay, uncorrected visual or auditory acuity, other mental or neurological dis- The British Dyslexia Association (BDA), for orders, psychosocial adversity, lack of profi - example, describes dyslexia as “characterised by ciency in the language of academic instruction, diffi culties with phonological processing, rapid or inadequate educational instruction” as exclu- naming, working memory, processing speed, and sionary conditions (US Department of Education the automatic development of skills that may not 2004). Though this defi nition eliminates the match up to an individual’s other cognitive abili- necessity of a signifi cant discrepancy between ties” (British Dyslexia Association 2009 ). intelligence and achievement for the diagnosis of Although this defi nition of dyslexia moves specifi c reading disability, it does not preclude its beyond the phonological processing defi cits use or usefulness as a diagnostic tool (Decker identifi ed by the IDA, it maintains two central et al. 2013 ; Swanson 2013 ). aspects of dyslexia: (a) that the disorder is caused 26 The Use of Intelligence Tests in the Diagnosis of Specifi c Reading Disability 417 by a defi ciency in certain cognitive processes and this case study, Morgan remarked that the word (b) that the reading performance is not commen- blindness experienced by his patient “follow[ed] surate with the person’s other abilities. upon no injury or illness, but [was] evidently congenital, and due most probably to defective development of that region of the brain” (Morgan Diagnosis of Specifi c Reading 1896 , p. 1378). This observation gave rise to the Disability idea of a congenital, biological basis for specifi c reading disability. To be effective, remedial instruction in reading Hinshelwood (1917 ), an ophthalmologist and must be preceded by careful diagnosis (Monroe surgeon at the Glasgow Eye Infi rmary, provided and Backus 1937 , p. 12). additional case study evidence to support the Historically, intelligence tests have been used view of specifi c reading disability as a congenital to diagnose specifi c reading disability by two pri- disorder of neurobiological origin. He described mary means: (a) the identifi cation of signifi cant a family in which six children, across two genera- discrepancies between an individual’s assessed tions, suffered from varying degrees of congeni- ability and academic performance (an ability- tal word blindness, characterized by an “inability achievement discrepancy) and (b) the detection to recognize by sight, words and letters…” of signifi cant performance variations among cog- (Hinshelwood 1917 , p. 72). Like Kussmaul nitive and academic abilities within an individual ( 1877) and Morgan (1896 ), Hinshelwood ( 1917 ) (often referred to as intraindividual variations or noted that his subjects suffered from no other a pattern of strengths and weaknesses). In this known functional defi cits and were otherwise next section, we discuss the historical anteced- typical in their development. ents and origins of these two approaches.

Origin of the Ability-Achievement Early Case Descriptions Discrepancy Procedure

The construct of reading disability long predates The practice of using information derived from recent debate regarding its defi nition. Children intelligence tests to help identify specifi c reading and adults with diffi culties consistent with read- disabilities is also well established in the histori- ing disability were fi rst described in the research cal literature. In 1925, Orton fi rst used the literature in the late nineteenth and early twenti- Stanford-Binet test of intelligence to identify sev- eth centuries. As an example, Kussmaul, a eral students whose reading abilities were well German neurologist, described an otherwise typi- below what would have been typical of children cally abled adult patient with what he character- of comparable intelligence (Hallahan and Mercer ized as word blindness and noted that word 2002 ). Less than a decade later, Monroe (1932 ) blindness “may exist although the power of sight, used a variety of achievement and intelligence the intellect, and the powers of speech are intact” tests, including the Stanford-Binet, to identify (Kussmaul 1877 , p. 595). By emphasizing the students who had discrepancies between their specifi city of the reading disability and the intact expected abilities (per the Stanford-Binet) and state of other areas of cognitive functioning, their performance on measures of reading Kussmaul gave birth to the construct of specifi c achievement. reading disability (Hallahan and Mercer 2002 ). Monroe, a psychologist and author of many of In 1896, Morgan used Kussmaul’s term (word the Dick and Jane readers, began her academic blindness) to describe the condition of an other- career as a research associate for Orton, codevel- wise typically developing 14-year-old boy who oper of the Orton-Gillingham reading method. In would have been “the smartest lad in the school if her classic book, Children Who Cannot Read , the instruction were entirely oral” (p. 94). In Monroe ( 1932) recommended the use of a reading 418 N. Mather and D. Schneider index that was designed to predict reading perfor- (p. 582). This obvious drawback notwithstanding, mance based on age and ability. The intent of this Monroe and Backus (1937 ) advised that all poor index was to quantify the concept of unexpected readers be given an intelligence test “…to deter- underachievement or reading achievement that mine whether the child has the mental capacity was signifi cantly lower than would be predicted for a higher level of reading than he is actually by chronological age, mental age, and academic achieving” (p. 21). Monroe and Backus (1937 ) accomplishments in other areas. added the qualifi cation, however, that the selected To that end, Monroe developed a reading test should not be too heavily weighted on one expectancy formula that was designed to deter- type of ability and that tests requiring consider- mine if an individual’s reading achievement was at able reading should be avoided because they a level consistent with her or his other achieve- would unduly penalize poor readers. ments. The factors used to determine reading expectancy comprised (a) chronological age, (b) mental age (based upon the Stanford-Binet test of Origin of the Intraindividual intelligence), and (c) current levels of performance Variation Procedure in arithmetic computation. This expectancy score was then compared to the average level of reading In addition to developing the concept of an achievement across several standardized assess- ability- achievement discrepancy, Monroe and ments. Describing her expectancy formula, Backus (1937 ) pioneered the use of standardized Monroe explained: “It seems that we are measur- tests of intelligence to identify patterns of ing a discrepancy between reading and other strengths and weaknesses in students’ intellectual accomplishments, which may occur in either profi les. They observed that “[a]n analysis of the direction at any intellectual level” (p. 17). Upon various sub-tests in the Stanford-Binet or other the identifi cation of such a discrepancy, Monroe intelligence tests may give an indication of the recommended that remedial reading instruction be child’s best and poorest intellectual abilities” provided until reading performance became con- (p. 23). They did not, however, go so far as to sistent with intellectual ability and academic advocate the use of intraindividual variations in accomplishments in other areas (p. 177). cognitive performance as a diagnostic tool in its Despite her endorsement of this index, Monroe own right. noted that her method was not without problems. Travis, widely recognized in the United States The most signifi cant of these was that the original as the founding father of the discipline of speech- Stanford-Binet, used to calculate intelligence for language pathology, was among the fi rst practi- the reading expectancy formula, required reading tioners to examine the unique patterns of on various subtests that were not designed primar- cognitive and academic strengths and weaknesses ily to measure reading ability. Many students particular to individuals with reading disabilities. referred to Monroe were unable to read at all at In a chapter written for the Yearbook of the the time of testing, resulting in unreliable scores National Society for Studies in Education , Travis and the likely under-prediction of intelligence and ( 1935) described children of typical intelligence reading potential (Monroe 1932 ). Monroe and her who struggled to learn to read or spell but per- colleague Backus (1937 ) observed, “Sometimes formed profi ciently in other areas. These chil- children of good general intelligence show retar- dren, whom he characterized as possessing a dation in some of the specifi c skills which com- special disability, showed striking variations in pose an intelligence test” (p. 22). This observation abilities among subjects as well as notable dis- was consistent with observations of Monroe’s crepancies between measured intelligence and mentor, Dr. Samuel Orton (1925 ), who noted that expected performance in subject-specifi c areas. “…psychometric tests as ordinarily employed Some of these children could not read, although give an entirely erroneous and unfair estimate of they could comprehend material read to them; the intellectual capacity of these children” others could read profi ciently but could not 26 The Use of Intelligence Tests in the Diagnosis of Specifi c Reading Disability 419

comprehend what they had read. Travis explained Because specifi c reading disability falls under that the clearest expression of this type of special the more generic category of specifi c learning disability was an academic profi le comprising disability, federal guidelines for SLD identifi ca- consistently low scores on tests in a given subject tion within public schools also apply to this sub- and average or above average scores on tests in category. Currently, under the Individuals with other subjects. He explained that “…in the case Disabilities Education Improvement Act (IDEA of a reading disability, a child might obtain scores 2004) (US Department of Education 2004 ), three placing him in the ninth grade in arithmetic…and methods may be used for the diagnosis of SLD: in the third grade in reading. Here we would have (a) ability-achievement discrepancy, (b) response evidence of a striking reading disability” (p. 43). to intervention, and (c) alternative research-based A key advantage cited by early advocates of procedures. In this next section, we discuss con- the intraindividual variation procedure was the temporary uses of each of these methods, as well educationally useful information provided by as some of the controversies and concerns related students’ performance on the various subtests to each. within a comprehensive assessment battery. Perhaps this idea was best expressed by Gallagher (1966 ), who noted that “[t]he information pro- Ability-Achievement Discrepancy vided by this patterning of abilities is much more important than [a] single mental age score…It The US Offi ce of Education fi rst codifi ed the use not only establishes the fact of developmental of an ability-achievement discrepancy procedure imbalance, but it locates the areas of specifi c dis- to identify and diagnose reading disabilities in ability” that, presumably, could be addressed 1976 (US Offi ce of Education 1976 ). Under the through targeted remediation (Gallagher 1966 , federal regulations of the day, “[a] specifi c learn- p. 30). The use of intraindividual variations in the ing disability [could] be found if a child [had] a diagnosis of specifi c learning disabilities, how- severe discrepancy between achievement and ever, was not without its critics. Bijou (1942 ), a intellectual ability in one or more of several contemporary of Travis, fi rst observed that cogni- areas: oral expression, written expression, listen- tive tests were not designed to diagnose disability ing comprehension or reading comprehension, through pattern analysis and might therefore be basic reading skills, mathematics calculation, invalid for such a purpose. This observation may mathematics reasoning, or spelling” (US Offi ce have been prescient, as later research (e.g., of Education 1976 , p. 52405). For three decades, Kavale and Forness 1984 ; O’Donnell 1980 ; spanning the mid-1970s to the mid-2000s, fed- Proctor and Prevatt 2003 ) has likewise called into eral law required that SLD status be determined question the existence of a unique pattern of cog- primarily by the documentation of a signifi cant nitive strengths and weaknesses that character- ability-achievement discrepancy. While the diag- izes specifi c reading disability. Nevertheless, all nosis of an ability-achievement discrepancy was historical approaches to SLD identifi cation have no longer required as of the 2004 reauthorization emphasized an analysis of the spared or intact of IDEA (legally in force as of 2006), the proce- abilities that “stand in stark contrast to the defi - dure is still widely used in American public cient abilities” (Kaufman 2008 , p, 7). schools today. An ability-achievement discrepancy is diagnosed when performance in one or more areas of Present-Day Methods achievement, per standardized testing, varies for Diagnosing SRD signifi cantly from values predicted by overall intellectual ability per standardized testing. In The determination of a disability should defi - the majority of cases in which the discrepancy nitely constitute more than a simple “exercise in model is used, diagnosis of a specifi c learning arithmetic” (Willis and Dumont 2002 , p.173). disability and concomitant provision of special 420 N. Mather and D. Schneider education services require several components. their peers (Fletcher et al. 1992 ; Kavale 2002 ; These include but are not necessarily limited to Speece 2002 ). This can result in educators (a) documentation of a statistically signifi cant waiting for students to fail before referring discrepancy, often 1.5 standard deviations or them to special education evaluation and ser- more, between one or more areas of academic vices. Delays in service provision caused by achievement and intellectual ability; (b) determi- such wait-to-fail approaches can exacerbate nation of a link between the identifi ed discrep- academic diffi culties and cause students to ancy and the poor academic performance; (c) a lose ground outside of their immediate areas team-based determination that an individual’s of disability. educational needs would be best met through the 2. Aspects of the learning disability itself, such provision of special education services; and (d) as impaired processing speed or poor working consideration of exclusionary criteria, such as memory, may disproportionately impact over- intellectual disability or English language all performance on standardized tests of intel- learner status (Restori et al. 2009 ). ligence, thus diminishing the size of the Despite its long-standing use and codifi cation discrepancy. This effect tends to compound in law, the diagnostic use of the ability- with time, as poor reading skills impact achievement discrepancy procedure has proven vocabulary development and overall knowl- to be one of the most contentious issues in the edge, decreasing overall scores and reducing fi eld of learning disabilities (Swanson 2013 ). The the size of the discrepancy. ability-achievement discrepancy procedure was 3. The size of the discrepancy required for diag- designed as one means by which to identify the nosis of SRD varies widely among school dis- existence of unexpected underachievement (i.e., tricts and states (Berninger 1996 ), resulting in achievement far below what would be predicted the uneven provision of special education ser- based on overall intelligence); it was not initially vices across the nation. intended to be used as the primary criterion upon 4. Simple discrepancy calculations fail to account which to diagnose SLD – and research has not for the variability in the origin and presenta- supported its use as such (Fletcher et al. 1992 ; tion of reading diffi culties (Catts et al. 2003 ; Fletcher et al. 1994 ). Swanson et al. 2006 ), and the identifi cation of For example, in a meta-analysis of 46 studies an ability-achievement discrepancy does little comparing groups of struggling readers with to enhance understanding of the reasons for ability-achievement discrepancies to those with- reading failure or provide guidance for the out discrepancies, Stuebing et al. (2002 ) con- development of an intervention plan. cluded that the use of an ability-achievement 5. Poor readers of varying intelligence show sim- discrepancy for the identifi cation of a reading ilar reading, spelling, language, and memory disability had weak validity. This is likely at least defi cits (Siegel 1989 ). These fi ndings call in to partially attributable to an operational mismatch question the practice of denying the diagnosis between the defi nition of specifi c learning dis- of specifi c learning disability to students with ability, which characterizes SLD as a defi cit in IQ scores well below average for their age. psychological processing, and the purpose of the Clearly, rigid adherence to formula-based ability- ability-achievement discrepancy procedure: the achievement models for the diagnosis of SRD identifi cation of unexpected underachievement. lacks empirical support and may also result in Other criticisms of the use of ability-achievement decreased attention to signifi cant qualitative and discrepancy of SLD identifi cation have focused quantitative information that can be found within on the following: intelligence tests. Furthermore, contrary to the 1. The magnitude of the discrepancy tends to theoretical assumptions underlying the ability- increase with age, and young children often achievement discrepancy procedure, intelligence lack an ability-achievement discrepancy suf- tests and achievement tests do not fall into a neat fi cient for diagnosis, even when well behind binary of abilities and achievement skills (Lohman 26 The Use of Intelligence Tests in the Diagnosis of Specifi c Reading Disability 421

2006 ). In fact, far more overlap exists between the may be attributable to a number of factors, skills and abilities measured on tests of intelli- including natural variability in skills (e.g., verbal gence and tests of achievement than many realize. abilities higher than reading abilities), strengths Lohman explained that “Most novices believe that in specifi c areas that are measured by intelligence ability is innate and achievement acquired, tests (e.g., analytic reasoning) and weaknesses in whereas experts see the two as different aspects of areas that are measured by the achievement tests the same thing” (p. 1). Assuming an absolute dis- (e.g., decoding), or a lack of appropriate instruc- tinction exists between tests of achievement and tion or remediation in reading. Nevertheless, the tests of intelligence is erroneous as they often detection of a signifi cant ability-achievement dis- measure similar skills and abilities. For example, crepancy warrants further investigation, whatever most intelligence tests have measures of oral the cause. vocabulary, whereas achievement tests often have Moreover, in some students with SRD, an similar vocabulary tests that require reading. Both ability-achievement discrepancy can be attrib- are measures of lexical knowledge, and perfor- uted to the reading disability itself. For such indi- mance would tend to be similar unless an individ- viduals, an ability-achievement discrepancy ual has poor reading skills. Anastasi (1980 ) alluded actually operationalizes the concept of unex- to this phenomenon when she lamented, “If a pected underachievement (Ferrer et al. 2010 , benevolent wizard were to give me the power to p. 10). This is particularly true for individuals of eliminate four words from the tester’s vocabulary, high intelligence whose reading lags far behind I would choose ‘intelligence,’ ‘aptitudes,’ ‘abili- their overall abilities. These students, often ties,’ and ‘achievement.’ Then if a malevolent wiz- referred to as twice exceptional, may read at a ard were suddenly to appear and demand that I level consistent with their peers, yet underachieve take back one word, I would chose to retain ‘abili- signifi cantly in relation to their academic poten- ties’” (p. 1). Anastasi referred to all abilities as tial and intelligence (Mather and Gerner 2009 ). “developed” skills that evolve through cultural Although their decoding skills are signifi cantly experiences and interests, suggesting that abilities weaker than would be expected based on their and skills are simply two sides of the same coin. overall intelligence, they rarely qualify for spe- While cognitive abilities are grounded in biologi- cial education services using methods other than cal processes, they, like the skills in which they are an ability-achievement discrepancy (Crepeau- manifested, are developed through education and Hobson and Bianco 2011 ). While the overall instruction. Consequently, all cognitive and abilities of twice exceptional students often allow achievement tests are, to a great or lesser extent, them to compensate for their poor reading skills, measures of developed abilities (Lohman 2006 ), they would nevertheless benefi t from targeted and the diagnostic utility of the ability- achievement interventions and special education services discrepancy is therefore questionable. (Peterson and Pennington 2012 ). Otherwise, they Nevertheless, the existence of a discrepancy are likely to remain “slow, dysfl uent readers” between intelligence tests and reading achieve- (Mather et al. 2013 , p. 714), hampering their abil- ment can provide important information in cer- ity to achieve on a level commensurate with their tain situations. While it is not universally the overall intelligence. case, approximately one half of students diagnosed with SLD do have signifi cant ability- achievement discrepancies (Kavale and Reese Response to Intervention 1992 ). Furthermore, Kavale et al. ( 1994 ), in a reanalysis of ability-achievement testing data, In lieu of an ability-achievement discrepancy, found that ability-achievement discrepancies can IDEA 2004 now permits schools to use a process be used to differentiate students with specifi c that determines whether or not a student responds reading disabilities from other low achievers with to evidence-based intervention as part of SLD a moderate degree of fi delity. Such discrepancies evaluation procedures. Response to intervention 422 N. Mather and D. Schneider

(RTI) is an educational approach that is primarily diagnose or discriminate between the various concerned with general, not special, education. causes of poor reading. RTI should not be viewed Two essential components of RTI are universal as a method for the identifi cation of specifi c read- screening and progress monitoring using ing disability, but as a preventative method of curriculum- based measurement (CBM) instru- early identifi cation and intervention for reading ments at specifi ed intervals throughout the aca- diffi culties of all types. demic year. Based upon their overall response to While RTI approaches have several notable evidence-based instruction and instructional benefi ts, concerns remain. As with the ability- interventions, students are assigned to one of achievement discrepancy procedure, a major three tiers of academic support: (a) Tier 1, in limitation of RTI models is that they do not pro- which high-quality core reading instruction and vide information concerning the underlying interventions are provided to all students in align- cause of poor reading achievement (Mather and ment with curricular goals; (b) Tier 2, in which Kaufman 2006 ). Furthermore, RTI data appear small group instruction and targeted interven- insuffi cient for the early, accurate identifi cation tions are provided to students who fail to make of those with specifi c reading disabilities, as satisfactory progress through Tier 1 instruction; opposed to those with other causes of reading and (c) Tier 3, in which one-on-one instruction underachievement. Recent research has sug- and targeted interventions are provided to stu- gested that the one-stage screening procedures dents who fail to respond to Tier 2 reading that are often employed in RTI models result in instruction (Jimerson et al. 2007). In many dis- unacceptably high rates of false positives (Fuchs tricts, Tier 3 is not synonymous with special edu- et al. 2012 ). However, Fuchs et al. (2012 ) found cation. This tier is designed to address the needs that use of a two-stage screening procedure that of students who continue to struggle to meet included several cognitive (e.g., rapid naming, reading goals, even when provided with small phonological processing, oral language compre- group instruction. Contrary to the traditional hension, and nonverbal reasoning) and untimed model of SLD, which is exclusive of simple low and timed reading measures greatly improved achievement or other conditions that might cause classifi cation accuracy. Therefore, comprehen- poor performance in the classroom (e.g., intellec- sive cognitive and achievement assessments tual disability or English language learner status), should be considered for those students who fail RTI Tier 3 is inclusive of all causes of low read- to respond to Tier 2 interventions or who are sus- ing achievement. pected of having a reading disability. The results In a very real sense, RTI poses a challenge to of such assessments can provide valuable infor- the basic construct of learning disability as a con- mation concerning the causes of underachieve- dition in which underachievement is unexpected ment, as well as an identifi cation of an individual’s given an individual’s overall intelligence and strengths and weaknesses that can be used to other academic accomplishments (Kavale et al. inform Tier 3 interventions or suggest the need 2005 ). Vaughn and Fuchs (2003 ) indicated that for more tailored special education services. RTI can be seen as an effort to redefi ne SLD such that the students with the greatest academic need, not just those with unexpected underachieve- Pattern of Strengths ment, are identifi ed and served. Likewise, Feifer and Weaknesses (2008 ) suggested that RTI is one means by which to address concerns associated with the so-called In addition to ability-achievement and RTI “test and place” philosophy of SLD identifi cation approaches, IDEA 2004 provides a third option and the often arbitrary numerical cutoffs of the to schools for SLD identifi cation. IDEA regula- ability-achievement discrepancy procedure. tion § 300.309 (a)(2)(ii) permits consideration of While these goals are laudable, CBM measures a pattern of strengths and weaknesses among alone do not provide suffi cient evidence to various aspects of cognitive ability and/or academic 26 The Use of Intelligence Tests in the Diagnosis of Specifi c Reading Disability 423 achievement as part of the SLD identifi cation instruments and assessments with a high degree process. The pattern of strengths and weaknesses of predictive and diagnostic validity; and (d) the (PSW) approach, also referred to as the intraindi- use of sound, empirically based decision-making vidual variations or intraindividual discrepancies strategies (Schultz et al. 2012 ). Likewise, when procedure, typically involves the administration selecting among assessments for a PSW evalua- of tests of cognitive abilities and related achieve- tion, evaluators should fi rst determine (a) the areas ment tests. Performance on these measures is of concern, (b) the age of the referred student, (c) evaluated by a qualifi ed professional for the the student’s grade in school, and (d) which cogni- presence of statistically signifi cant variations tive abilities and processes are most likely to be or discrepancies among or between areas of per- related to the student’s particular educational formance that might be relevant to the diagnosis challenges (McGrew and Wendling 2010 ). of a specifi c reading disability. As with the ability- Two of the most common approaches to the achievement procedure and the RTI approach, a clinical evaluation of patterns of strengths and determination must also be made that the under- weaknesses are the discrepancy/consistency achievement is not primarily the result of a factor approach (D/C) (Naglieri 1999 ) and the cognitive other than a learning disability. hypothesis testing (CHT) approach (Hale and A principal advantage of the PSW approach is Fiorello 2004 ). The D/C approach (Naglieri 1999 ) the educationally useful information provided by involves the analysis of cognitive processes using the assessments. According to the Standards for the planning, attention, simultaneous, and succes- Educational and Psychological Testing published sive (PASS) theory as operationalized by the fi rst by the American Educational Research Association (Naglieri and Das 1997 ) and second (Naglieri (AERA), American Psychological Association et al. 2014 ) editions of the Cognitive Assessment (APA), and National Council on Measurement in System (CAS), as well as the evaluation of obser- Education (NCME) ( 1999), the PSW approach vational data and data from achievement testing. allows a qualifi ed examiner to interpret a variety of Following data collection, CAS, achievement, skills and abilities in the context of the broader and observational data are systematically ana- assessment and other educational data, permitting lyzed to determine whether relationships exist the identifi cation of “problem[s] that may not be between weaknesses in cognitive processing and apparent if scores on different kinds of tests are weaknesses in academic achievement. The results combined” (p. 123). For example, an examiner of this analysis are then used to inform educa- might note that an individual performs poorly on tional planning and identify appropriate instruc- timed tests of reading and also has slow processing tional interventions (Schultz et al. 2012 ). speed scores relative to her or his performance in The CHT approach involves the selection of other areas. While such a fi nding might be indica- cognitive tests based upon empirical evidence that tive of a specifi c reading disability, it would also the cognitive factors to be evaluated are likely to be useful in selecting appropriate accommoda- be associated with the specifi c academic prob- tions (e.g., provision of additional time for assign- lems presented by an individual student (Schultz ment completion) and targeted remediation (e.g., et al. 2012 ). In this model, data concerning areas an intervention to improve reading speed). of academic diffi culty are typically derived from No universally agreed upon process exists for RTI, descriptions of the presenting problem(s), the evaluation of cognitive strengths and weak- and student case histories (Schultz et al. 2012 ). nesses, and disagreements abound among practi- Cognitive testing data are collected using stan- tioners. Nevertheless, the following practices are dardized instruments rooted in CHC theory (e.g., foundational to the accurate and effective clinical Woodcock-Johnson Tests of Cognitive Ability IV use of the PSW approach: (a) the use of data from (Schrank et al., 2014 )). As with the D/C approach, a wide variety of sources; (b) the use of data ana- data resulting from CHT approach assessments lytic techniques specifi cally designed for the are analyzed to identify the connections between identifi cation of relevant patterns; (c) the use of academic diffi culties and areas of cognitive 424 N. Mather and D. Schneider weaknesses. Findings are then used to target The cognitive-psychological model is a widely educational interventions to a student’s individual used approach to understanding the nature and needs, building on strengths and remediating etiology of specifi c reading disability. This model weaknesses. As an example, a student with low attempts “to explain both the consistencies and processing speed would likely benefi t from inconsistencies within performance” (p. 700) extended time, but remediation of phonics skills through psychological study, often including would be unnecessary. assessments designed to identify the cognitive Despite its previously outlined advantages, the factors and processes associated with particular PSW approach is not without limitations. The patterns of performance (Mather et al. 2013 ). A most important among these is a lack of validated number of cognitive factors and processes are diagnostic criteria or well-established guidelines associated with reading disability. Understanding for the use of the procedure. Furthermore, a spe- those factors and processes, and how weaknesses cifi c pattern of strengths and weaknesses that con- in one or more of them can contribute to reading fi rms or disconfi rms the presence of SRD has not diffi culties, is useful to both the diagnosis and been identifi ed. In general, PSW methods do not remediation of specifi c reading disability. While (a) address the fact that cognitive and achievement research does not support the assertion that there tests often share similar content, such as measures is a single cognitive profi le indicative of SRD, of vocabulary; (b) recognize that error is inherent analysis of specifi c cognitive factor scores in all measurement and may lead to an unaccept- derived from standardized cognitive batteries can ably high level of false positives; or (c) specify the provide evidence of a pattern of strengths and magnitude of the difference among abilities neces- weaknesses indicative of a reading disability, sary for the diagnosis of SRD (Stuebing et al. while providing a direct link to intervention 2012 ). Furthermore disagreements exist regarding hypotheses (Decker et al. 2013 ). In this next sec- the use of an intraindividual versus an interindi- tion, we discuss two theoretical models of cogni- vidual or normative approach. When using an tive abilities and processes, as well as several intraindividual approach, variations within an indi- specifi c cognitive factors that have been associ- vidual’s unique profi le are considered in determin- ated with reading disability. ing the existence of a reading disability, whether or not achievement falls below an absolute threshold (which would be relevant in a case of a twice Planning, Attention, Simultaneous, exceptional individual who may have superior and Successive (PASS) Theory intelligence, but reading in the low average range). When using a normative approach, however, Modern neurological science supports the concep- achievement must fall below a predetermined tion of the human brain as an organ comprised of threshold for a diagnosis of SRD to be made. differentially functioning but interdependent Regardless of which approach is taken, an evalua- regions (Das 2009 ; Das et al. 1994 ; Naglieri and tor must still use clinical judgment when diagnos- Otero 2011 ). The planning, attention, simultane- ing SRD or developing an individualized program ous, and successive (PASS) theory of intelligence of education or intervention. is heavily indebted both to modern neurological science and to Luria’s (1966 ) pioneering work on the modularization of the brain and psychological Cognitive Factors and Processes processes. PASS theory hypothesizes the existence Associated with Specifi c Reading of three major functional units of the brain: atten- Disability tion, simultaneous and successive processing, and planning. The attention function of the brain One factor alone is often not suffi cient to inhibit involves maintaining optimal levels of arousal, the act of reading; each case must be investigated focusing attention on salient stimuli, and resisting with the hopes of locating as many contributing distractions; the simultaneous and successive pro- factors as follows (Monroe and Backus 1937 ). cessing functions of the brain involve integrative 26 The Use of Intelligence Tests in the Diagnosis of Specifi c Reading Disability 425 and serial information processing, respectively; tially presented information, as evidenced by low and the planning function of the brain involve exec- successive and spelling scores, a strategy such as utive functioning and the organization and control chunking can be used to teach the student how to of behavior (Das 2002; Naglieri and Otero 2011 ). manage academic tasks that place large demands A major factor distinguishing PASS theory from on memory. Furthermore, researchers (Haddad other models of human intelligence is its emphasis et al. 2003 ; Iseman and Naglieri 2011 ; Naglieri on defi ning human abilities on the basis of cogni- and Gottling 1995 , 1997 ; Naglieri and Johnson tive processes rather than models that focus on test 2000 ) have demonstrated that the use of PASS- content (e.g., verbal, nonverbal), modality (e.g., based planning strategies can improve academic visual, auditory), or other attributes such as mem- performance. Elementary school students using ory and reasoning. the PASS Remedial Program (PREP; Das 1999 ) The cognitive processes theorized in the PASS have shown improvements in reading skills theory were fi rst operationalized in the Cognitive (Boden and Kirby 1995 ; Carlson and Das 1997 ; Assessment System (CAS) (Naglieri and Das Das et al. 1995 , 2000 ; Parrila et al. 1999 ). 1997) and more recently in the CAS-2 (Naglieri et al. 2014), two psychometric instruments designed to measure cognitive processes in each Cattell-Horn-Carroll Theory of the three functional areas described by Luria. The instrument provides scores for planning, Another prominent model of cognitive abilities is attention, simultaneous, and successive cognitive Cattell-Horn-Carroll (CHC) theory (Carroll 1993 ; processes, as well as a full scale (Naglieri and Cattell 1941 ; Horn 1985; Horn and Noll 1997 : Das 1997 ). Research suggests that the CAS is McGrew and Flanagan 1998 ). This model has well suited to the differential diagnosis of learn- proven particularly infl uential in the fi elds of edu- ing, attentional, and cognitive disabilities (Das cational psychology and psychoeducational 2002 ; Naglieri and Conway 2009 ; Naglieri et al. assessment (Konold et al. 2003 ; McGrew 2005 ) 2004 ). In fact, the CAS has shown the ability to and has been operationalized in the Woodcock- discriminate between children with attention def- Johnson Tests of Cognitive Abilities (WJ IV icit/hyperactivity disorder (ADHD), children COG) (Schrank et al., 2014 ). In CHC theory, nine with a specifi c reading disability (SRD), and chil- broad stratum abilities and over seventy narrow dren in general education with a high degree of abilities are described. The broad stratum abilities fi delity (Naglieri et al. 2004 ). In research per- include the following: formed by Naglieri et al. (2004 ), children with 1. Crystallized intelligence (Gc): previously SRD showed defi cits in successive processing acquired knowledge; the ability to retrieve, relative to their overall PASS scores, whereas apply, and communicate that knowledge; and children with ADHD obtained lower planning the ability to make reasoned judgments based and, to a lesser extent, attention scores than on upon that knowledge the simultaneous and successive scales. The cal- 2. Fluid intelligence (Gf): the ability to reason, culated effect sizes for the differences in perfor- form judgments, acquire novel concepts, and mance among the groups ranged from moderate solve problems by applying novel concepts or to large (Naglieri et al. 2004 ), suggesting that the procedures CAS can serve as a valuable tool in the diagnosis 3. Quantitative reasoning (Gq): the ability to of specifi c reading disabilities. understand and acquire quantitative concepts, PASS theory has also been linked to academic engage in quantitative reasoning, identify interventions. The book Helping Children Learn numerical and quantitative relationships, and (Naglieri and Pickering 2010 ) provides cognitive solve numerical and quantitative problems strategies that address specifi c PASS weaknesses 4. Reading and writing ability (Grw): the abili- related to academic problems. For example, if a ties to perform tasks requiring word reading student is having problems working with sequen- and spelling 426 N. Mather and D. Schneider

5. Short-term working memory (Gwm): the abil- et al. 1999 . Discrepant performance among ity to hold and manipulate novel information cognitive abilities can provide important informa- in one’s immediate consciousness tion about which areas are most likely to require 6. Long-term storage and retrieval (Glr): the remediation and which interventions are most ability to store and fl uently retrieve previously likely to be benefi cial for an individual student. learned associations 7. Visual processing (Gv): the ability to identify, store, recall, understand, and analyze visual Phonological Awareness patterns and representations 8. Auditory processing (Ga): the ability to iden- Phonological awareness comprises the ability to tify, understand, and discriminate sounds, perceive and identify individual speech sounds as including speech sounds well as the ability to perceive and understand the 9. Processing speed (Gs): the ability to perform phonological structure of spoken words. These cognitive tasks with speed, fl uency, and abilities are subsumed under auditory processing automaticity (Ga) in the CHC model. While the capacity to pro- Several of the broad and narrow cognitive duce oral language is inherent, reading and writing, abilities described in CHC theory have particular and the skills upon which they are dependent, must relevance to the identifi cation and diagnosis of be learned. In order to read and write, an individual specifi c reading disability. These include the must develop phonological awareness and learn to broad domain of crystallized knowledge (Gc) and distinguish among the discrete sounds of human three of its narrow component abilities: language speech, fi nally mapping those sounds onto the development, lexical knowledge, and listening symbols with which they are encoded in written ability (McGrew and Wendling 2010 ). Memory language. Furthermore, in order to read and write span and working memory, inclusive of induction effi ciently, this process must occur with automatic- and general sequential reasoning, are also corre- ity and ease. Phonological awareness is positively lated to reading performance, as are phonological correlated to reading ability (Ehri et al. 2001 ), and awareness (subsumed under Ga) in processing a signifi cant majority of individuals with specifi c speed (Gs) and rapid automatized naming (sub- reading disabilities perform poorly on these mea- sumed under Glr) (McGrew and Wendling 2010 ). sures (Catts and Kamhi 2005 ; Wagner and Torgesen Therefore, an evaluator should consider the selec- 1987 ). Defi cits in the areas of phonological aware- tion of tests designed to measure these key abilities ness are major indicators of specifi c reading disor- when evaluating an individual for the presence or der, particularly in cases in which other cognitive absence of a specifi c reading disability. abilities are intact or largely intact. Individuals who are diagnosed with signifi cant weaknesses in phonological awareness should be offered interven- Cognitive Abilities Related tions designed to improve skills in these areas, to Specifi c Reading Disability followed by, or in concert with, systematic instruction in phoneme-grapheme relationships. Individual variations in cognitive abilities have been shown to be important mediators of academic outcomes (Swanson and Hoskyn 1998). Rapid Automatized Naming Both children and adults with identifi ed reading disabilities have cognitive defi cits that are perva- Tests of rapid automatized naming (RAN) mea- sive across the life span. These include poor pho- sure the rapid retrieval of continuously presented, nological awareness (Wagner and Torgesen 1987 ), highly familiar symbols of a single type (e.g., slow naming speed (Bowers and Wolf 1993 ; Wolf objects, colors, letters, or numbers). Tests of rapid and Bowers 1999 ), poor verbal memory (Swanson alternating stimuli (RAS), a measure closely 2013 ), and slow processing speed (Feldmann related to RAN, involve the rapid retrieval of a et al. 2004; Kail 1991; Kail and Hall 1994; Kail combination of these common symbol types. 26 The Use of Intelligence Tests in the Diagnosis of Specifi c Reading Disability 427

Although the cognitive abilities underlying perfor- spoken words; storing and processing letters, mance on RAN tests are not fully understood, such letters in written words, and written words; nam- tasks appear to capture a unique cognitive ability ing orthographic symbols; and integrating the or set of abilities. Performance on RAN tasks is visual appearance of written words with internally highly correlated with reading ability (Powell stored representations (Berninger and Swanson et al. 2007 ; Wolf and Bowers 1999 ), and poor 2013). They may also benefi t from explicit instruc- RAN performance is a cognitive marker for spe- tion in memorization strategies such as chunking, cifi c reading disability (Georgiou and Parrila repetition, rehearsal, and the use of mnemonic 2013 ) that predicts reading ability independent of devices, as well as the use of memory aids, includ- other cognitive abilities, such as phonemic aware- ing planners, organizers, and note cards. ness (Powell et al. 2007 ). Measures of RAN capture a cognitive skill or set of skills whose importance increases with age, most likely because Processing Speed of increased demands on visual word recognition and automaticity (Vaessen and Blomert 2013 ). Processing speed, or Gs in the CHC model, com- These measures appear to be more related to facil- prises the rapidity, fl uency, and effi ciency with ity in reading irregular words than to phonic skills which an individual performs cognitive tasks. (Abu-Hamour et al. 2012 ). Moreover, RAN tasks This cognitive ability appears to be another have been shown to reliably predict future reading important mediator and correlate of reading disability in preliterate children (Furnes and ability in both children and adults (Evans et al. Samuelsson 2011 ; Lervåg and Hulme 2009 ), mak- 2002 ; Gregg et al. 2006 ; Kail 1991 ; Kail et al. ing RAN a useful screening tool for the early iden- 1999 ; Konold et al. 2003; Shanahan et al. 2006 ; tifi cation of specifi c reading disability. Willcutt et al. 2005). Even among individuals who have strong decoding skills, poor processing speed has been shown to hamper reading effi - Working Memory ciency and fl uency (Jacobson et al. 2011 ). Poor performance on measures of processing speed is In addition to phonologically related processes, associated with reduced reading rate and diffi - memory-related processes also appear to contrib- culty in fl uently decoding words with irregular ute to specifi c reading disability (Crews and spelling patterns (e.g., yacht). Furthermore, D’Amato 2009 ; Jacobson et al. 2011 ; Swanson impaired processing speed may hamper the abil- et al. 2009). Working memory comprises the ability of students to develop fl uent oral reading ity to hold and manipulate newly acquired infor- (Jacobson et al. 2011 ). When compounded with mation and concepts in immediate consciousness. defi cits in phonological awareness and/or work- It is relevant to every aspect of learning and areas ing memory, defi cits in processing speed can fur- of academic achievement, including decoding, ther impair reading ability. Individuals with such encoding, and reading comprehension. In a meta- defi cits often require interventions and strategies analysis of the results of 43 studies conducted designed to build reading speed, as well as addi- between 1963 and 2006, Swanson et al. (2009 ) tional time for assignments requiring reading. found that children with reading disabilities performed poorly on tasks that required simultaneous processing of information, nonword repetition, as Multiple Defi cit Model well as the immediate recall of letter, number, and word strings. Poor performance in such tasks is Some abilities measured by comprehensive cog- associated with defi cits in working memory, nitive and achievement batteries are highly cor- particularly in the area of verbal short-term related; this is particularly the case in the various memory. Students diagnosed with signifi cant defi - abilities that comprise the capacity to read and cits in working memory may require explicit write with fl uency and accuracy. The understanding instruction in storing and processing sounds in that many learning disorders are multifactorial in 428 N. Mather and D. Schneider origin and etiology is called the multiple defi cit history of core symptoms, such as poor phono- model (Pennington 2006). As an example, Konold logical awareness, inaccurate and slow oral read- et al. (2003 ) assessed how children performed on ing, and poor spelling (Shaywitz. and Shaywitz measures of phonological awareness, vocabulary 2013 ). While the SRD evaluation process requires and listening comprehension, processing speed, the consideration of information from a variety of and short-term memory. Among children with the sources, including academic records, achieve- strongest overall reading ability, performance ment testing, and classroom observation, cogni- tended to be high across all four measures; among tive testing remains fundamental to informed and children with the poorest overall reading ability, accurate SRD identifi cation. Cognitive testing is however, performance tended to be low across all essential to differential diagnosis, distinguishing four measures. There is also substantial evidence among multiple possible causes of reading under- that reading disability, slow processing speed, and achievement. Furthermore, when used by skillful attention-defi cit/hyperactivity disorder (ADHD), a clinicians, cognitive testing is not only useful and disorder in which a high degree of distractibility effective for diagnostic purposes but provides and impulsivity is characteristic, have a high degree information about individual needs that can be of comorbidity (McGrath et al. 2011 ; Willcutt et al. used to inform educational decisions and guide 2001 ; Willcutt et al. 2010 ). Furthermore, a mount- interventions (Decker et al. 2013 ). ing body of research suggests that speech and lan- Sometimes critics of cognitive ability testing guage disorders, phonological processing defi cits, claim that the results from intelligence tests are and specifi c reading disabilities co-occur at levels not useful for instructional planning. Often, how- far higher than can be accounted for by chance ever, the fi ndings from these tests lead directly to (Willcutt et al. 2010 ). specifi c alterations in a program. As an example, Nevertheless, some of the abilities measured it would be very important for a teacher to know on intelligence tests are not highly correlated. For that a student had high verbal abilities but low example, in homogenous samples of young reading skills. For this student oral evaluations adults, Horn and Blankson (2012 ) reported, “… would be the least biased, and the most accurate, measures in which there is much emphasis on way of assessing his or her actual knowledge and speediness correlate near zero, perhaps nega- progress. If, on the other hand, a student had low tively, with tests that require solving diffi cult verbal abilities as well as reading skills, he or she problems” (p. 91). Thus, it is not uncommon for will likely need a modifi cation in the diffi culty an individual with a specifi c reading disability to level of evaluations and reading assignments. have high scores on measures of problem solving Another student with adequate verbal abilities and reasoning but low scores on speeded tests. and good reasoning may have a signifi cant weak- Such variations within individual profi les should ness in memory. This student may need specifi c not be seen as diagnostic of reading disability in compensatory aids to address this weakness, such themselves, but should be considered in relation as open book/open note testing and/or take-home to their relevance to specifi c areas of academic exams. Another student with good memory but diffi culty. A sound, empirically based decision- slow processing speed may have diffi culty with making procedure should then be used to deter- completing clerical operations and more detailed mine whether or not special education services procedures, such as note taking. An important would be likely to benefi t the child. accommodation for this student may be a note taker or a complete set of instructor notes. Historically, school psychologists have over- The Future Use of Intelligence Tests relied on full scale intelligence test scores and arbitrary cutoff points for making special educa- Specifi c reading disability, as with all other learn- tion eligibility decisions. This approach lacks ing disabilities, cannot be diagnosed with a single psychometric and social validity and has conse- test; it is a clinical diagnosis that requires a quently fallen out of favor in recent years. 26 The Use of Intelligence Tests in the Diagnosis of Specifi c Reading Disability 429

Likewise, the future of intelligence testing is explained: “If these tests will give us a basis unlikely to involve reliance on the ability- from which we can start to understand a child’s achievement discrepancy procedure for SRD diffi culties, they will have justifi ed the time identifi cation. While this procedure may have spent on them. Anything that helps educators or some application, as with the identifi cation of parents to understand any phase of development twice exceptional individuals, it is neither suffi - or lack of development is of immeasurable ciently reliable nor valid to warrant continued value” (p. 189). More recently, Carroll ( 1993 ) widespread use. observed that for several thousand years, “it has While some have looked to RTI to replace the been recognized that there are individual differ- ability-achievement discrepancy procedure in ences in cognitive abilities, and that these differ- the identifi cation of children with SRD, the RTI ences have something to do with the roles and process is not diagnostic and does not differenti- behaviors of individuals in society” (Carroll ate among the various causes of reading under- 1993 , p. 25). Clearly, the future use of intelli- achievement (Mather and Kaufman 2006 ). gence tests lies in increasing our understanding Furthermore, it provides little educationally use- of how various cognitive processes predict and ful information concerning the specifi c cognitive contribute to SRD, so students can receive weakness underlying any specifi c case of timely, individualized, interventions. SRD. Consequently, cognitive testing is likely to remain an important tool in the identifi cation and Acknowledgments We would like to express our grati- diagnosis of specifi c reading disability for the tude to Dr. Michael E. Gerner and Dr. Tobias Schneider- foreseeable future. In the absence of these tests, Richardson for their helpful comments. it is diffi cult to distinguish between the diverse causes of reading disability or to select interventions that are targeted to the specifi c strengths References and weaknesses of individual students. Abu-Hamour, B., Urso, A., & Mather, N. (2012). The rela- In the future, the primary role of intelligence tionships among cognitive correlates and irregular testing will likely lie in the focused use of cogni- word, non-word, and word reading. International tive assessment to detect reading disabilities Journal of Special Education, 27 (1), 144–159. (Fiorello and Primerano 2005) with the aim of American Educational Research Association (AERA), American Psychological Association (APA), & understanding the child and identifying specifi c National Council on Measurement in Education cognitive strengths and weaknesses (Kaufman (NCME). (1999). Standards for educational and psy- 2006 ). The information derived from these chological testing . Washington, DC: AERA. assessments will be used both for identifi cation American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). and diagnostic purposes, as well as for informing Washington, DC: American Psychiatric Publishing. educational decisions and guiding interventions. American Psychiatric Association. (2013). Diagnostic In fact, an understanding of an individual’s com- and statistical manual of mental disorders (5th ed.). prehensive cognitive profi le is key to interpreta- Arlington: American Psychiatric Publishing. Anastasi, A. (1980). Abilities and the measurement of tion and effective remediation (Kaufman et al. achievement. In W. B. Schrader (Ed.), Measuring 2005 ). As observed by Kaufman (2006 ): “They achievement: Progress over a decade: Proceedings of aren’t intelligence tests anymore; they are mea- the 1979 ETS invitational conference (pp. 1–10). San sures of cognitive processing.” This observation Francisco: Jossey Bass. Berninger, V. W. (1996). Reading and writing acquisition: not only predicts the future of cognitive testing A developmental neuropsychological perspective . but also takes us back to the past, prior to the use Boulder: Westview Press. of an ability-achievement discrepancy, when the Berninger, V., & Swanson, H. L. (2013). Diagnosing and major purpose of cognitive assessments was to treating specifi c learning disabilities in reference to the brain’s working memory system. In H. L. Swanson, increase understanding of an individual’s pre- K. R. Harris, & S. Graham (Eds.), Handbook of learn- senting problem, as well as his or her specifi c ing disabilities (2nd ed., pp. 169–185). New York: abilities. In 1937, Stanger and Donohue Guilford Press. 430 N. Mather and D. Schneider

Bijou, S. W. (1942). The psychometric pattern approach from the National Reading Panel’s meta-analysis. as an aid to clinical analysis—a review. American Reading Research Quarterly, 36 , 250–287. Journal of Mental Defi ciency, 46 , 354–862. Evans, J. J., Floyd, R. G., McGrew, K. S., & Leforgee, Boden, C., & Kirby, J. R. (1995). Successive processing, M. H. (2002). The relations between measures of phonological coding and the remediation of reading. Cattell-Horn-Carroll (CHC) cognitive abilities and Journal of Cognitive Education, 4 , 19–31. reading achievement during childhood and adoles- Bowers, P. G., & Wolf, M. (1993). Theoretical links cence. School Psychology Review, 31 , 246–262. between naming speed, precise timing mechanisms Feifer, S. G. (2008). Integrating response to intervention and orthographic skill in dyslexia. Reading and (RTI) with neuropsychology: A scientifi c approach to Writing: An Interdisciplinary Journal, 5 , 69–85. reading. Psychology in the Schools, 45 , 812–825. British Dyslexlia Association. (2009). Dyslexia research Feldmann, G., Kelly, R., & Diehl, V. (2004). An interpre- information . Retrieved from, http://www.bdadyslexia. tative analysis of fi ve commonly used processing org.uk/about-dyslexia/further-information/dyslexia- speed measures. Journal of Psychoeducational research-information-.html Assessment, 22 , 151–163. Carlson, J., & Das, J. P. (1997). A process approach to Ferrer, E., Shaywitz, B., Holahan, J., Marchione, K., & remediating word decoding defi ciencies in Chapter 1 Shaywitz, S. (2010). Uncoupling of reading and IQ children. Learning Disabilities Quarterly, 20 , 93–102. over time: Empirical evidence for a defi nition of dys- Carroll, J. B. (1993). Human cognitive abilities: A survey lexia. Psychological Science, 21 , 93–101. of factor-analytic studies . Cambridge, MA: Cambridge Fiorello, C. A., & Primerano, D. (2005). Research into University Press. practice: Cattell-Horn-Carroll cognitive assessment in Cattell, R. B. (1941). Some theoretical issues in adult practice: Eligibility and program development issues. intelligence testing. Psychological Bulletin, 38 , 592. Psychology in the Schools, 42 , 525–536. Catts, H. W., Hogan, T. P., & Fey, M. E. (2003). Fletcher, J. M., Francis, D. J., Rourke, B. P., Shaywitz, Subgrouping poor readers on the basis of individual S. E., & Shaywitz, B. A. (1992). The validity of differences in reading-related abilities. Journal of discrepancy- based defi nitions of reading disabilities. Learning Disabilities, 36 (2), 151–164. Journal of Learning Disabilities, 25 , 555–561. Catts, H. W., & Kamhi, A. G. (Eds.). (2005). The connec- Fletcher, J. M., Shaywitz, S. E., Shankweiler, D. P., Katz, tions between language and reading disabilities . L., Liberman, I. Y., Stuebing, K. K., … & Shaywitz, Mahwah: Lawrence Erlbaum. B. A. (1994). Cognitive profi les of reading disability: Crepeau-Hobson, F., & Bianco, M. (2011). Identifi cation Comparisons of discrepancy and low achievement of gifted students with learning disabilities in a defi nitions. Journal of Educational Psychology , 86 , response-to-intervention era. Psychology in the 6–23. Schools, 48 , 102–109. Fuchs, D., Fuchs, L. S., & Compton, D. (2012). Smart Crews, K., & D’Amato, R. C. (2009). Subtyping chil- RTI: A next-generation approach to multilevel preven- dren’s reading disabilities using a comprehensive neu- tion. Exceptional Children, 78 , 263–279. ropsychological measure. International Journal of Furnes, B., & Samuelsson, S. (2011). Phonological aware- Neuroscience, 119 , 1615–1639. ness and rapid automatized naming predicting early Das, J. P. (1999). PASS Reading enhancement program . development in reading and spelling: Results from a Deal: Sarka Educational Resources. cross-linguistic longitudinal study. Learning and Das, J. P. (2002). A better look at intelligence. Current Individual Differences, 21 (1), 85–95. Directions in Psychological Science, 11 (1), 28–33. Gallagher, J. J. (1966). Children with developmental Das, J. P. (2009). Reading diffi culties and dyslexia: An imbalances: A psychoeducational defi nition. In W. M. interpretation for teachers. New Delhi: SAGE Cruickshank (Ed.), The teacher of brain-injured chil- International Publishers. dren (pp. 23–43). New York: Syracuse University Das, J. P., Naglieri, J. A., & Kirby, J. R. (1994). Assessment Press. of cognitive processes: The PASS theory of intelli- Georgiou, G. K., & Parrila, R. (2013). Rapid automatized gence . Needham Heights: Allyn & Bacon. naming and reading. In H. L. Swanson, K. R. Harris, Das, J. P., Mishra, R. K., & Pool, J. E. (1995). An experi- & S. Graham (Eds.), Handbook of learning disabilities ment on cognitive remediation or word-reading diffi - (2nd ed., pp. 169–185). New York: Guilford Press. culty. Journal of Learning Disabilities, 28 , 66–79. Gregg, N., Coleman, C., Davis, M., Lindstrom, W., & Das, J. P., Parrila, R., & Papadopoulos, R. (2000). Hartwig, J. (2006). Critical issues for the diagnosis of Cognitive education and reading disability. In learning disabilities in the adult population. A. Kozulin & Y. Rand (Eds.), Experiences of mediated Psychology in the Schools, 43 , 889–899. learning (pp. 274–291). Amsterdam: Pergamon Press. Haddad, F. A., Garcia, Y. E., Naglieri, J. A., Grimditch, Decker, S. L., Hale, J. B., & Flanagan, D. P. (2013). M., McAndrews, A., & Eubanks, J. (2003). Planning Professional practice issues in the assessment of cog- facilitation and reading comprehension: Instructional nitive functioning for educational applications. relevance of the PASS theory. Journal of Psychology in the Schools, 50 , 300–313. Psychoeducational Assessment, 21 , 282–289. Ehri, L., Nunes, S., Willows, D., Schuster, B., Yaghoub- Hale, J. B., & Fiorello, C. A. (2004). School neuropsy- Zadeh, Z., & Shanahan, T. (2001). Phonemic aware- chology: A practitioner’s handbook . New York: ness instruction helps children learn to read: Evidence Guilford Press. 26 The Use of Intelligence Tests in the Diagnosis of Specifi c Reading Disability 431

Hallahan, D. P., & Mercer, C. D. (2002). Learning dis- Kavale, K.A. (2002). Discrepancy models in the identifi - abilities: Historical perspectives. In R. Bradley, cation of learning disability . In R. Bradley, L. L. Danielson, & D. P. Hallahan (Eds.), Identifi cation of Danielson, & D.P. Hallahan (Eds.), Identifi cation of learning disabilities: Research to practice (pp. 1–67). learning disabilities: Research to practice (pp. 369– Mahwah: Lawrence Erlbaum Associates. 426). Mahwah, NJ: Erlbaum. Hinshelwood, J. (1917). Congenital word-blindness . Kavale, K. A., & Forness, S. R. (1984). A meta-analysis of London: Lewis. the validity of Wechsler Scale profi les and recategori- Horn, J. L. (1985). Remodeling old models of intelli- zations: Patterns or parodies? Learning Disability gence. In B. B. Wolman (Ed.), Handbook of intelli- Quarterly, 7 , 136–156. gence: Theories, measurements, and applications Kavale, K. A., & Reese, J. H. (1992). The character of (pp. 267–300). New York: Wiley. learning disabilities: An Iowa profi le. Learning Horn, J. L., & Blankson, A. N. (2012). Foundations for Disability Quarterly, 15 , 74–94. better understanding of cognitive abilities. In Kavale, K. A., Fuchs, D., & Scruggs, T. E. (1994). Setting D. Flanagan & P. Harrison (Eds.), Contemporary the record straight on learning disabilities and low intellectual assessment: Theories, tests, and issues achievement. Learning Disabilities Research & (3rd ed., pp. 73–98). New York: Guilford. Practice, 9 , 70–77. Horn, J. L., & Noll, J. (1997). Human cognitive capabili- Kavale, K. A., Holdnack, J. A., & Mostert, M. P. (2005). ties: Gf-Gc theory. In D. P. Flanagan, J. L. Genshaft, & Responsiveness to intervention and the identifi ca- P. L. Harrison (Eds.), Contemporary intellectual tion of specifi c learning disability: A critique and assessment: Theories, tests, and issues (pp. 53–91). alternative proposal. Learning Disability Quarterly, New York: Guilford Press. 28, 2–16. International Dyslexia Association (2007). Frequently Klassen, R. M., Neufeld, P., & Munro, F. (2005). When IQ asked questions: What is dyslexia? Retrieved from is irrelevant to the defi nition of learning disabilities. http://www.interdys.org/FAQWhatIs.htm . School Psychology International, 26 , 297–316. Iseman, J. S., & Naglieri, J. A. (2011). A cognitive strategy Konold, T. R., Juel, C., McKinnon, M., & Deffes, R. instruction to improve math calculation for children (2003). A multivariate model of early reading acquisi- with ADHD and LD: A randomized controlled study. tion. Applied Psycholinguistics, 24 , 89–112. Journal of Learning Disabilities, 44 , 184–195. Kussmaul, A. (1877). Disturbances of speech (trans: Jacobson, L. A., Ryan, M., Martin, R. B., Ewen, J., McCreery, J. A.). In H. von Ziemssen (Ed.), Mostofsky, S. H., Denckla, M. B., & Mahone, E. M. Cyclopedia of the practice of medicine (p. 595). (2011). Working memory infl uences processing speed New York: William Wood. and reading fl uency in ADHD. Child Neuropsychology, Lervåg, A., & Hulme, C. (2009). Rapid automatized nam- 17 , 209–224. ing (RAN) taps a mechanism that places constraints on Jimerson, S. R., Burns, M. K., & VanDerHeyden, A. M. the development of early reading fl uency. (Eds.). (2007). Handbook of response to intervention . Psychological Science, 20 , 1040–1048. London: Springer. Lohman, D. F. (2006). Beliefs about differences between Kail, R. (1991). Developmental change in speed of pro- ability and accomplishment: From folk theories to cessing during childhood and adolescence. cognitive science. Roeper Review, 29 , 32–40. Psychological Bulletin, 109 , 490–501. Luria, A. R. (1966). Human brain and psychological pro- Kail, R., & Hall, L. K. (1994). Processing speed, naming cesses. New York: Harper & Row. speed, and reading. Developmental Psychology, 30 , Lyon, G. R., Shaywitz, S. E., & Shaywitz, B. A. (2003). 949–954. Defi ning dyslexia, comorbidity, teachers’ knowledge Kail, R., Hall, L. K., & Caskey, B. J. (1999). Processing of language and reading: A defi nition of dyslexia. speed, exposure to print, and naming speed. Applied Annals of Dyslexia, 53 , 1–14. Psycholinguistics, 20 , 303–314. Mather, N., & Gerner, M. (2009). Postsecondary students Kaufman, A. S. (2006). From David Wechsler to the new with high abilities and reading disabilities: Case anal- IDEA guidelines: 35 years in the eye of the IQ storm . yses and commentary. Learning Disabilities: A Anaheim: National Association of School Multidisciplinary Journal, 15 , 121–130. Psychologists Annual Convention. Mather, N., & Kaufman, N. L. (2006). Introduction to the Kaufman, A. S. (2008). Neuropsychology and specifi c special issue, part one: It’s about the what, how well, learning disabilities: Lessons from the past as a guide and the why. Psychology in the Schools, 43 , 747–752. to present controversies and future clinical practice. In doi: 10.1002/pits.20185 . E. Fletcher-Janzen & C. R. Reynolds (Eds.), Mather, N., Shaywitz, B. A., & Shaywitz, S. E. (2013). Neuropsychological perspectives on learning disabili- Toward a synthesis of cognitive-psychological, ties in the era of RTI: Recommendations for diagnosis medical-neurobiological, and educational models for and intervention (pp. 1–13). Hoboken: Wiley. the diagnosis and management of dyslexia. In D. H. Kaufman, A. S., Lichtenberger, E. O., Fletcher-Janzen, E., Saklofske, C. R. Reynolds, & V. L. Schwean (Eds.), & Kaufman, N. L. (2005). Essentials of KABC-II The Oxford handbook of child psychological assess- Assessment . New York: Wiley. ment (pp. 698–721). New York: Oxford University Press. 432 N. Mather and D. Schneider

McGrath, L. M., Pennington, B. F., Shanahan, M. A., Naglieri, J. A., Das, J. P. & Goldstein, S. (2014). Cognitive Santerre-Lemmon, L. E., Barnard, H. D., Willcutt, Assessment System – Second Edition. Austin: E. G., … & Olson, R. K. (2011). A multiple defi cit PRO-ED. model of reading disability and attention-defi cit/ O'Donnell, L. E. (1980). Intra-individual discrepancy in hyperactivity disorder: Searching for shared cognitive diagnosing specifi c learning disabilities. Learning defi cits. Journal of Child Psychology and Psychiatry , Disability Quarterly, 3 , 10–18. 52 , 547–557. Orton, S. T. (1925). Word-blindness in school children. McGrew, K. S. (2005). The Cattell-Horn-Carroll theory of Archives of Neurology and Psychiatry, 14 , 581–615. cognitive abilities: Past, present, and future. In D. P. Parrila, R. K., Das, J. P., Kendrick, M., Papadopoulos, T., Flanagan, J. L. Genshaft, & P. L. Harrison (Eds.), & Kirby, J. (1999). Effi cacy of a cognitive reading Contemporary intellectual assessment: Theories, remediation program for at-risk children in Grade 1. tests, and issues (pp. 136–182). New York: Guilford. Developmental Disabilities Bulletin, 27 , 1–31. McGrew, K., & Flanagan, D. (1998). The intelligence test Pennington, B. F. (2006). From single to multiple defi cit desk reference: Gf-Gc cross-battery assessment . models of developmental disorders. Cognition, 101 , Needham Heights: Allyn & Bacon. 385–413. McGrew, K. S., & Wendling, B. J. (2010). Cattell–Horn– Peterson, R. L., & Pennington, B. F. (2012). Developmental Carroll cognitive-achievement relations: What we dyslexia. The Lancet, 379 (9830), 1997–2007. have learned from the past 20 years of research. Powell, D., Stainthorp, R., Stuart, M., Garwood, H., & Psychology in the Schools, 47 , 651–675. Quinlan, P. (2007). An experimental comparison Monroe, M. (1932). Children who cannot read . Chicago: between rival theories of rapid automatized naming University of Chicago Press. performance and its relationship to reading. Journal of Monroe, M., & Backus, B. (1937). Remedial reading . Experimental Child Psychology, 98 (1), 46–68. Boston: Houghton Miffl in. Proctor, B., & Prevatt, F. (2003). Agreement among four Morgan, W. P. (1896). Word blindness. British Medical models used for diagnosing learning disabilities. Journal, 1378 , 98. Journal of Learning Disabilities, 36 , 459–466. Naglieri, J. A. (1999). How valid is the PASS theory and Restori, A. F., Katz, G. S., & Lee, H. B. (2009). A critique CAS? School Psychology Review, 28 , 145–162. of the IQ/Achievement discrepancy model for identi- Naglieri, J. A., & Das, J. P. (1997). Cognitive Assessment fying specifi c learning disabilities. Europe’s Journal System. Administration and scoring manual. of Psychology, 5 (4), 128–145. Interpretive handbook . Austin: PRO-ED. Schultz, E. K., Simpson, C. G., & Lynch, S. (2012). Naglieri, J. A., & Gottling, S. H. (1995). A study of plan- Specifi c learning disability identifi cation: What consti- ning and mathematics instruction for students with tutes a pattern of strengths and weaknesses? Learning learning disabilities. Psychological Reports, 76 , Disabilities: A Multidisciplinary Journal, 18 , 87–97. 1343–1354. Schrank, F. A., McGrew, K. S., & Mather, N. (2014). Naglieri, J. A., & Gottling, S. H. (1997). The PASS theory Woodcock-Johnson IV Tests of Cognitive Abilities . and mathematics instruction: A summary of initial Rolling Meadows, IL: Riverside Publishing. studies. Journal of Cognitive Education, 5 , 209–216. Shanahan, M. A., Pennington, B. F., Yerys, B. E., Scott, Naglieri, J. A., & Johnson, D. (2000). Effectiveness of a A., Boada, R., Willcutt, E. G., & DeFries, J. C. (2006). cognitive strategy intervention to improve math calcu- Processing speed defi cits in attention defi cit/hyperac- lation based on the PASS theory. Journal of Learning tivity disorder and reading disability. Journal of Disabilities, 33 , 591–597. Abnormal Child Psychology, 34 , 585–602. Naglieri, J. A., & Conway, C. (2009). The Cognitive Shaywitz, S. E., & Shaywitz, B. A. (2013). Making a hid- Assessment System. In J. A. den disability visible: What has been learned from the Naglieri, J. A., & S. Goldstein (2009). A practitioner’s neurobiological studies of dyslexia. In H. L. Swanson, guide to assessment of intelligence and achievement K. R. Harris, & S. Graham (Eds.), Handbook of learn- (pp. 3–10). New York: Wiley. ing disabilities (2nd ed., pp. 643–657). New York: Naglieri, J. A., & Pickering, E. B. (2010). Helping chil- Guilford Press. dren learn: Intervention handouts for use at school Siegel, L. S. (1989). IQ is irrelevant to the defi nition of and home (2nd ed.). Baltimore: Paul H. Brookes learning disabilities. Journal of Learning Disabilities, Publishing. 22 , 469–486. Naglieri, J. A., Salter, C. J., & Edwards, G. H. (2004). Siegel, L. S., & Mazabel, S. (2013). Basic cognitive pro- Assessment of children with attention and reading dif- cesses and reading disabilities. In H. L. Swanson, fi culties using the PASS theory and Cognitive K. R. Harris, & S. Graham (Eds.), Handbook of learn- Assessment System. Journal of Psychoeducational ing disabilities (2nd ed., pp. 186–213). New York: Assessment, 22 , 93–105. Guilford Press. Naglieri, J. A. & Otero, T. (2011). Cognitive Assessment Snowling, M. J., & Hulme, C. (2012). Annual research System: Redefi ning intelligence from a neuropsycho- review: The nature and classifi cation of reading disor- logical perspective. In A. Davis (Ed.). Handbook of ders—a commentary for proposals on DSM-5. Journal pediatric neuropsychology (pp. 320–333). New York: of Child Psychology and Psychiatry, 53 , 593–607. Springer Publishing. 26 The Use of Intelligence Tests in the Diagnosis of Specifi c Reading Disability 433

Speece, D. (2002). Classifi cation of learning disabilities: nosis (pp. 37–47). Bloomington: Public School Convergence, expansion, and caution. In R. Bradley, Publishing Company. L. Danielson, & D. Hallahan (Eds.), Identifi cation of U. S. Department of Education. (2004). Individuals with learning disabilities: Research to practice (pp. 467– disabilities education improvement act, 20 U.S.C. § 519). Mahwah: Erlbaum. 1400 . Washington, DC: Author. Stanger, M. A., & Donohue, E. K. (1937). Prediction and U.S. Offi ce of Education. (1976). Proposed rulemaking. prevention of reading diffi culties. New York: Oxford Federal Register, 41 (230), 52404–52407. University Press. Vaessen, A., & Blomert, L. (2013). The cognitive linkage Stuebing, K. K., Fletcher, J. M., LeDoux, J. M., Lyon, and divergence of spelling and reading development. G. R., Shaywitz, S. E., & Shaywitz, B. A. (2002). Scientifi c Studies of Reading, 17 , 89–107. Validity of IQ-discrepancy classifi cations of reading Vaughn, S., & Fuchs, L. S. (2003). Redefi ning learning disabilities: A meta-analysis. American Educational disabilities as inadequate response to instruction: The Research Journal, 39 , 469–518. promise and potential problems. Learning Disabilities Stuebing, K. K., Fletcher, J. M., Branum-Martin, L., & Research & Practice, 18 , 137–146. Francis, D. J. (2012). Evaluation of the technical ade- Wagner, R. K., & Torgesen, J. K. (1987). The nature of quacy of three methods for identifying specifi c learn- phonological processing and its causal role in the ing disabilities based on cognitive discrepancies. acquisition of reading skills. Psychological Bulletin, School Psychology Review, 41 , 3–22. 101 , 192–212. Swanson, H. L. (2013). Meta-analysis of research on Willcutt, E. G., Pennington, B. F., Boada, R., Ogline, J. S., children with learning disabilities. In H. L. Swanson, Tunick, R. A., Chhabildas, N. A., & Olson, R. K. (2001). K. R. Harris, & S. Graham (Eds.), Handbook of learn- A comparison of the cognitive defi cits in reading dis- ing disabilities (2nd ed., pp. 627–642). New York: ability and attention-defi cit/hyperactivity disorder. Guilford Press. Journal of Abnormal Psychology, 110 (1), 157–172. Swanson, H. L., & Hoskyn, M. (1998). Experimental Willcutt, E. G., Pennington, B. F., Olson, R. K., intervention research on students with learning dis- Chhabildas, N., & Hulslander, J. (2005). abilities: A meta-analysis of treatment outcomes. Neuropsychological analyses of comorbidity between Review of Educational Research, 68 (3), 277–321. reading disability and attention defi cit hyperactivity Swanson, H. L., Howard, C. B., & Saez, L. (2006). Do disorder: In search of the common defi cit. different components of working memory underlie Developmental Neuropsychology, 27 , 35–78. different subgroups of reading disabilities? Journal of Willcutt, E. G., Betjemann, R. S., McGrath, L. M., Learning Disabilities, 39 , 252–269. Chhabildas, N. A., Olson, R. K., DeFries, J. C., & Swanson, H. L., Trainin, G., Necoechea, D. M., & Pennington, B. F. (2010). Etiology and neuropsychol- Hammill, D. D. (2003). Rapid naming, phonological ogy of comorbidity between RD and ADHD: The case awareness, and reading. A meta analysis of the corre- for multiple-defi cit models. Cortex, 46 , 1345–1361. lational evidence. Review of Educational Research, Willis, J. O., & Dumont, R. P. (2002). Guide to identifi ca- 73 , 407–444. tion of learning disabilities (3rd ed.). Peterborough, Swanson, H. L., Zheng, X., & Jerman, O. (2009). Working NH: Authors. Available from authors: print copy from memory, short-term memory, and reading disabilities: [email protected] or CD from dumont @ A selective meta-analysis of the literature. Journal of fdu.edu. Learning Disabilities, 42 , 260–287. Wolf, M., & Bowers, P. G. (1999). The double-defi cit Travis, L. E. (1935). Intellectual factors. In G. M. Whipple hypothesis for the developmental dyslexias. Journal of (Ed.), The thirty-fourth yearbook of the National Educational Psychology, 91 , 415–438. Society for the Study of Education: Educational diag- Executive Functioning and Intelligence 27

Emily C. Duggan and Mauricio A. Garcia-Barrera

Over the last few decades, the conceptualization of a theory of frontal lobe relations to human of the most complex cognitive processes of the “higher” cognitive behavior is more recent. Early human brain has received tremendous attention descriptions of this neuroanatomical and func- from researchers. As a result, cognitive models tional relationship include Harlow’s famous case and theories of higher-level functioning are of Phineas Gage and the passage of an iron bar becoming increasingly detailed, sophisticated, through his head, originally presented in 1848 as and evidence-based. The proliferation of research a letter to the editor of the Boston Medical and in these areas, particularly intelligence and exec- Surgical Journal and recently included in The utive functioning, is driven by several key fac- Journal of Neuropsychiatry and Clinical tors, including progress in neuroscience research Neurosciences (Harlow 1848 as reproduced by approaches and the ever-increasing interest Neylan 1999 ). Mr. Gage was severely injured on among researchers, clinicians, and policy makers his left, and probably right, prefrontal areas, but to understand the social importance and implica- was able to walk and talk immediately after the tions of these cognitive abilities. Though we have accident, failing to demonstrate signs of behav- made exceptional progress in our efforts to ioral change in the eyes of Harvard’s Medical understand their principal neural correlates and School examiners. Twenty years later, Harlow underlying neurophysiologic mechanisms, intel- ( 1868 ) published a compilation of the behavioral ligence and executive functioning are still nebu- changes observed in Mr. Gage during the 12 lous—in all regards—and this in itself is the crux years he lived after the injury. The list included of the problem. impulsivity, impatience, and irreverence, among Interest in studying higher cortical processing, other symptoms that interfered with Gage’s its relation to a speciﬁ c brain area, and its impair- capacity to perform at work and in personal rela- ments after neurological damage can be traced tionships. These historical notes constituted the back to the Egyptians (circa 3500 BC) and their earliest description of what became known as “a early yet sophisticated studies about human frontal lobe syndrome.” Interestingly, the early behavior and illness. However, the development reports on Gage’s recovery appear to identify dissociation in the presentation of his symptoms; that is, while Gage’s behavior was childish and E. C. Duggan , M.Sc. (*) • M. A. Garcia-Barrera , Ph.D. immature, his overall intellectual capacities and Department of Psychology , University of Victoria , ability to independently navigate the demands of PO Box 1700 STN CSC , V8W 2Y2 Victoria , BC , Canada the environment seemed to have been mostly e-mail: [email protected]; [email protected] spared from damage. In fact, a characteristic of

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 435 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_27, © Springer Science+Business Media New York 2015 436 E.C. Duggan and M.A. Garcia-Barrera the frontal lobe syndrome post brain injury has tions of intelligence share fundamental attributes been a relatively intact intellectual capacity with (Sternberg and Detterman 1986 ; Terman 1921 ). respect to premorbid levels but a marked impair- According to these symposia, intelligence—at a ment of executive functioning. This dissociation minimum—involves basic mental processes, lies at the foundation of this chapter. Are execu- adaptation to the environment, and the ability to tive functions associated with intelligence? What learn from experience. This defi nition has broad- are the possible relationships between these two ened over time to emphasize the importance of constructs? Given our recent progress in teasing metacognition and higher-order thinking (e.g., out and characterizing cognitive functions, is decision making, problem solving, and reason- there enough evidence of dissociation between ing). In a survey of 1,020 experts, essential ele- these two complex umbrella constructs? ments of intelligence were identifi ed based on the To address these questions, we will discuss the percentage of respondents rating them as impor- defi nitions of intelligence (briefl y) and executive tant. These core elements with the highest per- functions (to a larger extent), followed by a centages included: abstract thinking or reasoning review of the empirical evidence provided by (99.3 %), problem solving ability (97.7 %), psychometric and neurophysiologic approaches capacity to acquire knowledge (96.0 %), memory to the examination of these two constructs’ over- (80.5 %), and adaptation to one’s environment lap, unique features, and their interactions. (77.2 %) (Snyderman and Rothman 1987 ). Additional defi nitions of intelligence provided by prominent researchers over nearly 100 D e fi ning and Conceptualizing years include similar themes. Binet and Simon Intelligence: A Brief Review ( 1916) regarded intelligence as comprising of three distinct components: direction (identifying In addressing the concept of intelligence, Robert what needs to be done and how to do it), adapta- Sternberg once wrote that when “looked at in one tion (customizing an environment-appropriate way, everyone knows what intelligence is; looked strategy for executing a task, then monitoring at in another way, no one does” (Sternberg 2000 , and adjusting the strategy through its implemen- p. 3). Traditionally, theories of intelligence have tation), and criticism (critiquing one’s own been founded on the existence of a general factor thoughts and actions). In Spearman’s two-factor that contributes to performance on diverse cogni- theory of intelligence, the “general intelligence,” tive tests. This factor was labeled general intelli- or g, factor pervades all intellectual perfor- gence, or g (Spearman 1927 ). Though g has mances and accounts for the general mental served great psychometric utility, its defi nition energy that is involved in the most complex men- does little to practically capture the essence of tal activities such as deductive operations actual intelligence. Additionally, the study of (Spearman 1904 , 1927 ). Intelligence has been intelligence has a rich history and defi nitions of viewed as the ability to accomplish “abstract this construct are numerous and diverse. Despite thinking” (Terman 1921 ) and to undertake activ- the absence of a universal defi nition of intelli- ities delineated not only by abstractness but also gence, there are key elements common to nearly by diffi culty, complexity, economy, adaptiveness all of its defi nitions and conceptualizations. Thus, to a goal, and social value (Stoddard 1943 , p. 4). instead of rigorously defi ning intelligence, which Similarly, it has been defi ned as “the capacity to is the purpose of previous chapters in this book, reorganize one’s behavior patterns so as to act let us briefl y explore the core essential aspects of more effectively and more appropriately in novel intelligence that would inform our discussion situations […] and the ability to carry on abstract about the relationship between this construct and thinking […] and problem solve” (Freeman that of executive function. 1955 , pp. 149–50). Wechsler ( 1958) described Two symposia, one in 1921 and a second in intelligence as “the aggregate or global capacity 1986, asking a panel of experts to defi ne intelli- of the individual to act purposefully, to think gence demonstrated that even diverse concep- rationally and to deal effectively with his [or her] 27 Executive Functioning and Intelligence 437 environment” (p. 7), and Das (1973 ) defi ned it as say that intelligence constitutes a confl uence of “the ability to plan and structure one’s behavior the abilities to understand complex ideas and use with an end in view” (p. 27). experience and reasoning to solve problems and One of the most infl uential taxonomies of the adapt to the environment . Based on this defi ni- defi nition of intelligence was proposed by Horn tion, we can reasonably appreciate that intelli- and Cattell (1967 ). Their model included two gence represents the integration of numerous types of intelligence: fl uid intelligence (i.e., the cognitive functions including attention, percep- ability to think logically and solve problems in tion, working memory, abstract representation, novel situations) and crystallized intelligence planning, problem solving, memory, learning, and (i.e., the ability to use knowledge, skills, and language. In fact, the integrative nature of intelli- experience; Cattell 1963 ; Horn and Cattell 1967 ). gence has become a key area of focus within intel- Detterman (1986 ) commented, “intelligence can ligence research (Colom et al. 2010 ; Roca et al. best be defi ned as a fi nite set of independent abil- 2010) and may serve to help differentiate intelli- ities operating as a complex system” (p. 57), gence from other higher-level cognitive functions, while Estes (1986 ) asserted, “intelligence […] is as discussed below. a multifaceted aspect of the processes that enable It is worth noting that in the last decades, animate or inanimate systems to accomplish progress toward elucidating the underlying neu- tasks that involve information processing, prob- ral physiology and mechanisms of intellectual lem solving, and creativity” (p. 66). Gardner functioning has been made. Intelligence has been ( 1983 ) wrote that intelligence “must entail a set correlated with greater total brain volumes of skills of problem solving—enabling the indi- (Wickett et al. 2000; McDaniel 2005 ), frontal, vidual to resolve genuine problems or diffi culties” parietal, temporal, hippocampal, and cerebellar (pp. 60–1). According to Campione and Brown volumes (Luders et al. 2009 ; Toga and Thompson ( 1978), intelligence has two basic components: 2005 ), as well as total grey and white matter vol- an architectural (structural) system—comprised umes, with a particular emphasis on the latter of capacity, durability, and effi ciency—and an (Haier 2004 , 2005 ; Gignac et al. 2003 ). A review executive (control) system—constituting one’s of 37 structural and functional neuroimaging knowledge base, schemes, control processes, and studies provided substantial evidence for a fron- metacognition. Das et al. ( 1994) describe cogni- toparietal intelligence brain network that is tive ability as a function of planning (i.e., cogni- highly distributed, with dorsolateral prefrontal tive control, intentionality, knowledge, and cortex (Broca’s areas 9, 45, 46, 47) and parietal self-regulation), attention, simultaneous process- cortex (BAs 7, 40) and the integrity of this net- ing (i.e., perception of the whole), and successive work’s white matter connectivity—particularly processing (i.e., decision making). Finally, an the arcuate fasciculus—being perhaps the most American Psychological Association consensus important for human intelligence (the Parieto- panel stated that with regards to intelligence, Frontal Integration Theory, Jung and Haier 2007 ; “individuals differ from one another in their abil- see also Chiang et al. 2009; Colom et al. 2009 ; ity to understand complex ideas, to adapt effec- Colom and Thompson 2011 ; Gläscher et al. 2009 , tively to the environment, to learn from 2010 ; Haier 2011 ). The integrity of the frontopa- experience, to engage in various forms of reason- rietal network and its harmonious coactivation is ing, to overcome obstacles by taking thought” considered critical for successful intellectual (Neisser et al. 1996 , p. 77). functioning. Developmentally, general intelli- The considerable sampling of intelligence def- gence follows an inverted U-shaped trajectory initions summarized here illustrates the generally (Jung and Haier 2007 ), though specifi c aspects of consistent and longstanding consensus among intelligence have been found to have unique researchers despite their eclectic range of (often trajectories. For example, fl uid intelligence dissonant) orientations and perspectives—be it declines from the height of neurocognitive matu- psychometric, cognitive, social, developmental, ration around the mid-twenties, while crystal- educational, and so on. Overall, we can surmise to lized intelligence continues to accumulate over 438 E.C. Duggan and M.A. Garcia-Barrera the course of most of the life span—differences to exert a self-regulatory, goal- oriented, conscious which result from a dynamic set of infl uences control over action have remained a relevant issue. (Craik and Bialystok 2006 ). Importantly, current concerns about executive More specifi c fi ndings regarding the underly- function as a construct of interest often include: ing mechanisms of intelligence depend on the way (a) lack of a consensus about the defi nition, (b) intelligence is defi ned and assessed. In the con- lack of agreement about its structural fraction- temporary literature, intelligence is most typically ation (components), and (c) diffi culty to reliably framed in terms of psychometric g , fl uid intelli- measure it (Garcia-Barrera et al. 2011 ). For the gence, and/or crystallized intelligence (Cattell purpose of this chapter, we will only discuss 1963 ). As g is a higher-order, latent characteristic issues surrounding the conceptualization and (and a purely statistical concept) of a complex structure of executive functions. cognitive system (Detterman 2000 ), it is typically measured with tests addressing numerous cognitive domains (e.g., Intelligence Quotient, or “IQ”). Conceptualizing Executive Functions: Interestingly, although most researchers would Issues and Alternatives agree that fl uid and crystallized intelligence are important but not exclusive components to the One of the diffi culties that researchers encounter ecological manifestation of intelligence, often, when performing studies on executive functions is fl uid intelligence tasks of novel problem solving in choosing how to defi ne this construct. A product (e.g., Raven’s Progressive Matrices; Raven 1938 ; of the diversity of theories is the plethora of terms Raven et al. 1988) or crystallized intelligence mea- that have been associated with this complex cogni- sures (e.g., vocabulary) are used in lieu of a full- tive function. Often used terms range from “execu- scale IQ, as these tests have been shown to be tive function” to “executive functions” with a excellent predictors of g . Advocacy for the latent variety of alternatives in between, such as “execu- variables approach (see Evidence through psycho- tive functioning,” “central executive,” “executive metric studies below) has more recently encour- control,” and “self-regulation.” The term “execu- aged the use of diverse measures in the examination tive system” appears to be the most adequate term of higher- order cognition (Haier 2011 ; e.g., for a theoretical conceptualization that emphasizes Martinez et al. 2013 ); however, skewness toward a modular system, but its operationalization in single-measure instruments for fl uid intelligence terms of behavior is diffi cult. This construct has has been a trend in research studies. been associated with the names “metacognition,” Using the basic framework that intelligence is “metacognitive function,” and “metacognitive a conceptual integration of our ability to under- ability,” with an emphasis on its relationship to stand complex ideas, reason, and draw upon expe- awareness and self-regulation of problem solving, rience to solve problems and adapt to the decision making, and learning skills. Finally, it has environment, let us now turn to another seemingly been historically recognized as “frontal lobe func- related higher-order cognitive construct tied to the tion” or “frontal functioning,” terms that have been same frontoparietal network: executive function. persistently used in some disciplines and even in some instruments; however, such names appear localizationist in nature, deemphasizing one of the Executive Function: A Defi nition most important characteristics of executive systems, that is, an extraordinary connectivity and First defi ned by Luria (1973 ) and then labeled by coordinated networking within the brain. Lezak (1982 ), the concept of “executive function” Moreover, this label seems oblivious to the bulk of has proved to be at the top of the list of the most research on the combined roles of prefrontal and elusive constructs in neuropsychology (Jurado parietal cortices in executive functioning, raising and Rosselli 2007 ). Despite its controversial exis- the question whether it would be more accurate to tence, efforts to better understand humans’ ability call it “prefrontoparietal function” instead. 27 Executive Functioning and Intelligence 439

The term “executive function” seems to imply organization of internal representations (Miller the idea of a single-unit system, or acceptance of and Cohen 2001 ). When defi ned as outcomes, its role as an umbrella term; in contrast, the term executive functions are highly variable across “executive functions” would have the opposite subjects, although there is some evidence that effect, emphasizing the possibility of a range of individual variability may be explained at large by them. For the purposes of this chapter, the term genetic infl uences (Friedman et al. 2008 ). executive functions seems to be more appropriate, given three major reasons: fi rst, it emphasizes the idea of diversity while allowing room for concep- Operationalization of Executive tual unity as a characteristic of the system; sec- Functions: Gains from the Latent ond, the term stresses the functional dimension of Variables Approach this construct more than its morphological association, allowing for the idea of cybernetic control Signifi cant advances in the conceptualization and as one of its functions rather than its sole function; operationalization of the construct of intelligence and third, its behavioral operationalization were made possible with the introduction of latent appears more plausible, which is the ultimate variable analysis approach. Exploratory factor objective when this construct is compared to other analysis (i.e., principal components) on psycho- performance-based and behavioral-based con- metric tests yielded a set of factors (latent in structs, such as intelligence. nature), which served for the identifi cation of the Now, what are these executive functions? In a general factor g and its many underlying broad review of the neuropsychological literature, abilities (McGrew 2009 ). Similarly, and due to Jurado and Roselli (2007 ) offered one of the most the latent nature of executive functions, signifi - compelling and clinically applicable defi nitions cant progress in their conceptualization and oper- of this construct. These authors state that: ationalization has been accomplished in the last In a constantly changing environment, executive decade since the introduction of latent variable abilities allow us to shift our mind set quickly and analysis approach to elucidate the possible under- adapt to diverse situations while at the same time lying components within its complex structure inhibiting inappropriate behaviors. They enable us (Miyake et al. 2000 ). In fact, a considerable to create a plan, initiate its execution, and perse- vere on the task at hand until its completion. amount of research has been dedicated to examine Executive functions mediate the ability to organize the latent structure of executive abilities in pre- our thoughts in a goal directed way and are there- schoolers (e.g., Espy et al. 2011 ; Hughes et al. fore essential for success in school and work situa- 2009; Wiebe et al. 2011), kindergarteners (e.g., tions, as well as everyday living. (p. 214) Sadeh et al. 2012 ), school-aged children and ado- I n fl uenced by a confl uence between clinical lescents (e.g., Garcia-Barrera et al. 2011, 2013 ; neuropsychology and cognitive neuroscience per- Gur et al. 2012; Huizinga et al. 2006 ; Lehto et al. spectives, one integrated (representing unity) and 2003 ), young adults (e.g., Frazer 2012 ; Miyake yet multidimensional (representing diversity) def- et al. 2000 ; Friedman and Miyake 2004 ; Friedman inition should propose that executive behaviors et al. 2008 ), and middle-age and older adults (e.g., are the outcomes of the interactions between cog- Adrover-Roig et al. 2012 ; Hull et al. 2008 ). nitive and emotional control processes (Garcia- Three executive functions have gained increas- Barrera et al. 2012 , 2013 ), aimed at the production ing attention due to their more feasible operation- of volitional, purposeful, and effi cient guided alization (e.g., Miyake et al. 2000 ): inhibition of behavior (Lezak 1982 ). These executive interac- goal-irrelevant representations (inhibiting ), fl exi- tions are mediated by cortico-cortical connections ble attentional shift toward goal-relevant repre- within the prefrontal cortex and between prefron- sentations (shifting ), and fl uid update of mental tal cortex and parietal cortex, as well as other representations in working memory as the goal is corticothalamic-striato-cerebellar circuits (Fuster planned and executed (updating ). However, 2000 ), which are involved in rule setting and the efforts have been made to isolate other executive 440 E.C. Duggan and M.A. Garcia-Barrera components that may also serve a role in the approach has demonstrated that the development multiple system interactions that result in every- of executive functions is nonlinear and that dif- day executive behavior, for instance, problem ferent within-subject developmental trajectories representation and emotional control (Frazer can be observed across executive abilities even as 2012 ; Garcia-Barrera et al. 2011 , 2012 ). Problem early as between the ages of six to eleven (Garcia- representation refers to the identifi cation of nov- Barrera et al. 2013 ). Some executive abilities elty, organization of incoming and internalized reach full maturity and plateau in their develop- inputs, and use of strategic seriation to generate a ment by age 15 (e.g., shifting, inhibition; set of temporally organized rules (or syntax) for Huizinga et al. 2006 ), whereas other executive purposeful goals of action. Emotional control is abilities may continue unfolding well into young associated with valence and reward processing adulthood (e.g., working memory; Huizinga et al. and specifi cally refers to the top-down regulation 2006 ), reaching a plateau around the mid-20s of the impact of reward-saliency in behavior and (McAuley and White 2011 ). These differential biasing valence evaluation toward goal-based developmental trajectories have been observed behavioral action (Frazer 2012 ). independently from the gains in speed of process- The latent variable approach requires optimal ing information associated with age, which, as it operationalization of the executive functions. would be expected, count for a portion of the Therefore, one of the secondary gains of modeling variance in test performance. executive functions as latent variables has been a The multidimensionality of executive functions robust psychometric approach to examining their appears to remain somewhat stable during the ear- development, adding to the discussion about the lier adulthood (McAuley and White 2011 ), but lit- unity versus the diversity of executive functions tle is known about the patterns of differential across the life span. Although using mostly cross- decline through later adulthood, given inconsistent sectional design, some research studies support one research fi ndings (Jurado and Rosselli 2007 ). Some or the other side of the debate, and a few suggest researchers have postulated a dedifferentiation that executive functions may develop following a hypothesis, that is, the idea that cognitive abilities unity to diversity gradient, possibly characterized (including executive skills) become less differenti- by a baseline unidimensional (Wiebe et al. 2011 ; ated with age, as correlations between executive Hughes et al. 2009) or two-dimensional system functions signifi cantly increase with age, and (Espy et al. 1999 ; Miller et al. 2012 ), which may groups of executive tasks are reducible to a single fractionate into a range of three or more executive factor in older adults. In other words, this hypoth- subsystems between the ages of six to eight (Brocki esis postulates that through development, cognitive and Bohlin 2004 ; Garon et al. 2008 ; Lehto et al. abilities differentiate from an initial unity in early 2003 ). The multidimensionality of this latent vari- development, and then return to dedifferentiation able-based construct seems feasible in light of as adults age (Balinsky 1941 ; also discussed in models of neural specialization and fractionation of Anstey et al. 2003; and illustrated in Fig. 27.1 ). higher-order cognitive functions during the earlier However, some have argued that this dedifferentia- neurodevelopmental stages (Tsujimoto 2008 ). tion may be related to age-related changes in fl uid Further, an inverted U developmental trajec- intelligence more than executive functions per se tory of this system, delayed with respect to the (Salthouse 2001 ). It is known though that the pat- development of other cognitive systems, has also terns of cognitive decline associated with age fol- been documented (e.g., Craik and Bialystok low closely the reduction in network connectivity 2006). This delayed cognitive maturation is sup- and structural atrophy observed in the elderly ported by developmental trajectories observed in (Craik and Bialystok 2006 ; Jurado and Rosselli frontal and parietal association cortices (Gogtay 2007 ), keeping in mind the interactions between et al. 2004 ), which are strongly associated with brain changes and cognition are nonlinear, executive functions (Collette et al. 2006 ). dynamic, and signifi cantly moderated by genetic, Longitudinal examination using the latent environmental, and social factors (Baltes 1987 ). 27 Executive Functioning and Intelligence 441

Fig. 27.1 Speculative model of the developmental trajectory of three executive functions. Emphasis is placed on the different rates of development during childhood and the differential decline with aging

Findings from Psychometric Studies Understanding the Relationship Between Executive Functioning Historically, the construct of intelligence pre- and Intelligence ceded that of executive function by a large time difference. Since the creation of instruments to A glance into the aforementioned defi nitions of assess intelligence began, neuropsychologists intelligence and executive functions is enough to have been focused on the discussion about the identify the conceptual commonalities between limitations of interpreting the Intelligence these two constructs. When defi ned as the ability Quotient (IQ) as a single composite to measure a to problem solve, plan, and structure goal- range of cognitive abilities as diverse as atten- directed behavior, to produce convergent logical tion, memory, reasoning, and processing speed. reasoning in the face of novelty, and to make not Thus, the unity versus diversity debate, often only effi cient but effi cacious decisions, little is associated with executive functions (e.g., Miyake left for one to distinguish intelligence from exec- et al. 2000 ), had long before been associated with utive functioning. The fi eld of neuropsychologi- intelligence. Is intelligence best represented as a cal assessment has traditionally favored more single general cognitive capacity that changes general descriptors such as “general cognitive over time, or is it best conceived as the outcome ability” or simply “cognitive” or “mental” abili- of the interactions between several cognitive ties to refer to the group of processes that under- abilities? Two of the most robust evidenced- lie the umbrella term “intelligence” (Lezak et al. based premises favoring the idea of diversity over 2012 ), avoiding part of the confusion. that of a single g factor have been that “brain Despite the confl ation observed between function is too complex to be communicated by a these two constructs, a number of efforts have single score” and that a summation score does been made to examine possible divergent com- not have a “predictably direct relationship to the ponents and to better identify the commonalities. size of brain lesions” (Lezak et al. 2012 , p. 22). In Progress has been made particularly through the other words, for neuropsychologists concerned use of the psychometric approach, the study of about brain-behavior relationships, a single score neural correlates, and the examination of clinical suggests a single localization, or at least nodal populations; discussion of these lines of inquiry and vulnerable to the effects of the damage exten- follows. sion. Lezak and colleagues (2012 ) also pointed 442 E.C. Duggan and M.A. Garcia-Barrera out that the patterns of cognitive decline observed Whereas crystallized intelligence (Gc) was in typical and atypical aging are characterized by defi ned as: the different rates in which abilities change. … a person’s breadth and depth of acquired knowl- Similar assertions can be made for the earlier edge of the language, information and concepts of development of these abilities (Blair 2006 ). a specifi c culture, and/or the application of this In contrast, a psychometric perspective, using knowledge. Gc is primarily a store of verbal or a latent variable approach such as factor analysis language-based declarative (knowing what) and procedural (knowing how) knowledge acquired techniques, has been traditionally successful at through the investment of other abilities during confi rming a structure of intelligence that gener- formal and informal educational and general life ally integrates Spearman’s g factor at the top of experiences. (p. 5) the hierarchy. The so-called Cattell-Horn-Carroll It is worth noting from these defi nitions the (CHC) model of intelligence (Blair 2010 ) inte- infl uence of educational experience in crystal- grates the factor-analytical work of John Carroll lized intelligence (Jonson and Bouchard 2005 ) (1993 ) with the two-factor (i.e., fl uid and crystal- and—more important for our discussion in this lized) model of intelligence proposed by Cattell chapter—the resemblance between the defi nition and Horn ( 1967 ) and has become the landmark of fl uid intelligence and executive functions. psychometric structure of intelligence (McGrew Thus, how do fl uid intelligence and executive 2009 ) making it a robust and attractive approach. functions differ? This question and similar ones As Blair ( 2010 ) pointed out, and despite the (e.g., Are they differentially involved in behav- statistical robustness that the latent variable ior? Does one construct subsume the other?) have approach offers, the psychometric tradition does been pondered for decades, yet have been rarely come with several limitations. One of them empirically investigated (Floyd et al. 2010 ). includes the dependency of factor analysis on test Since the recognition in neuropsychology of the selection. Due to the need of including perfor- term executive functions, a strong conceptual mance measures of mental and cognitive abili- relationship between intelligence and executive ties, tests often involve types of abilities that are functions has been largely discussed in the litera- too academic and not very ecologically valid ture. Sternberg ( 1985 ) proposed that executive (e.g., related to everyday life behavior; Floyd functions are common to all cognitive tasks and et al. 2010 ). Another limitation involves the fact that psychometric g merely represents individual that the traditional psychometric instruments differences in executive functioning. involved in the derivation of the g factor do not In fact, signifi cant correlations between intel- include emotional and social aspects of intelli- ligence test scores and executive functioning gence (Lezak et al. 2012 ) or other multiple types measures have been demonstrated (Ardila et al. of intelligence (Gardner 1983 ). 2000 ; Arffa 2007 ; Brydges et al. 2012 ; Burgess Efforts in distinguishing the psychometric et al. 1998 ; Dempster 1991 ). Floyd and col- components involved in the structure of intelli- leagues (2010 ) examined the question “How do gence have produced interesting defi nitions of executive functions fi t with the Cattell-Horn- the two main core factors loading to g , that is, Carroll model?” by conducting a detailed psy- fl uid and crystallized intelligence. In a review of chometric analysis of the relationships between a theories, McGrew (2009 ) captured the defi nition well-established and known executive functions of fl uid intelligence (Gf) as: battery (the Delis-Kaplan Executive Function The use of deliberate and controlled mental opera- System-D-KEFS, Delis et al. 2001 ) and the tions to solve novel problems that cannot be performed automatically. Mental operations often Woodcock-Johnson III Tests of Cognitive include drawing inferences, concept formation, clas- Abilities (WJ-III; Woodcock et al. 2001 ). Their sifi cation, generating and testing hypothesis, identify- analyses demonstrated that measures of both ing relations, comprehending implications, problem constructs are not easily distinguished, that is, all solving, extrapolating, and transforming information. Inductive and deductive reasoning are generally con- D-KEFS subtests not only correlated with intel- sidered the hallmark indicators of Gf. (p. 5) ligence subtests but also measured both broad 27 Executive Functioning and Intelligence 443 ability factors and the general factor ability, as (Ackerman et al. 2005 ; Kane et al. 2005 ). We demonstrated in their exploratory and confi rma- know from the work of Carroll ( 1993) that fl uid tory factor analyses (Floyd et al. 2010 ). Further, a intelligence is closely associated with g , creating factor they label Executive Function (EF) had the the equation EFs(WM ) ~ Gf ~ g , that is, working second most robust loadings to the general factor memory as a proxy for executive functioning is g (0.85 ), surpassed only by their comprehension- closely associated with fl uid intelligence which is knowledge factor (0.95), the latter being a some- closely associated with g . what cluttered factor that included several To examine this issue using a latent variable D-KEFS tasks such as the sorting task (tapping approach, Friedman et al. (2006 ) investigated similar abilities to the Wisconsin Card Sorting the associations between performances on three Test), tasks for verbal reasoning and inference components of executive functioning (i.e., inhib- (20 questions test and word context), and tasks iting, updating working memory, and shifting) in from the WJ-III such as the analysis-synthesis, relation to performance on both fl uid and crystal- verbal comprehension, and general information. lized intelligence measures, in healthy young The comprehension-knowledge factor lies adults. Their fi ndings are consistent with closely between Carroll’s crystallized and fl uid Pennington’s only to some extent. These authors intelligence factors. However, when detailed con- found a robust relationship between their updat- fi rmatory factor analysis is applied, some ing working memory factor and both types of researchers have reported that crystallized knowl- intelligence. Their confi rmatory analysis fol- edge tasks are not as strongly associated with lowed by structural equation modeling confi rmed performance on measures of executive function- that when the correlations between the three ing (e.g., Espy et al. 1999; Krikorian and Bartok executive functions are accounted for (as per in 1998 ; Pennington 1997 ). In her wisdom, Denckla Miyake et al. 2000), the associations between (1996 ) reminds us that this complex overlap inhibiting and shifting with intelligence drop sig- between psychometric g and executive function- nifi cantly, demonstrating that it is through their ing should not be reduced to the idea that execu- shared variance with updating working memory tive functioning is simply “old wine in new that the association between executive function- bottles” (p. 268). Crinella and Yu (1999 ) exam- ing and intelligence operates. Even though these ined animal models, frontal lobe lesion studies, associations were high (loadings of 0.74 with and psychometric studies on clinical populations fl uid intelligence and 0.79 with crystallized intel- and concluded that executive functioning ligence) and accounted for a signifi cant portion although relevant to problem solving neither of the variance in intelligence measures (37–45 %), equals nor explains g in its totality. the authors recognized that there is more to intel- If the general factor g is not a psychometric ligence than just executive functioning or at least equivalent to executive functioning, is fl uid intel- than just inhibiting, shifting, and updating. ligence a considerable equivalent? In the tradi- Salthouse and colleagues (Kail and Salthouse tion of Baddeley’s central executive (Baddeley 1994 ; Salthouse and Pink 2008 ) assert that what 1996), Pennington (e.g., Pennington et al. 1996 ) is common between executive functions tasks proposed that executive functioning is best and intelligence is not working memory but defi ned as the outcome of working memory, speed of processing. Examples of these interac- which he believes is closely associated with fl uid tions are abundant in the recent literature. In intelligence, leaving the rest of g (crystallized Floyd et al. (2010 ), a processing speed factor intelligence) to be regarded as non-executive in accounted for a signifi cant portion of the variance nature. Pennington’s predictions about executive in two out of the four subtests comprising their functions may have left out relevant elements of executive function factor. Adrover-Roig and col- this complex system, but the relationship between leagues (2012 ) used a speed of processing factor working memory and fl uid intelligence has been as a moderator latent variable to effectively psychometrically demonstrated to be high examine the structure of executive functions in 444 E.C. Duggan and M.A. Garcia-Barrera older adults. Speed of processing may play a rel- understanding the underlying neurobiology and evant role in the development of cognitive and corresponding brain networks of intelligence and mental abilities of all sorts, and it is an important similarly, but to a lesser extent, of executive func- variable to account for when examining perfor- tioning; however, studies actively examining the mance in the tasks often used to test these mod- relationship between intelligence and executive els. However, the diffi culty in isolating the pure functioning have been fewer. Though entire vol- infl uence of this foundational process during any umes could be dedicated to documenting these cognitive task remains a challenge. lines of research and their points of convergence Despite the robustness of some of these psy- and divergence, here we attempt to briefl y review chometric fi ndings, Blair ( 2010 ) wisely points the most central points of this literature (with an out that: emphasis on recent studies) to provide a basis for Although working memory capacity would appear our following discussion of contemporary to be a primary determinant of Gf, and perhaps by research dedicated to examining executive func- extension the factor itself, somewhat ironically, Gf tioning and intelligence in tandem. and EF have been shown in clinical and develop- Executive neurologic substrates versus intel- mental research to be distinct from the general factor, as well as from other aspects of intelligence ligence neurologic substrates . Brain lesion analy- (Blair 2006 ). The fi nding that Gf is so highly simi- sis, or lesion-symptom mapping, has served as a lar to g in the psychometric literature but so clearly fundamental research methodology for gaining distinct from it in others, including clinical neuro- insights into brain-behavior relationships and psychology and developmental disabilities, […] has been a central point in the study of intelligence localizing cognitive functions. As previously […] and also of the neurobiology of intelligence. indicated, modern approaches to lesion analysis (p. 231) in relation to intelligence, executive functions, and frontal lobe functioning trace their history Acknowledging Blair’s (2010 ) assertion, we back to Harlow’s famous case of Phineas Gage now turn to reviewing the literature examining ( 1848 , 1868 ), as well as to the broader work of commonalities and differences between execu- numerous infl uential fi gures including Paul tive functioning and intelligence at the level of Broca (1848–1904), Karl Wernicke (1848–1905), their possible neural correlates and when exami- Hughlings Jackson (1835–1911), Norman nation of clinical populations is employed. Geschwind (1926–1984), and Alexander Luria (1902–1977). Research has demonstrated recruit- ment of prefrontal networks for performance on Examining the Neural Correlates both tests of general intelligence (Bishop et al. of Executive Functioning 2008 ; Duncan et al. 2000 ; Esposito et al. 1999 ; and Intelligence Prabhakaran et al. 1997 ) and executive functions (Duncan 2006 ; Duncan and Owen 2000 ; Miller Despite the wide and often confl icting array of 2000 ; Miller and Cohen 2001 ), and prefrontal viewpoints on intelligence and executive func- cortex lesions have been shown to result in a wide tioning, a substantial body of research evidence array of cognitive defi cits, including numerous indicates that both are multifaceted constructs aspects of executive and intellectual functioning, representing the cognitive products of similar depending on the damage (cf. Fuster 1988 ; (but not identical) neural networks. Overall, Goldman-Rakic 1987 ; Luria 1973 ; Roberts et al. lesion and neuroimaging studies have demon- 1998 ; Stuss and Benson 1984; Wise et al. 1996 ). strated the signifi cance of prefrontal and parietal Though it seems reasonable that the prefrontal areas and their connections to higher-level cogni- cortex could serve as the shared fundamental tion including intelligence (Haier 2011 ), execu- neurologic substrate of intelligence and executive tive functioning (Niendam et al. 2012 ), and functions, this argument has mixed support, espe- working memory (Wager and Smith 2003 ). An cially when it comes to localizing within the level extensive body of research has been dedicated to of the prefrontal cortex. 27 Executive Functioning and Intelligence 445

It is relatively common for patients with frontal Tranel et al. 2008). Nevertheless, lesion size and/ lobe lesions to demonstrate preserved psychomet- or distribution has been a point of criticism ric intelligence (Ackerly 1937 ; Hebb 1945 , 1949 ; (among several) in this literature, with similar Hebb and Penfi eld 1940 ; Teuber 1959 ; Weinstein methods being used to report affected executive, and Teuber 1957 ; Milner 1963 ; Bar-On et al. but not intellectual, functioning after focal frontal 2003 ; Black 1976 ; Janowski et al. 1989 ), and lesions. Furthermore researchers counter that frontal lobe damage has been associated with intelligence is not an exclusive manifestation of marked impairments in executive functioning in executive functioning; rather, executive function- spite of grossly preserved or normal intellectual ing represents a critical but not the only compo- abilities, as measured through conventional intel- nent of cognitive information processing required ligence and neuropsychological tests (Bechara for problem solving (Crinella and Yu 1999 ; Kane et al. 1998 ; Damasio et al. 1990 ; Eslinger and and Engle 2002). Although intelligence and Damasio 1985 ; Saver and Damasio 1991 ; Tranel executive functions share many similar neural et al. 2008 ). Although there is a considerable body correlates including prefrontal and parietal asso- of evidence demonstrating that the frontal lobes ciation areas, accumulating evidence indicates are required for human and animal models of that some areas including ventral and anterior executive functioning (e.g., Cummings and prefrontal cortices are likely more unique to Benson 1990 ; Eslinger and Damasio 1985; Teuber executive functioning (Collette et al. 2006 ; 1964 ), frontal lobe function and executive func- Juardo and Rosselli 2007 ; Wager and Smith tioning is not a direct “one-to-one relationship” 2003 ). (see Alvarez and Emory 2006 for a review), and As discussed above (see Sect. “Findings from tests commonly used to assess executive abilities Psychometric Studies ”), subsequent neuroscience traditionally associated with prefrontal regions do models of intelligence have typically fallen under not always show clear regional specifi city (e.g., the scope of one of two orientations (i.e., intelli- Wisconsin Card Sorting Test and dorsolateral pre- gence as a single general cognitive capacity frontal cortex, Milner 1963 ; Rezai et al. 1993 ), changing over time versus intelligence as the but also other frontal and non-frontal regions outcome of a numerous integrated cognitive (Anderson et al. 1991 ; Barcelo and Santome- abilities), both of which are supported by neuro- Calleja 2000 ; Buchsbaum et al. 2005 ; Horner physiologic data. First, deriving from Spearman’s et al. 1996 ; Stuss et al. 1983 ). (1904 , 1927 ) theory, the prefrontal cortex has Preserved psychometric intelligence (as mea- been proposed as a unifi ed neural architecture of sured with a conventional IQ) following frontal the general factor underlying a wide range of cog- lobe damage was challenged by Duncan et al. nitive abilities, or g (e.g., Duncan et al. 2000 ; cf. (1995 , 1996 ) who demonstrated intelligence Roca et al. 2010; Barbey et al. 2012 ). Second, impairments in relation to frontal lobe defi cits deriving from Thomson’s (1951 ) theory (as dis- using high g -loadings measures of fl uid intelli- cussed in Bartholomew et al. 2009 ), tests of intel- gence (e.g., Progressive matrices, Culture Fair), ligence have been proposed to represent the sum which they argue are better for detecting intelli- of numerous separate cognitive operations func- gence changes after brain damage. Such use of tionally mediated by specialized regions of the fl uid intelligence measures has received support brain, particularly the frontal and parietal cortices from imaging studies showing consistent activa- and the white matter association tracts that net- tion patterns within frontal and parietal areas and work these areas (e.g., Colom and Thompson highlighting the particular importance of the lat- 2011; Gläscher et al. 2009 , 2010 ; Jung and Haier eral prefrontal cortex, the supplementary motor 2007). Review of both models, interestingly area/anterior cingulate, the anterior insula/oper- enough, demonstrates that each can account for culum, and the intraparietal sulcus (Bishop et al. the pattern of positive intercorrelations among 2008 ; Duncan et al. 2000 ; Duncan and Owen cog nitive tests and current methods (i.e., psycho- 2000 ; Gray et al. 2003 ; Prabhakaran et al. 1997 ; physiological, neuroimaging, genetic) cannot 446 E.C. Duggan and M.A. Garcia-Barrera

distinguish between them (Bartholomew et al. Brain -lesioned patients and healthy controls . 2009 ; Deary et al. 2010 ). A third plausible model Recently, Roca and colleagues (2010 ) examined of intelligence, proposed by Van Der Maas and the role of fl uid intelligence in a variety of execu- colleagues (2006 ), posits that cognitive processes tive tasks in two experiments with patients with grow over time and mutually interact, which can focal frontal lesions. In one experiment, patients account for the positive correlations between vari- with frontal lesions were signifi cantly impaired ous cognitive tests; however, some aspects of the on measures of intelligence (Culture Fair IQ) and model have received criticism, and it remains to be executive functioning (Wisconsin Card Sort Test seen how successful it will fair when tested in less and Verbal Fluency), executive functioning was theoretical applications (Bartholomew et al. 2009 ). correlated with intelligence, and higher IQ was These models bear great signifi cance in the associated with better executive function task executive functions versus intelligence forum, performance, with differences between patients as they simultaneously implicate executive and controls “entirely explained by g ” (Roca functioning in their positioning (based on the et al. 2010 , p. 243). In a second experiment, signifi cant overlap between the conceptualiza- although executive functioning tasks (Ineco tions of intelligence and executive functions Frontal Screening, Hotel Task, Iowa Gambling defi nitions) and guide the conceptual and Task, Faux Pas test, and Mind in the Eyes task) experimental frameworks for a preponderance were correlated with IQ (Culture Fair), with bet- of current executive- intelligence research. ter performance associated with higher IQ, exec- Additionally, most lesion studies relevant to utive functioning differences between patients this discussion have been fl awed to some extent, and controls remained even after correcting for which has limited our ability to effectively IQ differences, and their results suggested these evaluate executive functioning and intelligence executive defi cits to be associated with greater with respect to one another. Barbey et al. lesion volume in the right anterior prefrontal cor- ( 2012b; see also Barbey et al. 2012a ) summa- tex. The results of this study are intriguing in that rize in their review: it appears that unique contributions of executive Of the neuropsychological patient studies that functioning tasks assessing abilities or behaviors have examined the neural basis of general intel- most different from those traditionally associated ligence (Basso et al. 1973; Black 1976; Eslinger with intelligence are more likely to persist after and Damasio 1985 ; Shallice and Burgess 1991 ; controlling for IQ. Also, considering the likeli- Bechara et al. 1994 ; Duncan et al. 1995 , 1996 ; Burgess and Shallice 1996 ; Isingrini and Vazou hood that components of both intelligence and 1997 ; Parkin and Java 1999 ; Blair and Cipolotti executive functions contribute to psychometric g , 2000 ; Kane and Engle 2002 ; Bugg et al. 2006 ; in so far as they are conceptualized and mea- Gläscher et al. 2009 , 2010 ; Roca et al. 2010 ; sured, we agree that it is benefi cial to examine Tranel et al. 2008 ) and executive functioning (Ptito et al. 1995 ; D’Esposito and Postle 1999 ; executive functioning after controlling for intel- Muller et al. 2002 ; D’Esposito et al. 2006 ; ligence, but we also suggest examining intelli- Baldo and Dronkers 2006 ; Volle et al. 2008 ; gence when controlling for executive functioning Tsuchida and Fellows 2009 ), all share one or (see the discussion on Dichter et al. 2006 , below more of the following features: diffuse (rather than focal) brain lesions, lack of comparison in Clinical Disorders in Adults , as well as the dis- subjects carefully matched for pre- and post- cussions on Martinez et al. 2013 ; Gläscher et al. injury performance measures, and exclusive use 2012 , for an alternative perspective). of general intelligence or executive function Using voxel-based lesion-symptom mapping, tests. (p. 1155) Barbey et al. (2012) found measures of g (WAIS) It is under this challenging context of issues and executive functioning (D-KEFS) were asso- and discrepancies, as well as myriad factors not ciated with circumscribed damage to left lateral- addressed here, that the most recent investiga- ized regions and interconnections of the frontal tions into intelligence and executive functioning and parietal cortices. Although their results indi- operate. cate that executive functioning and intelligence 27 Executive Functioning and Intelligence 447 fundamentally depend on shared neural sub- intelligence, fl uid intelligence, crystallized strates of the frontoparietal network, they also intelligence, and mental processing speed” found areas uniquely related to one but not the (p. 611). Though their study is geared more other, and vice versa. Specifi cally, the right supe- toward the intelligence component of higher- rior and inferior parietal lobe and left inferior order cognitive functioning, these fi ndings still occipital gyrus related to g , while the left anterior hold implication for executive networks, and frontal pole was related to executive function. their approach may be used in the future to exam- Furthermore, in a second study of patients ine executive functions and intelligence together. with focal dorsolateral prefrontal cortex lesions Second, using voxel-based lesion-symptom using a similar methodology, Barbey et al. (2012) mapping in a large sample of patients with focal found the dorsolateral prefrontal cortex function- brain lesions, Gläscher et al. (2012 ) characterized ally supports mechanisms for intelligence (cor- the regional specifi city of two functions (statisti- roborating the existing neuroscience literature cally isolated from performance across multiple and particularly the fi ndings of Woolgar et al. tasks) previously associated with prefrontal net- ( 2010), linking dlPFC to fl uid intelligence), but works: cognitive control and value-based deci- not for the mechanisms of executive functioning sion making. These authors’ fi ndings revealed (as measured with the D-KEFS). In discussing two distinct functional-anatomical prefrontal net- their fi ndings, Barbey et al. (2012) suggest that works. First, cognitive control (response inhibi- “ g and executive function draw on the tion, confl ict monitoring, and switching) was combination of conceptual knowledge and exec- associated with the anterior cingulate cortex and utive processes, and that the communication dorsolateral prefrontal cortex. Second, value- between areas associated with these capacities is based decision making (valuation, decision mak- of critical importance” (p. 1160). It is not readily ing, and reward learning) was associated with the apparent to us if such a conclusion does more to ventromedial, frontopolar, and orbitofrontal areas clarify or to obscure the already grey relationship of the prefrontal cortex. Additionally, “cognitive between executive functions and intelligence; control tasks shared a common performance fac- however, the strength of these authors’ approach tor related to set shifting that was linked to the clearly lies in their use of the latent variables, and rostral anterior cingulate cortex… [while] regions future studies in this fi eld would benefi t from in the ventral PFC were required for decision- implementation of similar methodological making” (p. 14681). Gläscher et al. conclude that procedures. their fi ndings provide detailed lesion-based evi- The use of a priori hypotheses has been con- dence for a remarkable level of functional- sidered another factor limiting the scope of neu- anatomical specifi city in the prefrontal cortex. roanatomical conclusions that can be made about While the research fi ndings presented by higher-order cognitive abilities; two studies Martinez et al. (2013 ) and Gläscher et al. (2012 ) abandoning the a priori approach have provided each contribute further to the understanding of interesting contributions to the debate at hand the neural substrates of executive functions and (Gläscher et al. 2012 ; Martinez et al. 2013 ). First, intelligence, it is their approaches that are per- Martinez et al. (2013 ) recently conducted a study haps even more interesting and which could lead in healthy controls to “identify brain networks to a potential research avenue that (somewhat) whose properties predicted higher-order cogni- bypasses the construct entanglement issue that is tive function” with fi ber-tracking techniques and inherent to intelligence and executive functions. network analysis (p. 608). These authors found Evidence from clinical studies . Perhaps the that “in general, specifi c networks were better diffi cultly in elucidating the nature of the rela- predictors for each different intelligence mea- tionship between executive functions and intelli- sure, but only one was shared by them all […] gence is more broadly related to the populations Temporal to lateral prefrontal connections were we are hoping to gain insights from. As high- relevant to explaining the variability in general lighted above, some similarities in the relevant 448 E.C. Duggan and M.A. Garcia-Barrera psychometric literature have been found distinctly rize that general cognitive ability may account for different in clinical research. Thus, we aim to group executive function differences (Osório review clinically driven areas of research that et al. 2012 ). Due to the characteristic cognitive provide a slightly different perspective on con- pattern in Williams syndrome, this line of research ceptualizing and exploring the executive- offers a potentially signifi cant opportunity to intelligence relationship. more closely examine crystallized aspects of In the context of developmental disorders, the intelligence in relation to fl uid intelligence and relationship between intelligence and executive executive functions, a research topic not readily functioning has perhaps been most thoroughly being addressed in the current literature. However, investigated in individuals with attention defi cit some limitations on this approach must be recog- hyperactivity disorder (ADHD). Many theorize nized. In particular, this study depended on an that symptoms of ADHD arise from executive inaccurate interpretation of the Miyake et al. dysfunction (e.g., self-regulation and behavioral (2000 ) executive function components (e.g., control problems), and early lines of research examining broader working memory rather than purported that executive dysfunction in ADHD updating) and selecting tasks that are best could be attributed to IQ differences, due to the explained by multiple executive domains (e.g., high correlation with many traditional executive Wisconsin Card Sorting Test). Future studies function measures. Executive impairments have should include tasks that examine more carefully since been shown to persist despite normal range the individual executive components and that psychometric intelligence in children with allow for consideration of relevant clinical factors ADHD, as well as in children with other develop- (e.g., verbal/visual strengths or weaknesses). mental disorders commonly linked to executive Another intriguing context in which executive dysfunction (e.g., phenylketonuria and specifi c functioning and intelligence relationships may be learning disabilities; Barkley 1997 ; Berlin 2003 ; considered is clinical disorders featuring promi- Diamond et al. 1997 ; McLean and Hitch 1999 ; nent frontal dysfunction occurring later in life. Stanovich et al. 1997 ; Swanson 1999 ). We identifi ed three recent studies that contribute An example of the examination of the relation- to this newer line of inquiry by investigating ship of executive functions to intelligence in Parkinson’s disease, Frontotemporal dementia, developmental disorders has most recently been and Schizophrenia. As a follow-up to their addressed by Osório and colleagues in individuals research examining how decline of fl uid intelli- with Williams syndrome, a rare neurodevelop- gence contributes to executive functioning defi - mental disorder characterized by a distinct cogni- cits after focal frontal lobe lesions (Roca et al. tive profi le of strengths and weaknesses (i.e., 2010, as discussed above), Roca, Duncan, and impairments in nonverbal domains and executive colleagues used identical methodologies to inves- functioning in contrast to relative verbal strengths) tigate intelligence and executive functioning in along with a pattern of other features (Osório individuals with Parkinson’s disease (Roca et al. et al. 2012 ). These researchers examined the rela- 2012 ) and Frontotemporal dementia (Roca et al. tionships between IQ (WISC FSIQ) and three 2013 ), two neurodegenerative disorders that executive function components selected based on prominently include cognitive changes in these the work of Miyake et al. (Friedman et al. 2006 ; domains. Though there were slight differences, Miyake et al. 2000 ): working memory (reverse overall results from the Parkinson’s disease and digit span), shifting (trail making test, Wisconsin Frontotemporal dementia studies were highly card sorting task), and inhibiting (Stroop task, go/ consistent with their focal frontal lesion data no-go). When compared with typically develop- (Roca et al. 2010 ): some frontal defi cits (as mea- ing controls, individuals with Williams syndrome sured with the “classic” executive tasks, i.e., demonstrated signifi cantly poorer executive func- Wisconsin Card Sorting Test and verbal fl uency) tioning; however, after controlling for IQ most were entirely explained by fl uid intelligence loss, differences were lost, leading the authors to theo- while others (multitasking and theory of mind) 27 Executive Functioning and Intelligence 449 were not. Interestingly, these researchers propose Considering all of these studies and their that the overlap between tests sensitive to frontal fi ndings, we encourage further examination of impairment and fl uid intelligence appears to be changes in executive functioning and intelligence on a continuum, and the dissociation between in these and other clinical neurodevelopmental fl uid intelligence and frontal defi cits depend “on and neurodegenerative disorders, along with con- somewhat different frontal regions, with fl uid tinued consideration of the neuroanatomical sub- intelligence dependent in particular on lateral and strates of age-related cognitive gains and declines dorsomedial regions (Bishop et al. 2008 ; Woolgar (Bugg et al. 2006; Craik and Bialystok 2006 ; et al. 2010), whereas more of the anterior frontal Salthouse 2011 ). cortex is crucial for multitasking and theory of mind” (Roca et al. 2012 , p. 2450). These authors’ investigation into executive functioning and Additional Considerations intelligence in the presence of neurodegenerative diseases contributes valuable insights and encour- As we have presented here, the primary body of ages future inquiry into these and related diseases literature examining the relationship between with prominent executive and intelligence executive functioning and intelligence has con- involvement. Similar to our discussion of their sisted of psychometric and/or imaging studies in original study (Roca et al. 2010 ) above, it is inter- brain-lesioned patients versus healthy controls, esting that executive tests most ecologically with a closely related secondary body of litera- unique from those of intelligence tests appear to ture focusing on these relationships in other rele- remain signifi cant after controlling for IQ, and vant clinical conditions. Numerous other factors collectively these studies highlight the continued and considerations, however, may hold signifi - diffi culty researchers face when selecting assess- cant bearing in resolving the ambiguity between ment tools and attempting to differentiate executive functioning and intelligence. Although between two highly overlapping constructs. not readily referred to in the current body of lit- In a unique approach, Dichter et al. (2006 ) erature, we would feel amiss not to acknowledge examined the relationships between intelligence at least some of these considerations including (WAIS estimated IQ), executive functioning intellectually gifted children (Arffa 2007 ), sex (Trail Making Test, Tower of London, Continuous differences in higher cognition or frontoparietal Performance Task, Wisconsin Card Sorting Test, network activation (cf. Jung and Haier 2007 ; and a visuospatial working memory task requir- Njemanze 2005 ), and drug treatments that affect ing recall of a dot’s location on a screen after a executive functioning but not intelligence (cf. 5-s delay), and P300 event-related potentials Crinella and Yu 1999 ). (elicited using visual and auditory oddball para- A further consideration pertains to training digms) in individuals with Schizophrenia. targeted to improve specifi c cognitive functions. Estimated IQ was signifi cantly correlated with Three compelling research studies recently most executive functioning scores in controls but examined the relationship between music train- not in individuals with Schizophrenia. ing, executive functioning, and intelligence Additionally, for both groups P3 amplitude and (Degé et al. 2011 ; Moreno et al. 2011 ; latency were signifi cantly correlated with execu- Schellenberg 2011 ). Some researchers have theo- tive functioning, but not with estimated IQ until rized that executive functions might serve as variance due to executive functioning was mediators of intelligence, explaining improve- removed. Based on their fi ndings, the authors ments in IQ associated with learning to play suggest that P3-intelligence relationships may be music, and that music training, targeting execu- mediated by executive function abilities and that tive functioning, changes functional brain plas- executive function abilities should be controlled ticity and enhances verbal intelligence through for when examining electrophysiological corre- high-level cognitive skill transfer. These stances lates of intelligence. have received both positive (Degé et al. 2011 ; 450 E.C. Duggan and M.A. Garcia-Barrera

Moreno et al. 2011) and negative (Schellenberg Along similar lines, it may be more conceptu- 2011) support, though several conceptual and ally and practically benefi cial to emphasize the methodological issues remain to be resolved differences between intelligence and executive (Hargreaves and Aksentijevic 2011 ). functioning, rather than their similarities. Recently, considerations regarding the differentiating roles of complexity and novelty suggest Conclusion these are important aspects that may further our understanding of, or differences between, cogni- So what conclusions can we draw about the tive abilities underlying higher-order cognitive relationship between intelligence and executive processes (Dumontheil et al. 2011 ; Duncan et al. functioning? The current evidence suggests that 2012 ; Garlick and Sejnowski 2006 ; Hampshire as we are still trying to understand the differ- et al. 2011 ), although such considerations have ences between these two complex outcomes, been made more commonly from the perspective there are some benefi ts in addressing them as of the fi eld of intelligence research, rather than different psychological constructs, despite the from the perspective of executive functioning fact that their defi nitions signifi cantly overlap research or a balance between the two. and that they seem to be drawing resources In Fig. 27.2 we attempt to visually represent a from the same underlying processes. Based on purely speculative and theoretically based sum- the weaknesses in the literature, we might mary of what we identifi ed to be the current expect to gain clarity by improving and differ- trends in explaining the relationships between the entiating the defi nitions of intelligence and constructs of intelligence and executive function- executive functions and their proposed compo- ing. Starting at the bottom of the fi gure, a three- nents. Additionally, critical evaluation and layered box contains examples of cognitive thoughtful selection of assessment measures processes frequently discussed in the conceptual- for higher-order functioning in coordination ization of executive functioning and intelligence with the application of strong analytic (e.g., problem solving, synthesis, forethought). approaches (e.g., latent variables) appear to be Some of them were discussed in earlier sections key elements to future research in this fi eld. of this chapter (e.g., updating, shifting, problem Valuable insights into intelligence versus exec- representation) and several others were not (e.g., utive functioning are likely to arise from devel- monitoring, prospective memory, categoriza- opments in examining the most evolutionarily tion). Notice that dotted lines mark the separation advanced (but perhaps least understood) corti- between the three layers, indicating a certain cal areas and networks, such as the frontopari- level of fl exibility and mobility between the cog- etal network (e.g., Ramnani and Owen 2004 ). nitive processes (or concepts) included in each Furthermore, it seems inevitable that we will one of them. Each layer represents a state varying ultimately have to adjust our current assump- across time (past, current, future). Mental repre- tions about the hierarchical structure of cogni- sentations and experiences consolidated through tive abilities and psychometric g (Haier 2011 ; knowledge (either educational or experiential, Schlinger 2003). For example, considering the and committed to long-term memory), and skills amount of variance shared between executive strengthened via repetition (e.g., ability to syn- functioning and intelligence (e.g., Barbey et al. thesize) are represented in the past or “crystal- 2012b ; Schretlen et al. 2000 ) and their consis- lized” states. Examples of mental operations and tent activation of highly similar brain networks, cognitive processes believed to be actively it may not be wise to make a priori assumptions engaged in responding to the ever-changing about a hierarchical relationship between the environmental demands are included in the layer two or that intelligence should always be con- representing the current or “fl uid” states. Future- trolled for before looking at executive function- oriented states involve examples of cognitive ing (or vice versa). processes that might be actively engaged in 27 Executive Functioning and Intelligence 451

Fig. 27.2 Speculative integration model explaining pos- resources, and the interactions between ﬂ uid and crystal- sible dissociation between executive functioning and lized states, as well as states inﬂ uencing future actions intelligence, both as outcomes of common cognitive

the mental operations applied to future actions this network may serve as a relevant underlying (e.g., prospective memory, decision making). neural support for the generation of behaviors Taken as a group, these processes are continu- often times referred to be of a “higher order,” ously interacting with each other at different rates which can be also categorized as intelligent and/ and, depending on the task, are based on goal- or executive behaviors. Although outside the directed behavior or environmental demands. The constraints of this chapter, we cannot neglect outcomes of these complex and multiple interac- three constructs representing basic feed-forward tions produce singularities, which can be concep- and feedback resources essential to the function- tually captured under the labels “Intelligence” and ing of the frontoparietal network. These are, pro- “Executive Functioning.” Taking a step forward, cessing speed (relevant to the effi ciency of the we propose to qualify each one of these constructs system), attention (moderating the levels of sys- as being products of differential type of demands, tem engagement), and working memory that is, intelligence (as in g ) fully emerges in situ- (reduced—solely for the purpose of this theoreti- ations when the “complexity” of the demands to cally based diagram—to be the resource repre- the system is high, whereas executive functioning senting the system’s capacity to hold and maintain emerges in cases in which the demands have a active the representations needed during all the unique “novelty” aspect to them. However, both mental operations outputted via the processing of constructs are outcomes of the same underlying the brain structures within the frontoparietal processes, mental operations, and resources. network). Furthermore, the processes contained within Though intelligence, executive functioning, the time-sensitive states may have a common and their corresponding relationships remain elu- neural base, the frontoparietal network and its sive, continued research—emphasizing collabor- rich cortico-subcortical extensions. Ultimately, ative and integrative approaches—in the areas of 452 E.C. Duggan and M.A. Garcia-Barrera lesion studies, psychometrics, neuroimaging, Barbey, A. K., Colom, R., & Grafman, J. (2012). genetics, clinical disorders, neuroplasticity, and Dorsolateral prefrontal contributions to human intelligence. Neuropsychologia . (In press) doi: 10.1016/j. numerous other lines of inquiry will most cer- neuropsychologia.2012.05.017 . tainly stimulate future advancement in this fi eld. Barbey, A. K., Colom, R., Solomon, J., Krueger, F., Forbes, C., & Grafman, J. (2012b). An integrative architecture for general intelligence and executive function revealed by lesion mapping. Brain, 135 (4), References 1154–1164. doi: 10.1093/brain/aws021 . Barcelo, F., & Santome-Calleja, A. (2000). A critical Ackerly, S. (1937). Instinctive, emotional and mental review of the specifi city of the Wisconsin card-sorting changes following prefrontal lobe extirpation. test for the assessment of prefrontal function. Revista American Journal of Psychiatry, 92 , 717–729. de Neurologia, 30 (9), 855. Ackerman, P. L., Beier, M. E., & Boyle, M. O. (2005). Barkley, R. A. (1997). Behavioral inhibition, sustained Working memory and intelligence: The same or differ- attention, and executive function: Constructing a uni- ent constructs? Psychological Bulletin, 131 , 30–60. fying theory of ADHD. Psychological Bulletin, 121 , doi: 10.1037/0033-2909.131.1.30 . 65–94. doi: 10.1037/0033-2909.121.1.65 . Adrover-Roig, D., Sesé, A., Barceló, F., & Palmer, A. Bar-On, R., Tranel, D., Denburg, N. L., & Bechara, A. (2012). A latent variable approach to executive control (2003). Exploring the neurological substrate of emo- in healthy ageing. Brain and Cognition, 78 (3), 284– tional and social intelligence. Brain, 126 , 1790–1800. 299. doi: 10.1016/j.bandc.2012.01.005 . doi: 10.1093/brain/awg177 . Alvarez, J. A., & Emory, E. (2006). Executive function Bartholomew, D. J., Deary, I. J., & Lawn, M. (2009). A and the frontal lobes: A meta-analytic review. new lease of life for Thomson’s bonds model of intel- Neuropsychology Review, 16 (1), 17–42. doi: 10.1007/ ligence. Psychological Review, 116 (3), 567. s11065-006-9002-x . doi: 10.1037/a0016262 . Anderson, S. W., Damasio, H., Jones, R. D., & Tranel, D. Basso, A., De Renzi, E., Faglioni, P., Scotti, G., & (1991). Wisconsin card sorting test performance as a Spinnler, H. (1973). Neuropsychological evidence for measure of frontal lobe damage. Journal of Clinical the existence of cerebral areas critical to the perfor- Experimental Neuropsychology, 13 , 909–922. mance of intelligence tasks. Brain, 96 , 715–728. doi: 10.1080/01688639108405107 . doi: 10.1093/brain/96.4.715 . Anstey, K. J., Hofer, S. M., & Luszcz, M. A. (2003). Cross- Bechara, A., Damasio, A. R., Damasio, H., & Anderson, sectional and longitudinal patterns of dedifferentiation S. W. (1994). Insensitivity to future consequences fol- in late-life cognitive and sensory function: The effects lowing damage to human prefrontal cortex. Cognition, of age, ability, attrition, and occasion of measurement. 50 (1), 7–15. doi: 10.1016/0010-0277(94)90018-3 . Journal of Experimental Psychology: General, 132 (3), Bechara, A., Damasio, H., Tranel, D., & Anderson, S. W. 470. doi: 10.1037/0096-3445.132.3.470 . (1998). Dissociation of working memory from deci- Ardila, A., Pineda, D., & Rosselli, M. (2000). Correlation sion making within the human prefrontal cortex. The between intelligence test scores and executive func- Journal of Neuroscience, 18 (1), 428–437. tion measures. Archives of Clinical Neuropsychology, Berlin, L. (2003). The role of inhibitory control and execu- 15 (1), 31–36. doi: 10.1016/S0887-6177(98)00159-0 . tive functioning in hyperactivity/ADHD. Comprehensive Arffa, S. (2007). The relationship of intelligence to execu- summaries of Uppsala dissertations from the Faculty of tive function and non-executive function measures in a Social Sciences 120. Uppsala: Acta Universitatis sample of average, above average, and gifted youth. Upsaliensis. Archives of Clinical Neuropsychology, 22 (8), 969– Binet, A., & Simon, T. (1916). The development of intel- 978. doi: 10.1016/j.acn.2007.08.001 . ligence in children (The Binet-Simon scale) translated Baddeley, A. (1996). Exploring the central executive. The from articles in L’Année psychologique from 1905, Quarterly Journal of Experimental Psychology: 1908, 1911 by Elizabeth S. Kite. (Publications of the Section A, 49(1), 5–28. doi: 10.1080/027249896392784 . training school at Vineland New Jersey). Baltimore: Baldo, J. V., & Dronkers, N. F. (2006). The role of Williams & Wilkins Company. inferior parietal and inferior frontal cortex in Bishop, S. J., Fossella, J., Croucher, C. J., & Duncan, working memory. Neuropsychology, 20, 529–538. J. (2008). COMT val158met genotype affects recruit- doi: 10.1037/0894-4105.20.5.529 . ment of neural mechanisms supporting fl uid intelli- Balinsky, B. (1941). An analysis of the mental factors of gence. Cerebral Cortex, 18 (9), 2132–2140. various age groups from nine to sixty. Genetic doi: 10.1093/cercor/bhm240 . Psychology Monographs, 23 , 191–243. Black, F. W. (1976). Cognitive defi cits in patients with Baltes, P. B. (1987). Theoretical propositions of life-span unilateral war-related frontal lobe lesions. Journal of developmental psychology: On the dynamics between Clinical Psychology, 32 , 366–372. growth and decline. Developmental Psychology, 23 , doi: 10.1002/1097-4679(197604)32:2<366::AID- 611–626. doi: 10.1037/0012-1649.23.5.611 . JCLP2270320234>3.0.CO;2-F . 27 Executive Functioning and Intelligence 453

Blair, C. (2006). Towards a revised theory of general intel- Neuroscience, 139 (1), 209–221. doi: 10.1016/j. ligence: Further examination of fl uid cognitive abili- neuroscience.2005.05.035 . ties as a unique aspect of human cognition. Behavioral Colom, R., & Thompson, P. M. (2011). Understanding and Brain Sciences, 29 , 145–153. doi: 10.1017/ human intelligence by imaging the brain. In S0140525X06419038 . T. Chamoroo-Premuzic, S. von Stumm, & A. Furnham Blair, C. (2010). Fluid cognitive abilities and general (Eds.), The Wiley-Blackwell handbook of individual intelligence. A life-span neuroscience perspective. In differences (pp. 330–352). Oxford: Wiley-Blackwell. R. M. Lerner (Series Ed.) & W. F. Overton (Volume Colom, R., Haier, R. J., Head, K., Alvarez-Linera, J., Ed.), Handbook of life-span development : Vol. 1. Angeles Quiroga, M., Chun Shih, P., & Jung, R. E. Cognition, biology and methods (pp. 226–258). (2009). Gray matter correlates of fluid, crystal- Hovoken: John Wiley & Sons. lized, and spatial intelligence: Testing the P-FIT Blair, R. J. R., & Cipolotti, L. (2000). Impaired social model. Intelligence, 37, 124–135. doi: 10.1016/j. response reversal: A case of “acquired sociopathy”. intell.2008.07.007 . Brain, 123, 1122–1141. doi: 10.1093/brain/123.6.1122 . Colom, R., Karama, S., Jung, R. E., & Haier, R. J. (2010). Brocki, K. C., & Bohlin, G. (2004). Executive functions in Human intelligence and brain networks. Dialogues in children aged 6 to 13: A dimensional and developmen- Clinical Neuroscience, 12 (4), 489. tal study. Developmental Neuropsychology, 26 (2), Craik, F. I., & Bialystok, E. (2006). Cognition through the 571–593. doi: 10.1207/s15326942dn2602_3 . lifespan: Mechanisms of change. Trends in Cognitive Brydges, C. R., Reid, C. L., Fox, A. M., & Anderson, M. Sciences, 10 (3), 131–138. doi: 10.1016/j. (2012). A unitary executive function predicts intelli- tics.2006.01.007 . gence in children. Intelligence, 40 (5), 458–469. Crinella, F. M., & Yu, J. (1999). Brain mechanisms and doi: 10.1016/j.intell.2012.05.006 . intelligence: Psychometric g and executive function. Buchsbaum, B. R., Greer, S., Chang, W. L., & Berman, Intelligence, 27(4), 299–327. doi: 10.1016/ K. F. (2005). Meta-analysis of neuroimaging studies S0160-2896(99)00021-5 . of the Wisconsin card-sorting task and component Cummings, J. L., & Benson, D. F. (1990). Subcortical processes. Human Brain Mapping, 25 , 35–45. mechanisms and human thought. In J. L. Cummings doi: 10.1002/hbm.20128 . (Ed.), Subcortical dementia (pp. 251–260). New York: Bugg, J. M., Zook, N. A., DeLosh, E. L., Davalos, D. B., Oxford University Press. & Davis, H. P. (2006). Age differences in fl uid intelli- D’Esposito, M., & Postle, B. R. (1999). The dependence gence: Contributions of general slowing and frontal of span and delayed response performance on prefron- decline. Brain and Cognition, 62 (1), 9–16. tal cortex. Neuropsychologia, 37, 89–101. doi: 10.1016/ doi: 10.1016/j.bandc.2006.02.006 . S0028-3932(99)00021-4 . Burgess, P. W., & Shallice, T. (1996). Response suppres- D’Esposito, M., Cooney, J. W., Gazzaley, A., Gibbs, S. E. sion, initiation and strategy use following frontal B., & Postle, B. R. (2006). Is the prefrontal cortex nec- lobe lesions. Neuropsychologia, 34, 263–272. doi: essary for delay task performance? Evidence from 10.1016/0028-3932(95)00104-2 . lesion and fMRI data. Journal of the International Burgess, P. W., Alderman, N., Evans, J., Emslie, H., & Neuropsychological Society, 12 , 248–260. Wilson, B. A. (1998). The ecological validity of tests doi: 10.1017/S1355617706060322 . of executive function. Journal of the International Damasio, A. R., Tranel, D., & Damasio, H. (1990). Neuropsychological Society, 4 (6), 547–558. Individuals with sociopathic behavior caused by fron- Campione, J. C., & Brown, A. L. (1978). Toward a theory tal damage fail to respond autonomically to social of intelligence: Contributions from research with stimuli. Behavioral Brain Research, 41 (2), 81–94. retarded children. Intelligence, 2 , 279–304. doi: 10.1016/0166-4328(90)90144-4 . doi: 10.1016/0160-2896(78)90020-X . Das, J. P. (1973). Cultural deprivation and cognitive com- Carroll, J. B. (1993). Human cognitive abilities: A survey petence. In N. R. Ellis (Ed.), International review of of factor analytic studies. New York: Cambridge research in mental retardation (Vol. 6, pp. 1–53). University. New York: Academic. Cattell, R. B. (1963). Theory of fl uid and crystallized Das, J. P., Naglieri, J. A., & Kirby, J. R. (1994). Assessment intelligence: A critical experiment. Journal of of cognitive processes: The PASS theory of intelli- Educational Psychology, 54 , 1–22. gence . Needham Heights: Allyn & Bacon. Chiang, M. C., Barysheva, M., Shattuck, D. W., Lee, A. D., Deary, I. J., Penke, L., & Johnson, W. (2010). The neuro- Madsen, S. K., Avedissian, C., Klunder, A. D., Toga, science of human intelligence differences. Nature A. W., McMahon, K. L., de Zubicaray, G. I., Wright, Reviews Neuroscience, 11(3), 201–211. doi: 10.1038/ M. J., Srivastava, A., Balov, N., & Thompson, P. M. nrn2793 . (2009). Genetics of brain fi ber architecture and intellec- Degé, F., Kubicek, C., & Schwarzer, G. (2011). Music les- tual performance. The Journal of Neuroscience, 29 (7), sons and intelligence: A relation mediated by execu- 2212–2224. doi: 10.1523/JNEUROSCI.4184-08.2009 . tive functions. Music Perception, 29 (2), 195–201. Collette, F., Hogge, M., Salmon, E., & Van der Linden, M. Delis, D. C., Kaplan, E., & Kramer, J. H. (2001). Delis (2006). Exploration of the neural substrates of execu- Kaplan executive function system . San Antonio: The tive functioning by functional neuroimaging. Psychological Corporation. 454 E.C. Duggan and M.A. Garcia-Barrera

Dempster, F. N. (1991). Inhibitory processes: A neglected Esposito, G., Kirkby, B. S., van Horn, J. D., Ellmore, dimension of intelligence. Intelligence, 15 (2), 157– T. M., & Berman, K. F. (1999). Context-dependent, 173. doi: 10.1016/0160-2896(91)90028-C . neural system-specifi c neurophysiological concomi- Denckla, M. B. (1996). A theory and model of executive tants of ageing: Mapping PET correlates during cogni- function. A neuropsychological perspective. In G. R. tive activation. Brain, 122 , 963–979. doi: 10.1093/ Lyon & N. A. Krasnegor (Eds.), Attention, memory, brain/122.5.963 . and executive function (pp. 263–278). Baltimore: Espy, K. A., Kaufmann, P. M., McDiarmid, M. D., & P.H. Brookes. Glisky, M. L. (1999). Executive functioning in pre- Detterman, D. K. (1986). Human intelligence is a com- school children: Performance on A-not-B and other plex system of separate processes. In R. J. Sternberg & delayed response format tasks. Brain and Cognition, D. K. Detterman (Eds.), What is intelligence? 41 (2), 178–199. doi: 10.1006/brcg.1999.1117 . Contemporary viewpoints on its nature and defi nition Espy, K. A., Sheffi eld, T. D., Wiebe, S. A., Clark, C. A., & (pp. 57–61). Norwood: Ablex Publishing Corporation. Moehr, M. J. (2011). Executive control and dimen- Detterman, D. K. (2000). General intelligence and the sions of problem behaviors in preschool children. defi nition of phenotypes. In G. R. Bock, J. A. Goode, Journal of Child Psychology and Psychiatry, 52 (1), & K. Webb (Eds.), The nature of intelligence, Novartis 33–46. doi: 10.1111/j.1469-7610.2010.02265.x . foundation symposium 233 (pp. 136–148). Chichester: Estes, W. K. (1986). Where is intelligence? In R. J. Wiley. Sternberg & D. K. Detterman (Eds.), What is intelli- Diamond, A., Prevor, M. B., Callendar, G., & Druin, D. P. gence? Contemporary viewpoints on its nature and (1997). Prefrontal cognitive defi cits in children treated defi nition (pp. 63–67). Norwood: Ablex Publishing early and continuously for PKU. Monographs of the Corporation. Society for Research in Child Development, 62 (4), Floyd, R. G., Bergeron, R., Hamilton, G., & Parra, G. R. 1–208. (2010). How do executive functions fi t with the Dichter, G. S., van der Stelt, O., Boch, J. L., & Belger, A. Cattell–Horn–Carroll model? Some evidence from a (2006). Relations among intelligence, executive func- joint factor analysis of the Delis–Kaplan executive tion, and P300 event related potentials in schizophrenia. function system and the Woodcock–Johnson III tests The Journal of Nervous and Mental Disease, 194 (3), of cognitive abilities. Psychology in the Schools, 179–187. doi: 10.1097/01.nmd.0000202490.97425.de . 47 (7), 721–738. doi: 10.1002/pits.20500 . Dumontheil, I., Thompson, R., & Duncan, J. (2011). Frazer, J. (2012). The integrative neuropsychological the- Assembly and use of new task rules in fronto-parietal ory of executive-related abilities and component cortex. Journal of Cognitive Neuroscience, 23 (1), transactions (INTERACT): A novel validation study . 168–182. doi: 10.1162/jocn.2010.21439 . Unpublished doctoral dissertation, University of Duncan, J. (2006). EPS Mid-Career Award 2004: Brain mech- Victoria, Victoria, Canada. anisms of attention. Quarterly Journal of Experimental Freeman, F. S. (1955). Theory and practice of psychologi- Psychology, 59 , 2–27. doi: 10.1080/17470210500260674 . cal testing . New York: Holt, Rinehart, and Winston. Duncan, J., & Owen, A. M. (2000). Common regions of Friedman, N. P., & Miyake, A. (2004). The relations the human frontal lobe recruited by diverse cognitive among inhibition and interference control func- demands. Trends in Neurosciences, 23 (10), 475–483. tions: A latent-variable analysis. Journal of doi: 10.1016/S0166-2236(00)01633-7 . Experimental Psychology. General, 133(1), 101. Duncan, J., Burgess, P., & Emslie, H. (1995). Fluid intel- doi:10.1037/0096-3445.133.1.101 . ligence after frontal lobe lesions. Neuropsychological, Friedman, N. P., Miyake, A., Corley, R. P., Young, S. E., 33 , 261–268. doi: 10.1016/0028-3932(94)00124-8 . DeFries, J. C., & Hweitt, J. K. (2006). Not all execu- Duncan, J., Emslie, H., Williams, P., Johnson, R., & Freer, tive functions are related to intelligence. Psychological C. (1996). Intelligence and the frontal lobe: The organi- Science, 17 , 172–179. zation of goal-directed behavior. Cognitive Psychology, Friedman, N. P., Miyake, A., Young, S. E., DeFries, J. C., 30 , 257–303. doi: 10.1006/cogp.1996.0008 . Corley, R. P., & Hewitt, J. K. (2008). Individual differ- Duncan, J., Seitz, R. J., Kolodny, J., Bor, D., Herzog, H., ences in executive functions are almost entirely genetic Ahmed, A., Newell, F. N., & Emslie, H. (2000). A in origin. Journal of Experimental Psychology: General, neural basis for general intelligence. Science, 289 , 137 (2), 201. doi: 10.1037/0096-3445.137.2.201 . 457–460. doi: 10.1126/science.289.5478.457 . Fuster, J. M. (1988). The prefrontal cortex: Anatomy, Duncan, J., Schramm, M., Thompson, R., & Dumontheil, physiology, and neuropsychology of the frontal lobe . I. (2012). Task rules, working memory, and fl uid intel- New York: Raven. ligence. Psychonomic Bulletin and Review, 19 (5), Fuster, J. M. (2000). Executive frontal functions. 864–870. doi: 10.3758/s13423-012-0225-y . Experimental Brain Research, 133 (1), 66–70. Eslinger, P. L., & Damasio, A. R. (1985). Severe distur- doi: 10.1007/s002210000401 . bance of higher cognition after bilateral frontal lobe Garcia-Barrera, M. A., Kamphaus, R. W., & Bandalos, D. ablation: Patient EVR. Neurology, 35 , 1731–1741. (2011). Theoretical and statistical derivation of a doi: 10.1212/WNL.35.12.1731 . screener for the behavioral assessment of executive 27 Executive Functioning and Intelligence 455

functions in children. Psychological Assessment, 23 , Gray, J. R., Chabris, C. F., & Braver, T. S. (2003). Neural 64–79. doi: 10.1037/a0021097 . mechanisms of general fl uid intelligence. Nature Garcia-Barrera, M., Frazer, J., & Areshenkoff, C. (2012). Neuroscience, 6 , 316–321. doi: 10.1038/nn1014 . Theoretical derivation and empirical validation of an Gur, R. C., Richard, J., Calkins, M. E., Chiavacci, R., integrative neuropsychological theory of executive- Hansen, J. A., Bilker, W. B., Loughead, J., Connolly, related abilities and component transactions J. J., Qiu, H., Mentch, F. D., Abou-Sleiman, P. M., IINTERACT) [Abstract]. Journal of the International Hakonarson, H., & Gur, R. E. (2012). Age group and Neuropsychological Society, 18(S2), 61. doi: 10.1017/ sex differences in performance on a computerized S1355617712001063 . neurocognitive battery in children age 8–21. Garcia-Barrera, M. A., Karr, J. E., & Kamphaus, R. W. Neuropsychology, 26 (2), 251. doi: 10.1037/a0026712 . (2013). Longitudinal applications of a behavioral Haier, R. J. (2011). Biological basis of intelligence. In screener of executive functioning: Assessing factorial R. J. Sternberg & S. B. Kaufman (Eds.), Cambridge invariance and exploring latent growth. Psychological handbook of intelligence (pp. 351–370). New York: Assessment, 25(4), 1300–1313. doi: 10.1037/a0034046 . Cambridge University Press. Gardner, H. (1983). Frames of mind: The theory of Haier, R. J., Jung, R. E., Yeo, R. A., Head, K., & Alkire, multiple intelligences . New York: Basic Books. M. T. (2004). Structural brain variation and general Garlick, D., & Sejnowski, T. J. (2006). There is more to intelligence. NeuroImage, 23 (1), 425–433. fl uid intelligence than working memory capacity and doi: 10.1016/j.neuroimage.2004.04.025 . executive function. Behavioral and Brain Sciences, Haier, R. J., Jung, R. E., Yeo, R. A., Head, K., & Alkire, 29 , 134–135. doi: 10.1017/S0140525X0630903X . M. T. (2005). The neuroanatomy of general intelli- Garon, N., Bryson, S. E., & Smith, I. M. (2008). Executive gence: Sex matters. NeuroImage, 25 (1), 320–327. function in preschoolers: A review using an integrative doi: 10.1016/j.neuroimage.2004.11.019 . framework. Psychological Bulletin, 134 (1), 31. Hampshire, A., Thompson, R., Duncan, J., & Owen, doi: 10.1037/0033-2909.134.1.31 . A. M. (2011). Lateral prefrontal cortex subregions Gignac, G., Vernon, P. A., & Wickett, J. C. (2003). Factors make dissociable contributions during fl uid reasoning. infl uencing the relationship between brain size and intel- Cerebral Cortex, 21 (1), 1–10. doi: 10.1093/cercor/ ligence. In E. Nyborg (Ed.), The scientifi c study of gen- bhq085 . eral intelligence: Tribute to Arthur R. Jensen Hargreaves, D. J., & Aksentijevic, A. (2011). Music, IQ, (pp. 93–106). Kidlington/Oxford: Elsevier Science, Ltd. and the executive function. British Journal of Gläscher, J., Tranel, D., Paul, L. K., Rudrauf, D., Rorden, Psychology, 102(3), 306–308. doi: 10.1111/j.2044-8295. C., Hornaday, A., Grabowski, T., Damasio, H., & 2011.02029.x . Adolphs, R. (2009). Lesion mapping of cognitive abil- Harlow, J. M. (1848). Passage of an iron rod through the ities linked to intelligence. Neuron, 61 (5), 681–691. head. Boston Medical and Surgical Journal, 39 , doi: 10.1016/j.neuron.2009.01.026 . 389–393. Gläscher, J., Rudrauf, D., Colom, R., Paul, L. K., Tranel, Harlow, J. M. (1868). Recovery from the passage of an D., Damasio, H., & Adolphs, R. (2010). Distributed iron bar through the head. Publications of the neural system for general intelligence revealed by Massachusetts Medical Society, 2, 327–347. doi: 10.11 lesion mapping. Proceedings of the National Academy 77/0957154X9300401407 . of Sciences, 107 (10), 4705–4709. doi: 10.1073/ Hebb, D. O. (1945). Man’s frontal lobes: A critical review. pnas.0910397107 . Archives of Neurology and Psychiatry, 54 , 10–24. Gläscher, J., Adolphs, R., Damasio, H., Bechara, A., Hebb, D. O. (1949). The organization of behavior . Rudrauf, D., Calamia, M., Paul, L. K., & Tranel, D. New York: Wiley. (2012). Lesion mapping of cognitive control and value- Hebb, D. O., & Penfi eld, W. (1940). Human behavior after based decision making in the prefrontal cortex. extensive bilateral removal from the frontal lobes. Proceedings of the National Academy of Sciences, Archives of Neurology and Psychiatry, 44 (2), 421. 109 (36), 14681–14686. doi: 10.1073/pnas.1206608109 . Horn, J. L., & Cattell, R. B. (1967). Age differences in Gogtay, N., Giedd, J. N., Lusk, L., Hayashi, K. M., cognitive capabilities. Acta Psychologica, 26 , 107– Greenstein, D., Vaituzis, A. C., Nugent, T. F., 3rd, 129. doi: 10.1093/geronj/21.2.210 . Herman, D. H., Clasen, L. S., Toga, A. W., Rapoport, Horner, M. D., Flashman, L. A., Freides, D., Epstein, J. L., & Thompson, P. M. (2004). Dynamic mapping C. M., & Bakay, R. A. (1996). Temporal lobe epilepsy of human cortical development during childhood and performance on the Wisconsin card sorting test. through early adulthood. Proceedings of the Journal of Clinical Experimental Neuropsychology, National Academy of Sciences of the United States 18 , 310–313. doi: 10.1080/01688639608408285 . of America, 101(21), 8174–8179. doi: 10.1073/ Hughes, C., Ensor, R., Wilson, A., & Graham, A. pnas.0402680101 . (2009). Tracking executive function across the Goldman-Rakic, P. S. (1987). Circuitry of primate pre- transition to school: A latent variable approach. frontal cortex and regulation of behavior by represen- Developmental Neuropsychology, 35(1), 20–36. tational memory. In F. Plum Editor (Ed.), Handbook of doi:10.1080/87565640903325691 . physiology: The nervous system (Vol. 5, pp. 373–417). Huizinga, M., Dolan, C. V., & van der Molen, M. W. Bethesda: American Physiological Society. (2006). Age-related change in executive function: 456 E.C. Duggan and M.A. Garcia-Barrera

Developmental trends and a latent variable analysis. Martinez, K., Villalon-Reina, J., Kessel, D., Joshi, A., Neuropsychologia, 44 (11), 2017–2036. doi: 10.1016/j. Pineda, J. A., Jahanshad, N., Nir, T., Eschenburg, K., neuropsychologia.2006.01.010 . Roman, F. J., Burgaleta, M., Solana, A. B., Angeles Hull, R., Martin, R. C., Beier, M. E., Lane, D., & Quiroga, M., Colom, R., & Thompson, P. M. (2013). Hamilton, A. C. (2008). Executive function in older Exploratory factor analysis of brain networks reveals adults: A structural equation modeling approach. sub-networks related to cognitive performance. Neuropsychology, 22 (4), 508. doi: 10.1037/0894-4105. International Symposium on Biomedical Imaging, 10 , 22.4.508 . 608–611. doi: 10.1109/ISBI.2013.6556548 . Isingrini, M., & Vazou, F. (1997). Relation between fl uid McAuley, T., & White, D. A. (2011). A latent variables intelligence and frontal lobe in older adults. International examination of processing speed, response inhibition, Journal of Aging and Human Development, 45 , 99–109. and working memory during typical development. doi: 10.2190/WHWX-YNVB-079V-2L74 . Journal of Experimental Child Psychology, 108 (3), Janowski, J. S., Shimamura, A. P., & Squire, L. R. (1989). 453–468. doi: 10.1016/j.jecp.2010.08.009 . Memory and metamemory: Comparisons between McDaniel, M. A. (2005). Big-brained people are smarter: patients with frontal lobe lesions and amnesic patients. A meta-analysis of the relationship between in vivo Psychobiology, 17 , 3–11. brain volume and intelligence. Intelligence, 33 (4), Johnson, W., & Bouchard, T. J., Jr. (2005). The structure of 337–346. doi: 10.1016/j.intell.2004.11.005 . human intelligence: It is verbal, perceptual, and image McGrew, K. S. (2009). CHC theory and the human rotation (VPR), not fl uid and crystallized. Intelligence, cognitive abilities project: Standing on the shoul- 33 (4), 393–416. doi:10.1016/j.intell.2004.12.002 . ders of the giants of psychometric intelligence Jung, R. E., & Haier, R. J. (2007). The Parieto-Frontal research. Intelligence, 37, 1–10. doi:10.1016/j. Integration Theory (P-FIT) of intelligence: Converging intell.2008.08.004 . neuroimaging evidence. Behavioral and Brain Sciences, McLean, J. F., & Hitch, G. J. (1999). Working memory 30 (2), 135–154. doi: 10.1017/S0140525X07001185 . impairments in children with specifi c arithmetic learn- Jurado, M. B., & Rosselli, M. (2007). The elusive nature ing diffi culties. Journal of Experimental Child of executive functions: A review of our current under- Psychology, 74, 240–260. doi: 10.1006/jecp.1999.2516 . standing. Neuropsychology Review, 17 (3), 213–233. Miller, E. K. (2000). The neural basis of top-down control doi: 10.1007/s11065-007-9040-z . of visual attention in the prefrontal cortex. In Kail, R., & Salthouse, T. A. (1994). Processing speed as a S. Monsell & J. Driver (Eds.), Attention and perfor- mental capacity. Acta Psychologica, 86 , 199–225. mance (p. 18). Cambridge, MA: MIT Press. doi: 10.1016/0001-6918(94)90003-5 . Miller, E. K., & Cohen, J. D. (2001). An integrative theory Kane, M. J., & Engle, R. W. (2002). The role of prefrontal of prefrontal cortex function. Annual Review of cortex in working-memory capacity, executive atten- Neuroscience, 24 (1), 167–202. doi: 10.1146/annurev. tion, and general fl uid intelligence: An individual- neuro.24.1.167 . differences perspective. Psychonomic Bulletin and Miller, M. R., Giesbrecht, G. F., Müller, U., McInerney, Review, 9 (4), 637–671. doi: 10.3758/BF03196323 . R. J., & Kerns, K. A. (2012). A latent variable Kane, M. J., Hambrick, D. Z., & Conway, A. R. A. (2005). approach to determining the structure of executive Working memory capacity and fl uid intelligence are function in preschool children. Journal of Cognition strongly related constructs: Comment on Ackerman, and Development, 13(3), 395–423. doi: 10.1080/15248 Beier, and Boyle. Psychological Bulletin, 131 , 66–71. 372.2011.585478 . doi: 10.1037/0033-2909.131.1.66 . Milner, B. (1963). Effects of different brain lesions Krikorian, R., & Bartok, J. (1998). Developmental data for on cards sorting: The role of the frontal lobes. Archives the Porteus Maze test. The Clinical Neuropsychologist, of Neurology, 9 , 90–100. doi: 10.1001/archneur.1963. 12 , 305–310. doi: 10.1076/clin.12.3.305.1984 . 00460070100010 . Lehto, J. E., Juujärvi, P., Kooistra, L., & Pulkkinen, L. (2003). Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, Dimensions of executive functioning: Evidence from chil- A. H., Howerter, A., & Wager, T. D. (2000). The unity dren. British Journal of Developmental Psychology, 21 , and diversity of executive functions and their contribu- 59–80. doi: 10.1348/026151003321164627 . tions to complex “frontal lobe” tasks: A latent variable Lezak, M. D. (1982). The problem of assessing executive analysis. Cognitive Psychology, 41 (1), 49–100. functions. International Journal of Psychology, 17 , doi: 10.1006/cogp.1999.0734 . 281–297. doi: 10.1080/00207598208247445 . Moreno, S., Bialystok, E., Barac, R., Schellenberg, E. G., Lezak, M. D., Howieson, D., Bigler, E., & Tranel, D. Cepeda, N. J., & Chau, T. (2011). Short-term music (2012). Neuropsychological assessment (5th ed.). training enhances verbal intelligence and executive New York: Oxford University Press. function. Psychological Science, 22 (11), 1425–1433. Luders, E., Narr, K. L., Thompson, P. M., & Toga, doi: 10.1177/0956797611416999 . A. W. (2009). Neuroanatomical correlates of intel- Muller, N. G., Machado, L., & Knight, R. T. (2002). ligence. Intelligence, 37 (2), 156–163. doi: 10.1016/j. Contributions of subregions of the prefrontal cortex to intell.2008.07.002 . working memory: Evidence from brain lesions in Luria, A. R. (1973). The working brain. New York: Basic humans. Journal of Cognitive Neuroscience, 14 , 673– Books. 686. doi: 10.1162/08989290260138582 . 27 Executive Functioning and Intelligence 457

Neisser, U., Boodoo, G., Bouchard, T. J., Jr., Boykin, A. W., Rezai, K., Andreasen, N. C., Alliger, R., Cohen, G., Brody, N., Ceci, S. J., Halpern, D. F., Loehlin, J. C., Swayze, V., II, & O’Leary, D. S. (1993). The neuro- Perloff, R., Sternberg, R. J., & Urbina, S. (1996). psychology of the prefrontal cortex. Archives of Intelligence: Knowns and unknowns. American Neurology, 50 , 636–642. doi: 10.1001/archneur.1993. Psychologist, 51(2), 77. doi: 10.1037/0003-066X.51.2.77 . 00540060066020 . Neylan, T. C. (1999). Frontal lobe function: Mr. Phineas Roberts, A. C., Robbins, T. W., & Weiskrantz, L. (1998). Gage’s famous injury. The Journal of Neuropsychiatry The prefrontal cortex: Executive and cognitive func- and Clinical Neurosciences, 11 (2), 280–281. tions . Oxford: Oxford University Press. Niendam, T. A., Laird, A. R., Ray, K. L., Dean, Y. M., Roca, M., Parr, A., Thompson, R., Woolgar, A., Torralva, T., Glahn, D. C., & Carter, C. S. (2012). Meta-analytic Antoun, N., Manes, F., & Duncan, J. (2010). Executive evidence for a superordinate cognitive control network function and fl uid intelligence after frontal lobe lesions. subserving diverse executive functions. Cognitive, Brain, 133 (1), 234–247. doi:10.1093/brain/awp269 . Affective, & Behavioral Neuroscience, 12 (2), 241– Roca, M., Manes, F., Chade, A., Gleichgerrcht, E., 268. doi: 10.3758/s13415-011-0083-5 . Gershanik, O., Arévalo, G. G., Torralva, T., & Duncan, Njemanze, P. C. (2005). Cerebral lateralization and gen- J. (2012). The relationship between executive func- eral intelligence: Gender differences in a transcranial tions and fl uid intelligence in Parkinson’s disease. Doppler study. Brain and Language, 92 , 234–239. Psychological Medicine, 42 (11), 2445. doi: 10.1017/ doi: 10.1016/j.bandl.2004.06.104 . S0033291712000451 . Osório, A., Cruz, R., Sampaio, A., Garayzábal, E., Roca, M., Manes, F., Gleichgerrcht, E., Watson, P., Ibáñez, Martínez-Regueiro, R., Gonçalves, Ó. F., Carracedo, A., Thompson, R., Torralva, T., & Duncan, J. (2013). Á., & Fernández-Prieto, M. (2012). How executive Intelligence and executive functions in frontotemporal functions are related to intelligence in Williams syn- dementia. Neuropsychologia, 51 (4), 725–730. drome. Research in Developmental Disabilities, 33 (4), doi: 10.1016/j.neuropsychologia.2013.01.008 . 1169–1175. doi: 10.1016/j.ridd.2012.02.003 . Sadeh, S. S., Burns, M. K., & Sullivan, A. L. (2012). Parkin, A. J., & Java, R. I. (1999). Deterioration of frontal Examining an executive function rating scale as a pre- lobe function in normal aging: Infl uences of fl uid dictor of achievement in children at risk for behavior intelligence versus perceptual speed. Neuropsychology, problems. School Psychology Quarterly, 27 (4), 236. 13 (4), 539. doi: 10.1037/0894-4105.13.4.539 . doi: 10.1037/spq0000012 . Pennington, B. F. (1997). Dimensions of executive func- Salthouse, T. A. (2001). Structural models of the relations in normal and abnormal development. In tions between age and measures of cognitive func- N. Krasnegor, R. Lyon, & P. Goldman-Rakic (Eds.), tioning. Intelligence, 29, 93–115. doi: 10.1016/ Development of the prefrontal cortex: Evolution, neu- S0160-2896(00)00040-4 . robiology, and behavior (pp. 265–281). Baltimore: Salthouse, T. A. (2011). Neuroanatomical substrates of Paul H. Brookes. age-related cognitive decline. Psychological Bulletin, Pennington, B. F., Bennetto, L., McAleer, O., & Roberts, 137 (5), 753. doi: 10.1037/a0023262 . R. J. (1996). Executive functions and working mem- Salthouse, T. A., & Pink, J. E. (2008). Why is working memory. Theoretical and measurement issues. In G. R. ory related to fl uid intelligence? Psychonomic Bulletin & Lyon & N. A. Krasnegor (Eds.), Attention, memory, Review, 15 (2), 364–371. doi: 10.3758/PBR.15.2.364 . and executive function (pp. 327–348). Baltimore: Saver, J. L., & Damasio, A. R. (1991). Preserved access P.H. Brookes. and processing of social knowledge in a patient with Prabhakaran, V., Smith, J. A., Desmond, J. E., Glover, acquired sociopathy due to ventromedial frontal dam- G. H., & Gabrieli, J. D. (1997). Neural substrates of age. Neuropsychologia, 29 (12), 1241–1249. fl uid reasoning: An fMRI study of neocortical activa- doi: 10.1016/0028-3932(91)90037-9 . tion during performance of the Raven’s progressive Schellenberg, E. G. (2011). Examining the associa- matrices test. Cognitive Psychology, 33 (1), 43–63. tion between music lessons and intelligence. doi: 10.1006/cogp.1997.0659 . British Journal of Psychology, 102(3), 283–302. Ptito, A., Crane, J., Leonard, G., Amsel, R., & Caramanos, doi:10.1111/j.2044-8295.2010.02000.x . Z. (1995). Visual-spatial localization by patients with Schlinger, H. D. (2003). The myth of intelligence. The frontal lobe lesions invading or sparing area 46. Psychological Record, 53 (1), 2. Neuroreport, 6 , 45–48. Schretlen, D., Pearlson, G. D., Anthony, J. C., Aylward, Ramnani, N., & Owen, A. M. (2004). Anterior prefrontal E. H., Augustine, A. M., Davis, A., & Barta, P. (2000). cortex: Insights into function from anatomy and neu- Elucidating the contributions of processing speed, roimaging. Nature Reviews Neuroscience, 5 , 184–194. executive ability, and frontal lobe volume to normal doi: 10.1038/nrn1343 . age-related differences in fl uid intelligence. Journal Raven, J. C. (1938). Guide to progressive matrices . of the International Neuropsychological Society, 6(1), London: H. K. Lewis. 52–61. doi: 10.1017/S1355617700611062 . Raven, J. C., Court, J. H., & Raven, J. (1988). Manual for Shallice, T., & Burgess, P. W. (1991). Defi cits in strategy Raven’s progressive matrices and vocabulary scales . application following frontal lobe damage in man. London: H. K. Lewis. Brain, 114 , 727–741. doi: 10.1093/brain/114.2.727 . 458 E.C. Duggan and M.A. Garcia-Barrera

Snyderman, M., & Rothman, S. (1987). Survey of Tranel, D., Manzel, K., & Anderson, S. W. (2008). Is the expert opinion on intelligence and aptitude prefrontal cortex important for fl uid intelligence?: testing. American Psychologist, 42(2), 137. doi: A neuropsychological study using matrix reasoning. 10.1037/0003-066X.42.2.137 . The Clinical Neuropsychologist, 22 (2), 242–261. Spearman, C. (1904). “General intelligence,” objectively doi: 10.1080/13854040701218410 . determined and measured. The American Journal of Tsuchida, A., & Fellows, L. K. (2009). Lesion evidence Psychology, 15 (2), 201–292. that two distinct regions within prefrontal cortex are Spearman, C. (1927). The abilities of man . London: critical for n-back performance in humans. Journal of Macmillan. Cognitive Neuroscience, 12, 2263–2275. doi: 10.1162/ Stanovich, K. E., Siegel, L. S., & Gottardo, A. (1997). jocn.2008.21172 . Converging evidence for phonological and surface Tsujimoto, S. (2008). The prefrontal cortex: Functional neural subtypes of reading disability. Journal of Educational development during early childhood. The Neuroscientist, Psychology, 89 , 114–127. doi: 10.1037/0022-0663. 14(4), 345–358. doi: 10.1177/1073858408316002 . 89.1.114 . van der Maas, H. L., Dolan, C. V., Grasman, R. P., Sternberg, R. J. (1985). Beyond IQ: A triarchic theory of Wicherts, J. M., Huizenga, H. M., & Raijmakers, human intelligence. New York: Cambridge University M. E. (2006). A dynamical model of general intelli- Press. gence: The positive manifold of intelligence by mutu- Sternberg, R. J. (2000). The concept of intelligence. In alism. Psychological Review, 113 (4), 842. R. J. Sternberg (Ed.), Handbook of intelligence doi: 10.1037/0033-295X.113.4.842 . (pp. 3–15). New York: Cambridge University Press. Volle, E., Kinkingnéhun, S., Pochon, J., Mondon, K., de Sternberg, R. J., & Detterman, D. K. (Eds.). (1986). What Schotten, M. T., Seassau, M., Duffau, H., Samson, Y., is intelligence? Contemporary viewpoints on its Dubois, B., & Levy, R. (2008). The functional archi- nature and defi nition . Norwood: Ablex Publishing tecture of the left posterior and lateral prefrontal cor- Corporation. tex in humans. Cerebral Cortex, 18 , 1093–1103. Stoddard, G. D. (1943). The meaning of intelligence . doi: 10.1093/cercor/bhn010 . New York: Macmillan. Wager, T. D., & Smith, E. E. (2003). Neuroimaging stud- Stuss, D. T., & Benson, D. F. (1984). Neuropsychological ies of working memory. Cognitive, Affective, & studies of the frontal lobes. Psychological Bulletin, Behavioral Neuroscience, 3 (4), 255–274. doi: 10.3758/ 95 (1), 3. doi: 10.1037/0033-2909.95.1.3 . CABN.3.4.255 . Stuss, D. T., Benson, D. F., Kaplan, E. F., Weir, W. S., Wechsler, D. (1958). The measurement and appraisal of Naeser, M. A., Lieberman, I., & Ferrill, D. (1983). adult intelligence (4th ed.). Baltimore: Williams & The involvement of orbitofrontal cerebrum in cog- Wilkins. nitive tasks. Neuropsychologia, 21, 235–248. Weinstein, S., & Teuber, H. L. (1957). Effects of penetrat- doi:10.1016/0028-3932(83)90040-4 . ing brain injury on intelligence test scores. Science, Swanson, H. L. (1999). Reading comprehension and 125 , 249–263. doi: 10.1126/science.125.3256.1036 . working memory in learning-disabled readers: Is the Wickett, J. C., Vernon, P. A., & Lee, D. H. (2000). phonological loop more important than the executive Relationships between factors of intelligence and brain system? Journal of Experimental Child Psychology, volume. Personality and Individual Differences, 29 (6), 72 , 1–31. doi: 10.1006/jecp.1998.2477 . 1095–1122. doi:10.1016/S0191-8869(99)00258-5 . Terman, L. M. (1921). A symposium: Intelligence and its Wiebe, S. A., Sheffi eld, T., Nelson, J. M., Clark, C. A., measurement. Journal of Educational Psychology, 12 , Chevalier, N., & Espy, K. A. (2011). The structure of 127–133. doi: 10.1037/h0064940 . executive function in 3-year-olds. Journal of Teuber, H. L. (1959). Some alterations in behavior after Experimental Child Psychology, 108 (3), 436–452. cerebral lesion in man. In A. D. Bass (Ed.), Evolution doi: 10.1016/j.jecp.2010.08.008 . of nervous control from primitive organisms to man Wise, S. P., Murray, E. A., & Gerfen, C. R. (1996). The (pp. 157–194). Washington, DC: America Association frontal cortex-basal ganglia system in primates. for the Advancement of Science. Critical Reviews in Neurobiology, 10 , 317–356. Teuber, H. L. (1964). The riddle of frontal lobe function in doi: 10.1615/CritRevNeurobiol.v10.i3-4.30 . man. In J. M. Warren & K. Akert (Eds.), The frontal Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). granular cortex and behavior (pp. 410–444). Woodcock-Johnson III tests of cognitive abilities . New York: McGraw-Hill. Itasca: Riverside Publishing. Thomson, G. H. (1951). The factorial analysis of human Woolgar, A., Parr, A., Cusack, R., Thompson, R., Nimmo- ability (5th ed.). London: University of London Press. Smith, I., Torralva, T., Roca, M., Antoun, N., Manes, F., Toga, A. W., & Thompson, P. M. (2005). Genetics of brain & Duncan, J. (2010). Fluid intelligence loss linked to structure and intelligence. Annual Review of restricted regions of damage within frontal and parietal Neuroscience, 28 , 1–23. doi: 10.1146/annurev.neuro. cortex. Proceedings of the National Academy of Sciences, 28.061604.135655 . 107 (33), 14899–14902. doi: 10.1073/pnas.1007928107 . The Evolution of Intelligence: Implications for Educational 28 Programming and Policy

Christopher Jones , Peggy L. Tarpley , and Douglas Blancero

When the opportunity arose to write this chapter, school’s academic community function, grow, we felt it would be a great opportunity to write succeed, or fail. about why teachers often reject changing based on research-guided knowledge, in this case specifi cally about intelligence but broadly about Carnine Revisited education. As a team of academics, practitioners, and coaches, we have worked in schools for over Educators many times seem reluctant to embrace 25 years. The focus on the rejection of research research diametrically opposed to their opinions as primary to what we have observed in schools or previous learning. Understanding the complex over time was an easy one. The issue is complex relationship between teaching and learning is with tendrils that reach out and affect all aspects often convoluted. Many teachers work incredibly of school reform. hard to educate children yet produce less than Over the years we have identifi ed several fac- stellar results. In education, we can often fi nd tors affecting the growth and improvement of excuses why children do not learn that are not schools. This chapter addresses several of them. aligned with the instruction. Avoiding culpability A common thread running through all of these insulates us from internal and external criticism factors is change. How we, the academics we, and allows us to continue to use ineffective meth- the academics and trainers/coaches, and the ods. While we were having extensive conversa- schools, the administrators, teachers, students, tions on this very topic, we wondered what it and families perceive and accept change affects would take to get teaching to become more of a how both a school and the members of the profession that valued research. After a lot of discussion on how to approach this chapter, we C. Jones decided to revisit Douglas Carnine’s refl ections Longwood University, Hull Building , on our profession. 201 High Street , Farmville , VA 23909 , USA During the winter of 2000, Douglas Carnine P. L. Tarpley wrote an article titled, Why education experts Department of Education and Special Education , resist effective practices (And what it would take Longwood University , 201 High Street , to make education more like medicine) . In his Farmville , VA 23909 , USA article he lamented about the teaching profes- * D. Blancero ( ) sion. He questioned why teachers embrace JP Associates , 7 Amherst Street , Red Hook , NY 12571 , USA methods that do not work. He wondered why the e-mail: [email protected] fi ndings of the national reading panel were still

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 459 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_28, © Springer Science+Business Media New York 2015 460 C. Jones et al. excluded from practice in many K-3 classrooms. and propose possible treatments to modify this He was also critical of college professors who characteristic, we are afraid the previous chapters are charged with the task of producing effective in this text might not have an impact on student teachers. He included Diane Ravitch’s trip to the learning. intensive care unit and her visions of what her Casting blame, devoid of any refl ection, on healthcare would be like if the physicians used how we got to this position will not result in an the same methodology that education professors increase in student academic performance or an used. Carnine called on our profession to increase in educators embracing evidence. It is embrace research at an individual teacher level. more important for educators to refl ect on how we It seems like it is a good time to revisit Carnine’s got here rather than who is at fault. Refl ecting on article to see how far we have come. It has been the original purpose of education is as good a place 15 years since Carnine’s article, and we have as any to start. Thomas Jefferson proclaimed: seen many reforms come and go with mixed Now let us see what the present primary schools results (e.g., Reading First; Reading First cost us, on the supposition that all the children of Implementation Evaluation: Final Report 2008 ). 10, 11 and 12 years old are, as they ought to be, at We have watched our science, reading, and math school: and, if they are not, so much the work is the system; for they will be untaught, and their igno- scores stay fl at, while other nations thrive rance & vices will, in future life cost us much ( http://www.oecd.org/pisa/ ). We now see new dearer in their consequences, than it would have initiatives, Race To the Top, aligned with new done, in their correction, by a good education standards (Common Core) with similar goals. (Ford 1892 –99). But the question continues to trouble us, why School is an intervention for ignorance. If our are we seeing academic success in other coun- primary intervention for ignorance is ineffective, tries, yet we continue to struggle with literacy then we have to modify it. and basic math skills in the United States? Has the United States reached their plateau with regard to academic performance or is it some- Shut Down the Schools? thing else? Is our lack of student achievement aligned more with the student variables or It has been 50 years since the verdict was passed teacher variables? If our academic stagnation is down on Brown versus Board of education. But caused by student variables, then indeed we the ramifi cations for school systems rejecting the might be doing as well as we can. court’s fi nding are still felt by many students These are tough questions that can possibly be who, through no fault of their own, experienced answered by historians, preachers, economists, very little or no education for over 7 years due to and a number of other groups. We take the posi- the closing of certain public schools. During this tion that instead of placing the primary blame on volatile time, many students in Prince Edward outside variables, maybe we should fi rst examine County, Virginia, received their education in the disposition of our teachers and their imple- church basements, private homes, or other mentation of best practices. makeshift facilities, while some just dropped out It is our opinion that we have not reached our of school. While it is not the focus of this chapter pinnacle. It is our belief that our stagnant aca- to revisit those dark times, it is important to demic performance is in part a by-product of the determine what was lost by those students who rejection of evidence-based practices. It may be dropped out. Conventional wisdom would say tempting to wag one’s disapproving fi nger at that they lost their opportunity; they lost their teachers and yell “Shame on you teachers,” but it chance to be competitive. Would our history be might be more productive to open a dialogue very different if our educators focused on the focusing on possible reasons why there is a rejec- evidence of effective teaching and Jefferson’s tion of the evidence in our fi eld. If we do not warnings about “future life cost” rather than pol- identify the causation of our rejection of evidence itics and fear? 28 The Evolution of Intelligence: Implications for Educational Programming and Policy 461

It is true that many believed, and still do, that family background and that school reform had it is a child’s history, upbringing, or genetics that little or no infl uence on student achievement. is the indicator of their future academic success. While subsequent research has validated the Many educators, at least intellectually, reject the importance of an effective teacher using evidence- notion of academic predestination. If education based procedures (Hattie 2009; Marzano 2003 ), is indeed the great equalizer (Mann 1848 ), then the importance of education will still be called it is safe to say that many of the students’ lives into question until we quantify and disseminate were made much more diffi cult because of the the value of an effi cient and effective teacher politics that resulted in the closing of schools. using best practices. Aside from politics and background, we must Fortunately, a great deal of research shows have evidence that highlights the importance of that schools and more specifi cally teachers have effective instruction. a signifi cant infl uence on student achievement. Hattie ( 2009 ) has shown that contributions Nye et al. ( 2004 ) reported that students who from the teacher, school, and curricula have a receive instruction from highly effective teach- signifi cant impact on student learning. Clearly, ers score between 13 and 18 percentile points how a child gets educated has a greater impact higher in reading and math. Marzano ( 2003 ) than the variables they bring to school. Lemov found that what goes on in the classroom ( 2010) identifi ed 49 techniques that improve stu- accounted for 20 % of the variance in student dent achievement that teachers use. We know achievement which was twice as much as the that if we have excellent delivery of phonics Coleman study (Marzano 2003). Hattie’s ( 2009 ) instruction, we will have a greater impact on lit- synthesis of over 800 meta-analyses relating to eracy skill development (Hattie 2009; National achievement reported the variables that produce Reading Panel 2000 ). If we combine systematic many of the greatest effect sizes that were explicit phonics instruction (d = 0.6) with deliv- teacher-related variables (e.g., providing forma- ery techniques that support student achievement tive evaluation: ES 0.90, acceleration: ES 0.88, (Lemov 2010), then we can positively impact microteaching: ES 0.88, teacher will clarity: ES student literacy achievement. And while this is 0.75, teachers challenging students: ES 0.66, all very encouraging, remember Chall reported high expectations: ES 0.53). Clearly, teachers the same fi ndings in 1967. One has to wonder using best practices can invoke great change in that if we were a profession focused more on student performance. Jefferson’s vision and the literacy achievement School reform should be a relatively simple evidence, would we have locked the doors in task. All we have to do is identify those variables Prince Edward County? that produce the greatest gains, replicate them in classrooms across the United States, and we should see dramatic increases in student perfor- The Importance of School mance. It just does not seem that diffi cult. Within a matter of months, we should be able to intro- It is safe to say that as long as we continue to duce into the classrooms those variables that value opinions, politics, and rhetoric over evi- have the greatest chance for student improve- dence, the fi ndings of Coleman and Jencks will ment and provide teachers with specialized pro- continue to infl uence our teachers. Coleman et al. fessional development that builds their ( 1966 ) asserted that student’s academic success competencies in best practices. We can then pro- is based on their background and social standing. vide evidence of our success through ongoing While Jencks et al. (1972 ) mirrored many of formative progress monitoring. Coleman’s fi ndings by stating that schools did Clearly, this has not been the case, so where very little to lessen the achievement gap between did the process break down? If we know that cer- the haves and the have-nots, Jencks also reported tain variables produce greater student achieve- that student achievement was a by-product of ment, then those variables should be taught in 462 C. Jones et al. higher education teacher preparation programs two approaches was passionate and colorful but and during the in-school professional develop- often failed to cite the empirical evidence that sup- ment training sessions. Unfortunately, this line of ports their use. Often the selection of an approach logic is often broken. There is a disconnect some- came down to what was taught to that individual times between the needs of the school divisions/ teacher at the university and college level. These districts, the professional development offered to debates went on for decades and succeeded in the teachers (we are not yet even considering the aligning teachers into different camps. Professors additional aspect of “how” we assess, analyze, from higher education programs that prepared and identify what professional development is teachers often found themselves in a position of needed or how it should be delivered in the school having to choose which approach they were going setting—stand up training, coaching, etc.), and to use in their programs. The choice often was what is taught in teacher preparation. In order to based on the professor’s preferences, previous rectify this, we have seen lucrative grants such as institution instructional practices, or current Reading First attempt to impose Scientifi cally trends. It was not until the National Reading Panel Based Reading Research (SBRR) paired with identifi ed the fi ve major components of good solid reading “Coaches.” These types of grants pro- reading instruction (phonemic awareness, alpha- duced, at best, mixed results (Reading First betic principle, fl uency with text, vocabulary, com- Implementation Evaluation: Final Report 2008 ). prehension) that we began to embrace the need for On paper Reading First seemed to have a real good solid systematic explicit phonics instruction. chance to make signifi cant progress on the war Or did we? Even in the face of the massive NRP against illiteracy. The grants required the use of fi nding, we still saw a proliferation of constructiv- SBRR and brought reading coaches into the build- ists’ literacy programs. It seems that when we get ing to assist teachers with their implementation of close to accepting an approach as having evidence the SBRR. While there are many variables that to support its use, we repackage old unproven caused problems with the fi nal results of Reading methods and present them as if they were some- First, one was the top-down approach to having thing new. The repackaging of unproven methods teachers implement SBRR. Many of the methods fulfi lls the need to try something new but rarely included in the SBRR stood in stark contrast with produces desired results. It appears that simply the preservice training teachers received. mandating best practices is not going to work. Seasoned teachers also rejected the SBRR on Teachers must enter the fi eld with an understand- grounds of autonomy or opinion. It became appar- ing and skill set that allows them to interpret what ent that each individual teacher must understand constitutes best practices. the nature of evidence-based practices and their infl uences on student achievement. If this condition is not met, then the intervention will not be What Is Done in Teacher successful. It became obvious that in order for Preparation Matters best practices to be embraced by many teachers, the teachers needed to value research. Great atrocities were often committed under noble pretenses. When educators make curricular decisions based on affect, opinion, or tradition, Teachers Who Value Research they are bound to make catastrophic mistakes. And even though many individuals do not Whole Language Versus Phonics feel like they are committing atrocities, generation after generation of at-risk students who Nothing evokes more heated dialogue in education do not possess basic reading skills or basic than the reading wars. Chall and Adams advocated mathematics skills is a testament to our failure. for phonics approaches, while Chomsky and Sending teachers into the fi eld who are not well Goodman advocated for more whole language versed in best practices but who possess a roman- methods Chall (1996 ). The rhetoric behind these tic idea of what constitutes good solid instruction 28 The Evolution of Intelligence: Implications for Educational Programming and Policy 463 is going to perpetuate continued student academic As a profession, when we reject evidence- failures. The simple truth is that many of our based practices, we fail those children who are struggling children will not benefi t from the meth- initially excited, optimistic, and want to learn. ods aligned with romantic ideas of what our And while children come to us with huge aca- teacher candidates consider good solid instruc- demic defi cits, with the proper instruction, they tion. Good intentions aside, the instructional can catch up, but if that child is taught by an edu- choices that we are making contribute to the con- cator who does not value research, it is a catastro- tinued underperformance of at-risk students. We phe waiting to happen. have a tendency to blame variables beyond our control for the student’s academic failures. Excuses such as their home environment or social Art of Teaching Versus Science economic status, clearly these have an infl uence of Teaching on their achievement, but these are variables beyond our control. The professional teacher Carnine (2000 ) wrote that teacher preparation would be better off focusing on variables within programs have a tremendous infl uence on teacher their control or infl uence. The most important development. They are incredibly infl uential in variable a teacher has control over is their instruc- the professional organizations that invoke aca- tion. But there is an appeal for attributing student demic policies. There was a time when teacher failure to variables beyond the teacher’s control or education programs were justifi ed in their inde- infl uence, and that is personal accountability. pendent approaches. Our teachers taught students Attributing student academic failure to variables who excelled on measures of science, reading, beyond the individual teacher’s control often and math. Our student’s performance on these eliminates the need for instructional adaptation or measures justifi ed what we were doing in the change. One of the problems with aligning stu- schools and in turn what we were doing in higher dent failure with variables beyond the teacher’s education. And while we continue to produce control is it legitimizes student failure. Even if the some of the world’s great minds, recent reports outside infl uences were correctly identifi ed as indicate that the United States’ academic perfor- reasons for academic failure, causation is not an mance in reading, science, and mathematics is excuse for continued academic failure. If a child fl at compared to students in Shanghai, Singapore, comes to us with an oral language defi cit caused and many other Asian providences (PISA— by their lack of verbal interactions, it is better for Organization for economic co-operation and us to seek evidence-based methods that assist the development http://www.oecd.org/pisa/ ). teacher in building the student’s oral language It is safe to say that a country’s future rests on vocabulary than to use their background as an their ability to produce students who can compete excuse for academic failure. Understanding the in a global economy. If this conventional wisdom cause of the student’s oral language defi cit does is true, the United States is at great risk for losing not let us off the hook. We still need to address the their competitive edge. Some individuals would child’s oral language defi cits and phonological question the importance or even the validity of awareness defi cits prior to beginning their reading these assessments. Many times these arguments instruction. The teacher who is well versed in only confuse teachers. And while arguments for research will know the methods that will effec- and against assessment spark great debate, we tively address the student’s academic weakness. cannot deny the staggering number of illiterate The teacher who is not well versed, or is more and undereducated students in the United States. aligned with the art of teaching rather than the sci- The drain on our economy and on our way of life ence of teaching, could continue to provide is undeniably apparent. The US Department of instruction that does not meet the student’s Education’s National Institute of Literacy in 2013 academic needs. The sad truth is we often deny found 14 % of the adults in the United States effective interventions either through our own could not read; that is over 32 million people. The ignorance, disposition, or just plain arrogance. number jumps to 21 % of the population who 464 C. Jones et al. reads below a fi fth grade level. All one has to do What are we willing to do to remain competi- to see the damaging effects of literacy is to enter tive with other countries and address the needs any prison and view a 65 % illiteracy rate. The of our most at-risk students? Our understanding true challenge is that in spite of massive reform of teaching sometimes convinces us that if we movements, the National Assessment of Adult can make this content interesting and lively, then Literacy has found no signifi cant change in liter- it will be effective. And while entertainment is acy scores for over a decade (http://nces.ed.gov/ defi nitely an element of teaching, it is not the naal/kf_demographics.asp). whole of teaching. The perception that a good How do we justify the stagnant growth when teacher knows intuitively if their students are we have a wealth of research that validates best learning confl icts with the need for accurate practices. One possible explanation is that we progress monitoring. reject that research in favor of the art of teaching. We want instruction to be lively and engaging, The art of teaching often becomes more impor- and yes a teacher can gauge the general perfor- tant than the science of teaching. Stanovich and mance of their students in some cases without Stanovich (2003 ) argued that teachers become assessments. But if these techniques were much more effective if their skill level in identi- enough, why do we continue to have students fying evidence-based practices is fully devel- lagging behind their peers. If teachers instinc- oped. Pearson (1999 ) agreed with Carnine that tively know how their students are performing education needs to be more in line with the medi- and what constitutes effective instruction, then cal model. It is our professional obligation to what reason do we have for student failure? Are utilize best practices that originate from valid teachers turning a blind eye to those students who research. If teachers want to support teacher are underperforming in their classroom? We have autonomy, it is imperative that they earn this priv- to believe that this is not the case. The more likely ilege by consistently aligning their pedagogy explanation is that teachers simply do not know based on evolving research. It is this exact method what to do when traditional methods are ineffec- that elevated the public’s respect for individuals tive. As a result, they continue to use ineffective in the fi eld of medicine. Unfortunately, many in methods hoping for a different result. And while the fi eld of education seem to reject empirical it is easy to blame the teachers for their skill defi - research preferring evidence from personal cits, higher education programs must own their observations. While clearly anecdotal reports accountability for teacher defi cits in understand- have their place in assessment, the overreliance ing what constitutes effective instruction. on qualitative data often results in a rejection of best practices (Pearson 1999 ). For example, Beck (1998 ) observed that many proponents of whole Professors Unchecked language when faced with the fi ndings of the national reading panel continue to reject system- Within Checks atic phonics instruction in favor of whole language. They did this with a clear conscience Professors of higher education often emerge from because they observed many students coming their Ph.D. program with a belief that they are the into their classroom who did not need explicit unchecked experts in their classrooms. They pos- instruction in the alphabetic principle. They sess a body of knowledge that has been infl uenced noticed that the students who were from print- rich by other professors, their experience, and their environment settings had highly developed liter- course work. While many Ph.D. programs in the acy skills. They believed that all one had to do fi eld of teacher education lay a solid pedagogical was to replicate the print-rich environment, and and theoretical knowledge base, it is imperative students would become literate. It is not their that teacher preparation program professors con- motivation and effort that comes into question, it sistently validate and update their instruction is their rejection of the research of best practices based on evidence. Teacher education professors that hampers student progress. are generally hard workers, but it is the lack of 28 The Evolution of Intelligence: Implications for Educational Programming and Policy 465 opportunities for internal and external feedback improvement specialists are highly experienced that could prevent them from relying more on and can work with a wide range of schools. By evidence-based methods. If education professors providing on-site professional development, they fail to overtly seek feedback, then they will be can have a direct impact on student learning and limited by their shortcomings. All one has to do is often collect much of the evidence that supports to look at the evaluation procedures established best practices (JP Associates website http: www. by respected professions, and one sees the need jponline.com ). It would seem that a partnership for sustained purposeful feedback. The fi eld of between school divisions, the on-site profes- medicine provides us with an example where sional development providers, and teacher prepa- multiple parties provide the checks and balances ration programs would benefi t the teacher needed to maintain a high level of effectiveness. education candidate. We might wonder what it would be like if our teacher preparation programs were more like the medical residency models. External Checks Within the residency model, not only do the candidates benefi t from professors’ instruction, but Doctors are exposed to various checks and bal- they also receive feedback from nurses, patient ances embedded within their profession. A doc- advocates, family members, and a host of other tor’s work can be evaluated by internal review interested parties. And while the sheer volume of boards, patients, and even lawyers. Those checks feedback could be overwhelming, it pushes the and balances, often absent in higher education, interns to know their craft. We can only imagine push those in the medical fi eld to be accountable the positive impact on teacher education if school for their decision. Many times education profes- divisions, teacher preparation programs, and on- sors receive very little feedback from their col- site professional development organizations all leagues and even less from students. work together in educating the future teacher. Unfortunately, if a college professor in teacher preparation programs is receiving very little feedback from their colleagues, then it is also Feedback likely that they are receiving very little feedback as to the effectiveness of their instruction from Actively seeking feedback is an indication that external parties. It is very rare to see a college one wants to perfect their abilities to teach. professor actively seeking feedback on their Reciprocal feedback, where we not only give classroom instruction from individuals who feedback on projects, tests, responses, etc., but work directly with teachers once they are in the seek feedback on the effectiveness of our meth- classroom. This lack of communication with ods is critical. Having the confi dence to actively professional development providers and school seek feedback brings you that much closer to systems prevents professors from receiving very valuing evidence. valuable information as to the quality of their A true professional, who is open to feedback, instruction and program. is seeking the truth. If they are interested in how Professional development providers have a effective their instruction is, then they will be unique opportunity to not only become well equally interested in what works. Wanting to versed in evidence-based procedures, but they know what works and being open to others’ can also see the impact of their professional empirically based ideas will push individuals to development session on individual classroom embrace evidence-based instruction. teachers. Many of these organizations have a fi rm If teacher education professors embrace the understanding of what constitutes good solid concepts associated with evidence-based instruction. In higher education we often do not instruction, they will model these concepts to the forge these very lucrative partnerships between teacher candidate. Hattie ( 2009) reported an teacher preparation programs and professional effect size of 0.77 for feedback. Teacher to stu- development service providers. Many of these dent feedback is important, but equally as impor- 466 C. Jones et al. tant is feedback from student to teacher. In fact, model of school improvement that is going to be some of the greatest gains in student achieve- useful to schools must focus explicitly on results, ment occur when students provide feedback to evidence of student learning, and student achieve- their teachers (Hattie 2009). And while teachers ment” (Assembly Required: A Continuous School report that they actively solicit feedback from Improvement System, p. 5, Lezotte). And later on their students, that is not always the case. in the same chapter, he writes: When asked if teachers accept feedback from …the system in place is ideally suited to producing students, 70 % of the teachers say they accept it, the results the school is currently getting … any whereas only 45 % of the students say the teach- change in the desired results, from the current sys- ers accept their feedback (Carless 2006 ). tem in place is going to require a change in mission, core beliefs, and core values that underpin the Students’ perception of the acceptance of their system, especially if the goal is to permanently feedback is critical. Students are recipients of sustain the desired change. (p. # ) that instruction, and their feedback is a good indicator of how effective the instruction is. When There are at the least two main aspects of professors actively seek student feedback, they change—accepting the need for change and man- can obtain a fairly accurate gauge as to the effec- aging the change. When we ask people, in this tiveness of their instruction. More importantly case teachers, to change what they are doing, they instill a belief in the teacher candidates that even when it is based on data, it is often perceived feedback is valuable. We believe teachers who as a challenge to their personal belief, the hard value feedback are more open to research. work they have invested. Sometimes it is the When we venture into areas that we are not actual act of changing that can be the challenge sure of, areas that are beyond our fi eld of exper- as much as what the change is. tise, we are much more likely to become defen- At one point in our work with schools, our sive if our assertions are challenged. But it is approach was to work with teachers, individually exactly those times we should be actively seeking and separate from leadership. We felt if we feedback. Unfortunately, it is very tempting to changed what was happening in the classroom, pretend and hope that individuals will not dis- we could change the school. It soon became evi- cover our weaknesses. What we fail to realize is dent that unless the people of that school con- that by modeling our search for information to nected and shared, unless the leader was clarify meaning, we instill in our candidates that providing a clear vision and supporting that they should actively seek feedback in times of vision effectively, then change was slow, or in the uncertainty. We should work with teachers to worse cases, change was resisted and repelled. actively investigate unfamiliar concepts. Changing the culture of the building is essen- Sometimes our narrow vision of professionalism tial to shifting mind-sets from the current system prevents us from seeking feedback. If seeking in place to one that will produce increased feedback is one of the fi rst steps in embracing success. Michael Fullan calls this reculturing. evidence, what would it take to get professionals This reculturing needs to occur on all levels to actively seek feedback? (higher education, schools, and classrooms). It calls for both a strong foundation in effective instruction (looking at the data) and a transfor- Change mational approach (relationships and sharing knowledge). At the beginning of this chapter, we briefl y addressed change and spoke of the need for an acceptance of research. This might be a good time Conclusions to take a closer look at change and its role in organizational growth, educator acceptance of research, School reform should be a relatively simple task, and higher education changing its approach to but recent history indicates otherwise. Throughout teacher prep. Lawrence Lezotte tells us, “Any this chapter we have identifi ed certain traits that 28 The Evolution of Intelligence: Implications for Educational Programming and Policy 467 prevent teachers from embracing evidence with the solace in the excuses that lack of educational prog- result of limiting our understanding of intelligence ress is somehow the fault of past teachers, the fam- and intellectual processes thereby fl attening math, ily, or the child himself or herself. Without these science, and reading scores for students. Educators excuses, teachers and administrators are smack in are often reluctant to embrace research that does the midst of accountability or culpability. not support their personal approach to instruction. Carnine (2000 ) writes about the pressure on Stagnant academic performance is a by-product of education from all sides. He suggests that all pro- the rejection of evidenced-based practices. This fessions have been at this stage of growth toward resistance to accept sound research takes on even a mature profession. The recipients of the less more importance when research indicates that con- than adequate treatment (whatever profession it tributions from the teacher, school, and curricula may be) and their advocates call for the incorpo- have a signifi cant impact on student learning. ration of more scientifi c methodology in the pro- It is apparent that it is not one individual at fession. He states further, “A mature profession, fault; rather it is a collective failure of education. It by contrast, is characterized by a shift from judg- is due, in part, to our tendency to blame variables ments of individual experts to judgments con- beyond our control for the student’s academic fail- strained by quantifi ed data that can be inspected ures. The variable that is within our control is by a broad audience…” (p. 12). While all, pre- teacher preparation, ongoing professional devel- and in-service teachers, teacher education fac- opment, and effective leadership that manages and ulty, school administrators, and professional leads the change necessary for success. development providers, contribute to the resis- We should instill in the preservice teacher the tance to education based on research, they also notion that we must work as a team to address hold promise to the acceptance and implementa- student achievement issues. Professors should be tion of research-based practices. part of a collaborative group that includes school divisions, professional development providers, and teacher preparation programs. Preservice Model for Success teachers need to actively seek feedback and embrace evidence-based practices. Teachers need We need a partnership between school divisions, to actively seek evidence that supports the meth- on-site professional development providers, and ods that they are using. They should be collecting teacher preparation programs based on the medi- formative data that supports the continued use of cal residency model where teacher professional these methods. If the data shows that students are development is a natural extension of teacher not responding to the intervention, then, the preparation. We learn best by doing. On-site pro- teacher should modify that intervention. For fessional development with both leadership and teachers to possess these skills, teacher teachers as they are working through their normal preparation professors must model and validate day can provide needed constructive feedback the use of evidence in their instruction. and guidance options that might not be seen by To summarize, teacher rejection of those in the “thick of the forest.” It can be the research- based ideas and practices impacting edu- training and proving ground for effective prac- cational decisions for students comes about because tices. Here in the fi eld, each individual teacher of several constructs. One, teachers may fi nd that can participate in evidence-based practices and the research clashes with their opinion or anecdotal observe the infl uences on student achievement. In observations. Further, this clash may come with the this way teachers, coaches/professional teacher’s past learning or teacher preparation pro- development providers, and administration are gram. Two, resistance may come because teachers all contributing to the evidence of best practices. and administrators fi nd the discomfort of changing A change in how schools work needs to be from what has always been done requires signifi - explored. A shift from simply managing schools cant effort and friction. Finally, educators fi nd to leading schools is essential. There needs to be 468 C. Jones et al. a move from top-down management to two-way Carnine, D. (2000). Why education experts resist effective communication that allows both administrators practices (And what it would take to make education more like medicine). Effective School Practices, 18 (3), and teachers together to explore what works for 6–14. students while at the same time recognizing deci- Chall, J. S. (1996). Learning to read: The great debate . sions need to be made and those decisions will be Fort Worth: Harcourt Brace College Publishers. questioned. We believe that teacher preparation Coleman, J. S., Campbell, E. Q., Hobson, C. J., McPartland, J., Mood, A. M., Weinfi eld, F. D., & York, R. L. (1966). professors should be in the discussion as well. If Equality of educational opportunity . Washington, DC: there is a “fair” process in place for such discus- U.S. Government Printing Offi ce. sion, consensus, collaboration, and cooperation Ford, P. L. (Ed.) (1892–99). The writings of Thomas can be reached. It calls for both a strong founda- Jefferson (10 vols). New York: G.P. Putnam's Sons, 1892–99. http://catalog.hathitrust.org/api/volumes/ tion in effective instruction (looking at the data) oclc/2428785.html and a transformational approach (relationships Fullan, M. (2001). Leading in a culture of change . and sharing knowledge). This speaks to the recul- California: Jossey-Bass. turing Michael Fullan recommends. Such a recul- Hattie, J. A. (2009). Visible learning: A synthesis of over 800 meta-analyses relating to achievement. New York, turing would allow educators to actively Ny: Routledge. investigate unfamiliar concepts, expand their Jencks, C., Smith, M. S., Ackland, H., Bane, M. J., Cohen, vision of professionalism, and actively seek feed- D., Grintlis, H., Heynes, B., & Michelson, S. (1972). back. It will require changing our mind-set and Inequality: A measurement of the effects of family and schooling in America . New York: Basic Books. providing a real vision for education that Lemov, D. (2010). Teach like a champion: 49 techniques embraces change and views research as a poten- that put students on the path to college. San Francisco, tial new tool for making decisions based on our CA: Jossey Bass. students’ needs and not ours. Lezotte, L. W., & McKee, K. M. (2002). Assembly required: A continuous school improvement system . Finally, all of these threads are connected by Okemos: Effective Schools Products, Ltd, Doug. the concept of change. Avoidance of change, Mann, H. (1848). Annual Report to the Massachusetts whatever the reason, prevents growth. Resistance Board of Education. to change is one of the primary reasons princi- Marzano, R. J. (2003). What works in schools: Translating research into action. Alexandria, Va: ASCD. pals, teachers, students, and practitioners fail at National Reading Panel. (2000). http://www.nichd.nih. school reform. If schools are going to improve gov/research/supported/Pages/nrp.aspx/. Accessed and if we as a country are going to reclaim our 9/24/2014. position as the premier nation for education, we Nye, B., Konstantopoulos, S., & Hedges, L. V. (2004). How large our teacher affects? Education Evaluation must embrace change—that means embracing and Policy Analysis, 26 (3), 237–257. new knowledge and research and effectively Pearson, P. D. (1999). A historically based review of pre- applying that information to educate our youth venting reading diffi culties in young children. Reading and ourselves along with them. Research Quarterly, 34, 231–246. PISA – Organization for Economic Co-operation and Development. Reading fi rst implementation evaluation: Final report, 2008. http://www.oecd.org/pisa/ keyfi ndings/pisa-2012-results.htm . References Stanovich, P. J., & Stanovich, K. E. (2003). Using research and reason in education: How teachers can use scien- Beck, I. (1998). Understanding beginning reading: A jour- tifi cally based research to make curricular & instruc- ney through teaching and research. In J. Osborn & tional decisions. University of Toronto. https://www. F. Lehr (Eds.), Literacy for all: Issues in teaching and nichd.nih.gov/publications/pubs/Pages/using_ learning (pp. 11–31). New York: Guilford Press. research_stanovich.aspx . Carless, D. (2006). Differing perceptions in the feedback U.S. Department of Education. (2013, Apr 28). National process. Studies in Higher Education, 31 (2), Institute of Literacy Research. http://www.statisticbrain. 219–233. com/number-of-american-adults-who-cant-read/ . Part VI

Conclusion The March of Reason: What Was Hidden in Our Genes 29

James R. Flynn

This chapter has eight parts: pessimism about Flynn model shows that it can. The relevant human genetic potential in the wake of the theory question for humanity is whether cognitive and of evolution, cognitive progress in the twentieth moral progress will persist over the next century. century as a historical refutation, how this was captured by massive IQ gains, a paradox posed by the existence of IQ tests, a paradox posed by Darwin and the “Scum Worthy” twin studies, moral progress, reason and morality, and prospects for future progress. Darwin had no concept of a gene as a unit of Despite our genes, social progress has heredity. However, he believed that all creatures enhanced rationality and morality. The industrial inherited characteristics that separated one spe- revolution had subtle effects on both, and we cies from another and also distinguished individ- enter an era in which the measurement of intelli- uals from one another within species. He was a gence (IQ tests) raised questions about whether thoroughly good man but refl ected the prejudices cognitive gains were equivalent to “intelligence” of his day regarding the inherited “weaknesses” gains. A division of labor solves this problem. of people at the bottom of the social scale. The measurement of intelligence properly refers Darwin (1871 , p. 510) lamented that physicians to assessing individual differences in cognitive prolong the lives of everyone and as a result “. . . skills within groups at a particular time and place. weak members of civilized societies propagate The measurement of cognitive progress properly their kind. No-one will doubt that this must be refers to people altering over time: whether they highly injurious to the race of man.” The man who can use reason to deal with a wider range of prob- also independently discovered the theory of natural lems (including moral problems), which is to say selection, Alfred Russell Wallace, records a conver- with cognitive history. Twin studies posed a sation (1890 , p. 93): Darwin is oppressed by the dilemma about the potency of environment to tendency of “the lower classes” to over-reproduce account for cognitive progress. The Dickens/ and characterizes the surplus as children of “the scum.” Wallace’s memory could be at fault. However, by 1890, Wallace had totally rejected this Adapted from material in James R. Flynn, Intelligence image of “civilized society.” He was adamant that and Human Progress: The Story of What was Hidden in our Genes , Elsevier, 2013 English society was too corrupt and unjust to allow any reasonable determination of who was fi t or J. R. Flynn (*) POLS Department , University of Otago , unfi t. He respected Darwin and was unlikely to so Box 56 , Dunedin 9001 , New Zealand describe his views without foundation.

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 471 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_29, © Springer Science+Business Media New York 2015 472 J.R. Flynn

The negative image of the lower classes had much alive: the notion that the genes of a substan- deep roots. In Rob Roy by Sir Walter Scott (1817 ), tial part of society mean that their IQ and other the depiction of near-imbecile servants is quite personal traits, such as resistance to education, extraordinary: fi delity to their master is their only welfare dependency, and criminality, are fi xed at a saving characteristic. A century later, during particular time and not subject to modifi cation by World War I, Lord Curzon observed British sol- new social conditions. Charles Murray believes, as diers bathing: “How is it that I have never been most of us do, that Americans in general deserve a informed that the lower orders have such white valued place in society appreciated by relatives skins?” (Blythe 1964 ). A pity the lower orders and associates. But he provides a table in which were useful as servants. Otherwise these strange we are told that, other things being equal, a loss of white-skinned creatures could have been kept in three IQ points over this generation will mean that zoos. During the intervention in Russia in 1918, the number of women chronically dependent on General Graves of Britain informed General welfare will increase by 7 %, illegitimacy by 8 %, Groves of American that he was getting a reputa- men interned in jail by 12 %, and the number of tion as a friend of the poor and that “you should permanent high school dropouts by nearly 15 % know that these people are nothing but swine” (Herrnstein and Murray 1994 ). (Melton 2001). The lower classes are scum, rab- Those who do not like the term “scum worthy” ble, riffraff, louts, peasants, and imbecile yokels can substitute “elimination worthy.” Surely that sucking on straws. is the cash value of “we want to eliminate your Most intellectuals greeted the spread of edu- genes because you are likely to have children like cation with a ferocious pessimism (Carey 1992 ). yourselves.” I reject the thesis of “scum today, Virginia Woolf and E. M. Forster were both scum tomorrow.” If you have a fi xed pool of devoted to adult education. Yet, Wolff refers to “scum,” and take their IQ at a given time as a the self-taught workingman as someone “we all badge of their inferiority, then if they multiply knew” to be egotistic, insistent, raw, striking, from one generation to another, the percentage of and ultimately nauseating. Forster has no sympa- scum increases. On the other hand, if the lower thy with a clerk whose attempts to educate him- classes can be drained of scum from one genera- self are “hopeless.” He is simply inferior, less tion to another, if they are not permanently scum intelligent, healthy, and loveable, typical of worthy, society may turn low-IQ parents into urbanized rural laborers who should be stripped higher-IQ offspring and even eliminate undesir- of their education and revert to what they can do able personal traits. As evidence: the whole drift well: breed yeomen. D. H. Lawrence, Pound, of the last century shows that modernity can alter Yates, H. G. Wells, George Bernard Shaw, T. S. the minds and capacities of people over time. Eliot, Aldous Huxley, Evelyn Waugh, and Graham Green also derided the capacities of the masses. A rare genius may be hidden among Cognitive Progress in the Twentieth them, but the masses will never match the intel- Century lectual attainments and social responsibilities of the elite. The common preference for tinned Let us forget for a moment that IQ tests were ever food is considered damning. invented and focus on people, those peculiar Darwin’s fear that the scum will multiply and beings that exist even when they are not being perpetuate themselves is based on the assumption tested. We will assume that we do have one mea- that the scum of one generation have something sure of cognitive skills, the humble Vocabulary about them, something that ensures that their test. Moreover, that it has been standardized from children will be the scum of the next generation. time to time on representative samples of the Today we would say that failure is in their American population ever since 1950. Therefore, genes. Although we would never be impolite we have a criterion as to what percentage of the enough to use the word “scum,” the thesis is very US population has a certain level of verbal 29 The March of Reason: What Was Hidden in Our Genes 473

profi ciency at any given time, and we can compare Let us go back 50 years to 1960. Jensen (1980 ) how that percentage altered over time. We also asserts that the average high school graduate was have data from various universities concerning at the 75th percentile and they had only a 50/50 what vocabulary level was a prerequisite for suc- chance of graduating from university. It may be cessful study, and census data on the occupa- said that elite jobs require a graduate degree. tional profi le of the US population. Jensen’s data assume that the average candidate In 1900, professionals were 3 % of the popula- in such a degree program was at the 95th percen- tion. By 1920, they were still only 5 %. They were tile and that the minimum standard was about the held in awe because of their cognitive achieve- 88th. Our data show that today the average is the ments. Even in 1957, when I went to Eastern 85th percentile and the minimum standard is Kentucky to lecture, I was referred to reverentially the 58th. So in 50 years, we have gone from the top as a “PhD man.” By the year 2010, 35 % of 15 % eligible to get elite credentials to the top Americans were in cognitively demanding jobs: 42 %. If the latter seems unrealistic, recall that 15 % highly paid professionals and another 20 % the top 35 % of Americans hold those jobs today. subprofessionals, that is, lower management or Once again, the objection can be put that the technical staff (Carrie 2012 ). There is one possi- universities have set standards below what a uni- ble rebuttal: elite jobs are less cognitively demand- versity education should require. Well if that is ing today. Medical colleagues tell me that doctors true, how can their graduates do jobs that are cog- have to know more science today, commerce col- nitively demanding, indeed more cognitively leagues tell me mangers have to plan with a wider demanding than they were 50 or 100 years ago? range of knowledge, and economics colleagues The standards of the universities pass what we tell me that today’s merchant bankers are virtuo- call the test of external validity. In any event, the sos of cognitive complexity. University academ- brute fact that the masses today fi ll a huge num- ics today sometimes give coherent lectures and do ber of elite jobs falsifi es the pessimism current in research; university technicians are infi nitely 1900. The genetic limitations on their rationality more knowledgeable than in the past. did not forbid the social roles once thought the The prerequisite for obtaining most of these exclusive property of the aristocracy. jobs is a university degree. Scores on the WAIS And what about altered behavior? WAIS (Wechsler Adult Intelligence Scale) Vocabulary vocabulary gains over time show that adult subtest can be equated with scores on the Americans gained the equivalent of 17 IQ points Scholastic Aptitude Test Reading Test (Rodrigo of active vocabulary in the second half of the de la Jara 2012). This is cheating a bit in that the twentieth century (Flynn 2013b ). This was thanks SAT is an offspring of IQ testing. Nonetheless, to the tertiary education revolution. If that gain is the equation tells us what percentage of the US projected back to 1900, before the secondary population is viable at leading US universities. school revolution took place, they made a total The universities will not reveal their minimum gain of 34 points. This is 2.27 standard deviations score, but there is data for the score that isolates above the mean and puts them at the 98th percen- the bottom 25 % of their students (Grove 2012 ). tile of the Americans of 1900. The professionals The average American (50th percentile) is viable of 1900 were the upper 3 %. Who would have at universities such as Corcoran Art & Design, thought that the average person with an average Michigan State, Louisiana Tech, Nevada-Las education could replicate the speech typical of Vegas, and Fairleigh Dickson (Flynn 2013b ). No professionals a century ago? university fl unks as many as 25 %, and therefore, There is additional historical data that attest as it is realistic to put the vocabulary threshold at a to how our minds have altered since 1900. When bit below the average, say at the 37th percentile. Luria (1976 ) interviewed peasants in Russia in If you used an IQ metric, that score would be the 1920s, he found that preindustrial people had only 5 points below average performance. certain cognitive traits in common. 474 J.R. Flynn

First, they did not classify. When he asked the latter expected only superfi cial knowledge what a fi sh and crow had in common, they would of such information and tested for understanding not say that they were animals. One fl ies, one complex relationships between concepts. swims, you can eat one and not the other. They Genovese (2002 , p. 101) concludes: “These fi nd- should not be lumped together because as objects ings suggest that there have been substantial in the concrete world, we use them differently. If changes in the cognitive skills valued by Ohio edu- you asked someone in 1900 what a rabbit and cators over the course of the 20th century.” dog had in common, you use dogs to hunt rabbits. The history of the twentieth century is a story The fact that they were mammals was too inci- of cognitive progress. The word “progress” is dental to be worthy of notice. Second, they did value laden so I will defi ne it by using a hypo- not take the hypothetical seriously. When asked thetical: if we grant that an expanded vocabulary whether granted that there were no camels in and our new habits of mind are necessary to com- Germany, would there be camels in German cit- prehend the universe and our own behavior and ies, they said there must be camels there if the the modern world, they constitute progress. city were large enough. Third, when he asked However, thus far, except for vocabulary, we have them to reason about abstractions such as “wher- no measure of the degree of cognitive progress. ever there is snow bears are white, there is snow at the North Pole, what color are the bears,” they stayed fi rmly rooted in their experience of the Massive IQ Gains over Time concrete world. They had never seen anything but brown bears. But they might believe a reliable You measure something when society decides it witness that came from the North Pole. In frustra- is valuable enough to measure. When people tion they asked Luria how they could solve prob- started to work at dawn and stopped at dusk, what lems that were not real problems. was the need for a personal timepiece? But when Today we all know that we do these three tasks the industrial revolution required people to get to readily. We use classifi cation as a means of order- work on time, we invented the factory whistle, ing the world as a prerequisite to understanding the clock on the mantle, and the wristwatch. it, for example, mammal versus reptile or primate When people inherited their jobs as they did their versus non-primate. We take the hypothetical names, what was the need for an IQ test? But seriously, for example, if medium-sized stars when the industrial revolution required a more eventually expand into red giants, our sun will do educated work force, we invented a measure of so and destroy the earth. We use logic to order who could profi t from education, who could universal assertions, for example, when light progress farthest, and who could become the elite behaves both as if it were a particle and a wave, of the modern world. In 1905, Alfred Binet you cannot classify it as one or the other. I call invented the IQ test. French school children told these cognitive traits new “habits of mind.” him that something new was worth measuring. They are clearly prerequisites for higher educa- It appears that shortly after a nation embarks tion and, as Carmi Schooler (1998 ) has shown, on the industrial revolution, IQs begin to rise. they allow one to perform the tasks of cognitively Thanks to birth date data (scores rising as the demanding jobs. These new habits of mind became subject’s date of birth rises from the past to the so essential that they affected how we educate our present), we know that Britain has made massive children. In 1900, our schools were still fi rmly IQ gains since 1872. There are data from about rooted in facts about the concrete world. Then they 30 nations all over the world, and at their peak, began to teach something new. Genovese com- gains run at the rate of at least 0.3 IQ points per pared the exams the state of Ohio gave to 14-year- year on Stanford-Binet and Wechsler tests, higher old schoolchildren between 1902 and 1913 and on tests like Raven’s Progressive Matrices. between 1997 and 1999. The former tested for in- Over the last century, IQ gains in Britain and depth knowledge of culturally valued information; America amounted to at least 30 IQ points. 29 The March of Reason: What Was Hidden in Our Genes 475

Scored against today’s norms, our ancestors had and the Netherlands appear to have emerged from a mean IQ of 70, the borderline for mental retar- the IQ gains period. The period for some nations dation. They were not retarded, of course. Their may fall well short of 100 years. China and Japan intelligence was pragmatic: it was focused on and Korea industrialized much later than America how to make use of the concrete world for their and Britain, and their rate of social change has own advantage. They lacked our “scientifi c spec- been dramatic. Rapid social change has put their tacles,” that is, the new habits of mind, the formal rate of gain well above the US-British rate, but the education that tutors the mind in logical analysis, price they pay may be a shorter cycle. Developing and the consequent broad range of vocabulary nations that have really begun to develop, and general information. The mind of 1900 that is Argentina, Brazil, Turkey, and Kenya, are just never exposed to such advantages is a far cry entering their massive gains phase. Much of the from a mind that cannot take advantage of them world is still in the doldrums. The next century when exposed (Flynn 2013b ). will be interesting. Developing nations are some Thus far, I have emphasized mass education. 10–30 IQ points behind the developed world. But In fact, causality operated on three levels. The there is strong evidence that those favored by eco- ultimate cause is the industrial revolution or nomic development (Latin America in particular) modernity. The intermediate causes are the indus- will catch the developed world within 40 years. trial revolution’s by-products, not just enhanced Much of the developed world is likely to remain schooling, but a host of other factors. Better stan- in the doldrums or regress under the impact of cli- dards of living nourish better brains. Family size mate change (Flynn 2013a ). drops so that adults and their speech dominate the Given what modernity has done to the human home’s vocabulary and modern parenting appears mind, what kinds of IQ tests or subtests would we (hothouse parenting or encouraging the child’s expect to be most affected? Every nation in its IQ potential for education). People’s professions gains phase has made enormous gains on Raven’s exercise their minds rather than asking for physi- Progressive Matrices. Indeed, the best estimate cally demanding repetitive work. Leisure allows (remember we have birth date data from 1872) is cognitively demanding activity rather than mere a total gain of over 50 points in 100 years. This recuperation from work. The world developed a test above all measures the use of logic on new visual environment so that abstract images abstractions (matrices patterns) removed from dominate our minds and we can “picture” the the concrete word. In essence it is a kind of anal- world and its possibilities rather than merely ogies test. describe it. The proximate causes are the minds Fox and Mitchum (2012 ) have analyzed just people take with them into the test room so they what has allowed each generation to do better on can answer more items correctly, not simply their Raven’s than the preceding generation. One hun- new “habits of mind” (classifi cation, logical anal- dred years ago, Americans could do simple analysis of abstractions) but also vocabularies, gen- ogies grounded in the concrete world: Domestic eral information, and visual awareness. cats are to wild cats as dogs are to what? (Wolves.) IQ gains are not eternal. Sooner or later, the This would do them no good on the kind of anal- intermediate causes gradually lose potency. ogies found on Raven’s. But by 1961, they could Education is widespread and adequate, family handle two squares followed by a triangle imply- size can go no lower, and leisure is as packed ing two circles followed by what? (A semicircle: with as many cognitively demanding pursuits and just as a triangle is half of a square, so a semicir- images as anyone can tolerate; even featherbed- cle is half of a circle.) By 2006, they could handle ding can produce no more elite jobs, so the trig- two circles followed by a semicircle implying gers of massive IQ gains stop. two sixteen’s followed by what? (Eight: you have America and Britain show IQ gains over 100 to see the relationship despite the transition from years or more and are still advancing. However, shapes to numbers.) Note how each step takes us more progressive societies such as Scandinavia further from the concrete world toward using 476 J.R. Flynn logic on abstractions, eventually abstractions information. Given the evidence, it would seem whose very identity shifts. Who can imagine the that those who hold IQ tests in the highest esteem average person in 1900 able to do all of that? Is it would be the fi rst to concede cognitive progress. any wonder that we get much higher scores on In fact, the opposite was true. Raven’s? Those who follow the late Arthur Jensen deny We have referred to Wechsler gains. Where that IQ gains over time are intelligence gains. have these been the largest? They have been the Jensen (1998 ) called them “hollow,” lacking real- largest, fi rst, on the Similarities subtest that forces world cognitive signifi cance because they could you to classify; second, on Analytic subtests that not pass what he called the “method of correlated force you to use logic to devise how blocks or vectors.” Here we must introduce g , often called objects can make certain designs; third, on the the general intelligence factor. There is nothing Pictorial subtests which ask you to fi nd the miss- mysterious about g . Something similar exists in ing piece of a picture or use pictures to tell a story; many areas. Some people have “musical g ”: what- and fourth, on the Vocabulary subtest where adults ever instrument they pick up, they learn quickly. made large gains thanks to more and more educa- Others have “athletic g ”: they shine at all sports. tion. In recent years, children have had no more There is a strong tendency for the same people to additional schooling, and their vocabulary gains score above or below average on all of the 10 or 11 have been modest (Flynn 2013b ) Wechsler subtests, no matter whether they test for In sum, the historical evidence and the pattern vocabulary, general information, mental arithme- of IQ gains match. The enormous score gains are tic, solving three dimensional jigsaw puzzles, or a measure of the enhanced cognitive traits that discerning logical relations conveyed by a matrix. distinguish the modern mind from the minds of Factor analysis measures the strength of the our immediate ancestors. tendency of various subtests to be intercorrelated. You can then go back to the subtests and calculate a hierarchy as to how much performance on But Are They Intelligence Gains? each of them predicts general performance across the whole set of subtests. This is their g loading. The argument thus far rests on two syllogisms. The best predictor is usually (not always) your First, the cognitive demands of elite jobs and performance on the Vocabulary subtest. Now you education are greater than 100 years ago; many can rank the ten subtests from those that have the more people can meet those demands; therefore, greatest “ g loading” down to those that have the there has been cognitive progress. Second, clas- least. Jensen then ranked the same tests from sifying the world, using hypotheticals, and using those whose score gains over time were the great- logic to render generalizations consistent are est down to those whose score gains were least. If more cognitively complex than simply taking the the subtest gains have a negative correlation with concrete world as a given; far more people can do the g loadings, and there is a mild tendency in the former; therefore, there has been cognitive that direction, you conclude that IQ gains are not progress. really intelligence gains. This assumes, of course, Even if no IQ tests existed, any aware person that it is legitimate to identify intelligence with g . can see that his or her mind differs profoundly We can see why Jensen thought the identifi ca- from the American mind in 1900. However, IQ tion appropriate. The impressive thing about the tests do exist and their record of gains over time g loadings of subtests is that they rise with offers a bonus: we can actually measure the the degree of cognitive complexity of the task the degree of cognitive progress modernity confers. subtests measure. As Jensen often pointed out, It would be odd if this were not the case. IQ tests the g loading of digit span forward, a simple task were designed to measure the traits that were of repeating a series of random numbers in the enhanced: logical analysis, analogies, classifi ca- order in which they are read out, has a low g tion, pictorial awareness, vocabulary, and general loading. Digit span backward, a more complex 29 The March of Reason: What Was Hidden in Our Genes 477 task of saying numbers in reverse of the order in from top to bottom. It actually responds to real- which they are read out, has a much higher g world social priorities. If it needs mass education loading. Speed of shoe tying would have a g load- and people to fi ll chattering jobs (law, teaching, ing of close to zero. Most of us feel that the more counseling), it will enhance vocabulary. If it needs cognitively complex a task, the more it measures executives to do lateral thinking, it will encourage intelligence. using the hypothetical. If it needs a wider range of And yet, Jensen’s demand leads to a paradox. information to cope with a more complex modern People over time have made huge gains on sub- world, it will enhance general information. In a tests every one of which poses problems of cog- post-sputnik era, if it wants more people adept at nitive complexity. Yet all of these gains are to be mathematics, it will push arithmetical skills – and dismissed because the gains did not privilege if it does not know how to improve them, gains tasks according to the magnitud e of their cogni- will be slight despite their high g loading. If the tive complexity. Imagine we added shoe tying as fact everyone has their own car enhances the need an 11th subtest and for some reason, perhaps for navigational skills, map-reading skills will go enhanced dexterity over time, people make by far up despite its low g loading. the largest gains on it. This would virtually guar- In other words, when society shifts its priori- antee a negative correlation between IQ gains ties for what mental skills are needed over time, over time and cognitive complexity. The solution it cares absolutely nothing for sheer cognitive to this paradox lies in whether the demand for a complexity. It makes no sense at all to advise it to hierarchy of cognitive complexity is a legitimate respect a g hierarchy. To demand this is to con- demand both for measuring intelligence and fuse society with a giant brain. measuring cognitive progress. I will argue that it Individuals have brains. Genes infl uence their is appropriate for the fi rst but not the second. overall quality; they probably give some people Take two people at a given place and time an optimum blood supply to the brain and an sharing the same cognitive environment (two optimum dopamine spraying system. Certain brothers in the same home). If one accesses that neurons spray dopamine, which strengthen the environment better than the other, it makes sense neural connections in the brain with use, rather to say he has the better mind. Moreover, he is like having a good sprinkling system for your likely to outstrip his brother in accord with cogni- lawn. When we compare generations over time, tive complexity. The less able brother will not we are not comparing one gigantic brain to fall far behind for simple cognitive skills, but he another, both operating in a common cognitive is more likely to fall behind for complex ones. environment, which the later brain accesses more After all, they live in a shared cognitive environ- effi ciently. We are comparing two complex social ment: both are subject to hothouse parenting, systems whose altering cognitive priorities create both will enjoy much the same amount of school- radically different cognitive environments. If the ing, both have modern habits of mind, and so environment has become more cognitively forth. I have complicated views about “intelli- demanding, there is cognitive progress. No one’s gence,” but in this context, I am willing to call the brain is any better at conception, and no one is difference between their IQs, particularly when more intelligent in the sense of adapting better to weighted for g loadings, an intelligence differ- a common environment. The rise in the average ence. Van Bloois et al. (2009 ) have done an excel- IQ compares two cognitive environments, not lent study showing that the gifted, the average, individual differences. Various mental abilities and the mentally retarded differ on Wechsler IQ alter autonomously, that is, without regard to g or subtests in accord with g loadings. relative cognitive complexity. Society, on the other hand, is quite different. It It may be asked: When cognitive skills are does not administer a gigantic IQ test, rank mental enhanced autonomously, does this have any real- skills in order of their degree of cognitive com- world signifi cance? An accumulating number of plexity, and then decide to enhance them going studies show that the answer is yes. 478 J.R. Flynn

Coyle and Pillow (2008 ) show that when you while digit span backwards (more complex) has a deduct g from performance on the SAT, the scores higher g loading. Making a souffl é has a higher g still predict university grades. Ritchie, Bates, and loading than scrambling eggs. Once we accept the Deary (under review) show that the effects of relationship, it is illuminating. Vocabulary (assum- education are benefi cial even though they are not ing equal opportunity) ranks minds for the cogni- mediated by g but consist of direct links to spe- tive complexity of the concepts they can absorb. cifi c subtests. Woodley et al. (Woodley 2012a , b ) Arithmetic ranks minds for how well they can show that education actually tends to promote plan a numerical strategy and carry it out mentally diversifi ed cognitive skills and that IQ gains over (without pen and paper). Which of the two time (which of course do not correlate with g ) involves more cognitive complexity? Vocabulary parallel and predict growth in GDP per capita. has the higher g loading – fascinating. Woodley (in press) concludes that autonomous The hierarchy of g loadings correlates with the skills allow one to adapt cognitively to modernity degree to which inbreeding (negatively) infl u- and thereby promote a better life. Armstrong and ences subtest performance. This shows that those Woodley ( under review ) show that modernity in areas of the brain that do cognitively complex general encourages greater sensitivity to a whole mental tasks have a genetic substratum more range of rules, ones that operate independently in fragile than those areas that do less complex a complex web of social situations, rather than tasks. They are more subject to damage by the collectively as assumed by g . Finally, for the spe- pairing of undesirable genes during sexual repro- cialists, Fox and Mitchum ( 2013) show that duction. This is what inbreeding enhances. We enhanced performance on Raven’s is not due to have a signifi cant contribution to our knowledge hollow skills (like test sophistication) but to real- of brain physiology. world cognitive advance, even though the skills enhanced are not correlated with g and are not factor invariant . The Tale of the Twins So now a simple division of labor has solved the paradox. We will restrict the use of g hierar- Twin studies (and other kinship studies) chal- chies to assessing individual differences between lenge not the signifi cance of cognitive progress people sharing a relatively homogenous cognitive but whether a coherent account can be given of environment at a given place and time. And we the causes of IQ gains. will eschew g when assessing what generational Take identical twins that were separated at differences over time occur as people’s minds birth and raised by different families. If they alter thanks to altered social priorities. Honor will grow up with identical IQs, the inference is that have been preserved for all. We will never con- identical genes trump dissimilar and enfeebled taminate g by calling cognitive progress “intelli- environments. If they grow up with IQs no more gence gains,” and we never dismiss cognitive alike than the rest of us, dissimilar environment gains by demanding that they be g gains. However, has trumped identical and enfeebled genes. The the two are kissing kin: both have to do with result: they are far more alike for IQ than ran- enhanced ability to solve cognitively complex domly selected individuals. By adulthood, all problems, one by individuals in pecking order, the kinship studies show that family environment has other by generations helter-skelter. faded away to zero. Adult IQs differ only to the I should add that I do not mean to imply that degree that chance events might cause them to the concept of g is trivial. The fact that g loadings differ (one is dropped on his head and the other correlate with cognitive complexity is illuminat- was not). It is hard to see how chance events ing. We must rely on our intuition to establish that could differ between the generations and cause the two correlate at all but specifi c cases are con- massive IQ gains over time. vincing. There is the fact that digit span forward Thus, environment is too feeble to have much (simple task of rote memory) has a low g loading, infl uence on IQ. Yet, massive IQ gains over time 29 The March of Reason: What Was Hidden in Our Genes 479 occur whose causes appear to be overwhelmingly this raised the skill level. Indeed the rising average environmental. We have a new paradox: How can performance became a causal factor in its own solid evidence show that environment is both right and a new feedback mechanism was born, feeble (kinship studies) and potent (IQ gains) at which we call the social multiplier . the same time? The Dickens/Flynn model solves To be above average, it was initially good this paradox (Dickens and Flynn 2001a , b , 2002 ). enough to shoot and pass well. Then ambidex- Let us see what happens to children that are trous people began to pass with either hand and genetically identical but grow up in different fi nd more open players, and the rising mean families. I will use basketball as an example. Joe forced everyone who wanted to keep up to do the and Jerry are identical twins separated at birth so same. Then people began to shoot with either that one is raised in Muncie Indiana and the other hand and get more opportunity to score baskets in Terre Haut. Thanks to their identical genes because they could go around a guard on either both will be four inches taller and a bit quicker the right or left side. Almost overnight basketball than average (faster refl ex arc). Indiana is a bas- was transformed from the stodgy sport of 1950 to ketball mad state, and at the start of school, both the incredibly fl uid and graceful sport that took boys get picked to play sandlot basketball more root in the 1960s. often than other kids. This is the beginning of The comparative potency of genes or environ- matching above average genes with an above ment depended on whose hand was on the throt- average environment. Moreover there is recipro- tle of a multiplier. Comparing individuals within cal causation between their skills and their envi- a cohort, genes co-opted environment and genes ronment: better skills mean a better environment, seemed omnipotent, thanks to the individual mul- which upgrades their skills, which means an even tiplier. Comparing generations over time, evolv- better environment, and so forth, essentially a ing environment broke free to raise the feedback mechanism. The Dickens/Flynn model performance in basketball to new heights, thanks calls this the individual multiplier . to the social multiplier. Next they make their grade school teams, I take it the analogy is obvious. Identical twins which upgrade their skills further, and they both in separated environments may have genes that make their high school teams and get profes- set them above (or below) the average person for sional coaching. These separated twins will end cognitive ability. If above, what are small genetic up with highly similar basketball skills, but why? differences at birth become potent because they Not merely because of their identical genes but co-opt matching and superior cognitive environ- also because of their highly similar basketball ments: more attentive teachers, superior peer histories. In the kinship studies, genes get all the interaction, honor streams, and better high schools credit and basketball environment gets nothing. and universities, factors hardly rendered impotent But this is a misinterpretation. It pretends that simply because they are co-opted by genes. Over environment is feeble, when in fact their genes time, things are different. Increasing the years of have co-opted something as potent as more play, schooling from six to twelve to more than twelve team play, and professional coaching. Potent (university) really does do something to enhance environment is disguised simply because it is the cognitive abilities of the whole society. The matched with identical genes. mere fact that genetic differences tend to deter- Now let us shift to factors that affect the collec- mine how many years of school a person gets at a tive basketball environment over time. The genes given time does not weaken the potency of addi- of people in general are essentially static over a tional years of schooling over time. few years, so now basketball environment is cut Just as the near identical scores of separated loose from genes and emerges in all its potency. identical twins do not rob environment of its After World War II, TV was invented and the potency, the huge environmentally induced IQ close-ups of basketball were exciting and popu- gains over time do not rob genes of their potency. larized the sport. Far more people participated and They are both potent enough to do their jobs, 480 J.R. Flynn explaining individual differences versus explaining practice never disappeared and the respectable group differences over time. classes’ image of mass violence persisted. Thirty The multipliers also solve a problem that baf- books that appear after 1840 express apprehen- fl ed the psychological community. If environ- sion: “Now it is the general complaint of the tav- ment is weak within groups, then to explain huge erns, the coffee-houses, the shopkeepers and environmental effects between generations over others, that their customers are afraid when it is time, you have to invent a factor X: a mysterious dark to come to their houses and shops for fear environmental factor that operated exclusively that their hats and wigs should be snitched from between groups or generations. their heads or their swords taken from their sides, We now see that much the same factors are or that they may be blinded, knocked down, cut operating within and between groups. Within or stabbed; nay, the coaches cannot secure them, groups, individuals are distinguished by factors but they are likewise cut and robbed in the public like better families, teachers, peers, universities, streets, etc.” (Shoemaker 2004 , p. 162). and jobs. These factors are made to seem feeble When we turn to at our genetic inheritance, because the individual multiplier correlates them the pessimism of the elite about moral progress with genetic differences, and twin studies show may seem to haves some substance. Our nearest them as having little impact beyond what genetic primate relatives suggest that over much of differences would dictate. Between groups, the human evolution, males and females were sub- two generations are also distinguished by factors ject to different selective pressures. like better parenting, more schooling, and more Males competed for access to females by cognitively demanding jobs. But thanks to the either violent combat or aggressive displays that social multiplier, they have huge effects simply intimidated rivals. Since aggressive males as environmental variables. They operate free of fathered the most offspring, their genes became genes because there are no real genetic differ- dominant. Females perpetuated their genes to the ences between the generations that they could be extent that they raised their children to maturity, correlated with. so that their children could reproduce. A bond In sum, the “weakness” of an environmental with a male helpmate was advantageous. factor within groups is a mere appearance and Therefore, genes for whatever helped domesti- does not translate into weakness between groups. cate males were positively selected. These pro- Much the same environmental factors operate clivities prepared the way for the emergence of both within and between groups and no mysteri- traits that statistically differentiate the genders. It ous factor X is necessary. The factors that sepa- is politically incorrect to assert that women are rate generations do not necessarily, of course, cleaner, more attentive to physical appearance, apply to ethnic groups. Black subculture digs a more skilled at arts that make home life attrac- gulf between black and white that is peculiar to tive, and more likely to use charm rather than those two groups (Flynn 2008 ). (overtly) aggressively behavior to attract the opposite sex. I will reply on those of both sexes who see through their eyes and not their The London Mob ideologies. However, there is reason to believe that our Many members (not all) of the Victorian elite genes have altered. Hallpike (2008 ) points out were pessimistic about moral progress. This was that male aggression began to pay decreased pro- partially based on “the London mob.” In 1780, creative dividends in the simplest Homo sapiens the House of Commons refused to debate a peti- societies, the hunter-gatherer societies universal tion against granting Catholics toleration. The until about 10,000 years ago. The simple societ- poor, criminals, and prostitutes rioted with hun- ies that survive today show that the collective dreds killed and some hanged (German and Rees action of other males can eliminate or expel an 2012 ). Although this was their last great riot, the overly aggressive male. The best hunter is 29 The March of Reason: What Was Hidden in Our Genes 481 expected to share his kill (spoilage makes most of tions against domestic violence. Some societies it worthless to him). About 10,000 years ago, lag. Many Sunni Muslims allow a man to divorce human beings started living in larger communi- his wife by saying “I divorce thee” three times. ties, which were functional only if aggression The husband is not responsible for the wife’s was restricted by rules. Just as people domesti- expenses (but responsible for the maintenance of cated animals like dogs and cats, people began to children until they are weaned). domesticate themselves. Just as domesticated Middle Eastern men have become aware of animals were selected for self-control of their what they face. Virk (2012 ) says that historically aggressive behavior, not to be directed at their men have tended to be free spirited, adventurous, masters but to be governed by rules the master and wild. He describes fi ve stages of domestica- set, so people were domesticated by genetic tion: courtship (a man wears clothes and uses per- selection for self-control and rule-bound behav- fumes agreeable to women and affects an interest ior (Wilson 1991 ; Leach 2003 ). in culture), declaration (he must express love Steven Pinker (2011 ) comes onstage at this rather than compliments), employment (he must point. The growth of larger cities and nations get a job so she can hold him in esteem), home increased the range of people that the inhabitants ownership (she tortures him with an account of were “trained” to forgo aggression against. Trade how their landlord tried to take sexual liberties and within and between nations was an important fac- suggests that rather than killing the landlord, the tor: you want to preserve a lucrative customer, obvious remedy, he buy a home of their own), and not kill him and confi scate his property. To put parenthood (she begins to call him childish names the point in evolutionary terms, assume that over such as “baby” and shifts child care onto him). His 1,000 generations law-abiding citizens have complete domestication is signaled when “they go outreproduced those predisposed to violence. If to market with a baby hanger on the husband’s so, human genes were selected so that we fi nd it back and a patent little handbag in the wife’s arm.” easier today to live together without physical This description reveals, I fear, a determination to aggression. This is plausible but unproven: evi- fi ght in the trenches. dence will follow. Now we turn to the evidence. If domestication Over the last 1,000 years, there has been has occurred, violence should have declined. The another domesticating trend. Males are respon- seventeenth and eighteenth centuries shift away sible for most acts of violence. Female domesti- from cruelty: fewer amusements like roasting a cation of males is signaled by the fact that males cat alive or men competing to batter a pig to death are violent primarily between puberty and some- with clubs. The last heretic was tortured; the last time in the 20, after which they are pacifi ed by witch burned in Europe. Slavery had existed for the responsibilities of marriage and child rearing thousands of years. In the nineteenth century, “an (Pinker 2011 ). As civilization developed, male overwhelming majority of Westerners came to competition for women focused less on violence feel that slavery was wrong.” Dueling is gone and and more on money, status, and amiability (a clan feuds, gang wars, and lynching are nothing nice guy). compared to a century ago. We no longer glorify Over the last few generations, some women winning the West by killing Indians. Slaughtering gained power to pacify males because of trends people on the highways so we can enjoy drink, that empower them in the home: the ability to the intoxication of speed, or the manly desire to fi nd employment so that they need not be totally use a car as a tool of combat is questioned dependent on males to support themselves and (Brinton 1959 ). their children, the presumption that both sexes Pinker (2011 ) adds quantifi ed evidence. As for will contribute to home maintenance and child violent death from war, hunter-gatherers (14000 rearing, the fact that division of property and BC to 1770 BC) get a rate of 15 per hundred child support means that a male cannot evade deaths. Beginning with the early cities and responsibilities through divorce, and legal sanc- empires of recorded history, the rate falls to 3–5%. 482 J.R. Flynn

Perhaps a better measure is the chance the average murder and innumerable innocent people were person has of dying from violence in a given year. executed. It is horrible to contemplate that some In thirteenth-century England, the homicide rate of them thought that they were guilty: What if was over 20 per hundred thousand per year. From they had wished the person dead or had dreamed the sixteenth to the twentieth century, the rate about their death? When murders occurred, using steadily dropped down to less than one throughout divination to establish the guilty party was coun- Europe. America shows about 5 people per terproductive. Many murderers walked free ready 100,000. The “far north” of America (New England to kill again. west to Oregon and Washington) is as safe as The personifi cation of animals was inherited Europe but homicides escalate as you go south. from tribal society. From the thirteenth to the Pinker calls the period after 1946 the long eighteenth centuries, animals thought complicit peace. There are civil wars and great powers in murder, assault, plague, or bestiality were tried bully minor ones. However, no great power has and executed throughout Europe. They included engaged in direct combat with another. Until pigs, horses, bulls, cows, sheep, rats, beetles, and recently, the expectation that great powers would insects. Some were clearly wronged: in 1474, a fi ght one another was normal. The Hapsburgs, rooster was prosecuted for laying an egg fathered Spain, France, England, Russia, Germany, by Satan. Some were exonerated: lawyers won America, Italy, Turkey, Japan, China, the famous victories representing rats and beetles Netherlands, and Sweden all did so. Finally, since (Evans 1906 ). 1989, there has been the new peace. Civil wars, Tertullian extolled holy ignorance: “We have genocides, repression by autocratic governments, no need of curiosity after Jesus Christ, nor of and terrorist attacks have all declined. We have a research after the Gospel.” Fortunately, Europe new commandment: no wars shall be fought to did not heed him. annex territory across national boundaries. By 1900, the new scientifi c ethos had blind faith on the defensive. However, it did little to banish the secular demons of racism and nation- Reason and Morality alism that culminated in the horrors of World War II and the holocaust. We were still like domesti- We return to our main theme: the consequences cated animals. We had selected ourselves to resist of the march of reason. Georg W. Oesterdiekhoff violence within groups but not between groups: (2009 ) has traced the effect on ethics running we were happy to coerce “inferior” races or kill from magic to religion to a scientifi c knowledge traditional enemies. Yet, over the last 70 years, of reality. Magical and religious beliefs produced these demons have been on the defensive. much immoral behavior in the past: human sacri- During the twentieth century, we made an fi ce to feed the gods (the Aztecs killed about 1.2 enormous leap forward in adopting a new moral million), the burning of witches, and the horrors notation. The story of mathematics is the story of of the Inquisition. In the Old Testament, God improved notation. Greek symbols were so cum- instructs his chosen people to slaughter animals. bersome that it took the genius of Archimedes to When Aaron’s two sons do so using the wrong represent large numbers. Roman numerals were kind of incense, he burns them alive. God is an advance but contemplate the task of dividing aware that a captured woman may not be in the MDCCCVIII by IV: the answer is 452 (1808 mood for sex having seen her husband and chil- divided by 4). The modern mind has a new way dren slain. God advises the Israelites to shave her of stating moral maxims. Remember Luria. We head, pare her nails, and imprison her until she are now ready to generalize, challenge general- sees the wisdom of being raped (Pinker 2011 ). izations by using hypotheticals, and demand that Are we fully aware of what it was like to live they be logically consistent with one another. everyday life surrounded by superstition? In In 1955, Martin Luther King began the many tribal societies, every natural death was a Montgomery bus boycott. My brother and I 29 The March of Reason: What Was Hidden in Our Genes 483 argued with our father: “What if you woke up of mind that upgrade moral debate. Valuing your tomorrow and had turned black?” Reply: “That is possessions is not the same as testing generaliza- the dumbest thing you have ever said, who do tions against logic. you know that turned black overnight?” He sim- The UNODC (2013 ) has found that national ply would not take the hypothetical seriously. My differences in homicide rates correlate better with Beyond Patriotism (2012) diagnoses the retreat intelligence (measured by IQ and school achieve- from nationalism since Vietnam. “What if your ment) than with years in school, GDP, less cor- home was hit by a drone because someone nearby ruption, and greater freedom and democracy. was sheltering a Taliban?” Or “If a war killed for- Hodson and Busseri (2012 ) found that low IQ in eigners to save 3,000 Americans, where would childhood predicted racism, homophobia, and you fall off the boat: at 10,000 or 100,000 or one membership in groups inclusive of hyper- million?” The answer tends to divide youth from nationalism. Still, too many variables are corre- age (the latter: “their government protects them lated with IQ to use this evidence to single out and our government protects us”). Inherited max- patterns of moral reasoning as a potent variable. ims can be very cruel. Islamic fathers shock the I can only appeal to the historical record. world when they kill a daughter because she has been raped. We would ask: “What if you had been knocked unconscious and sodomized?’ He Progress at Risk is unmoved. He sees moral maxims as concrete things, no more subject to logic than any other Will science and rationality spread to embrace concrete thing like a stone or tree. the world? Unfortunately, something that has Today we worry about the “collateral damage” pacifi ed humanity undermines steps to deal with of killing foreigners in Afghanistan and Pakistan. the most important threat we face, and some No military commander uses language like actors ignore the most important rule for main- “bombing the Vietnamese back to the stone age.” taining peace. In 1914, Thomas Mann says he had long felt the In the past, the more nations that enjoyed eco- need of a war to subordinate materialism to nomic progress and engaged in international “German Kultur. ” Rilke called the war the resur- trade the better. Today, the momentum of eco- rection of “the God of hosts.” Max Weber gushed nomic progress promises to make cutting carbon “this war is great and wunderbar. ” Even the saintly levels in the atmosphere impossible. There has Martin Buber lost his mind: “I know personally never been a time in the earth’s history when the that Belgian women amused themselves by put- carbon content of the atmosphere has been above ting out the eyes of wounded German soldiers and 1,000 ppm (parts CO2 per million) and when the forcing buttons ripped from their uniforms into the polar ice caps still existed. A race goes on empty eye sockets” (Elon 2003 ). List the men of between how much carbon dioxide we emit per letters who would talk like that today. unit of economic output, which is diminishing by There have always been people who were 1.3% per year, and how fast economic growth antiracist and anti-nationalistic and subscribed to escalates, which is about 3.45 % per year. something like the golden rule: the innocent Projections: we will pass 500 ppm by 2050 and should not suffer; put yourself in their place. But the no polar ice caps value of 1,000 ppm soon for the overwhelming majority of humanity the after 2100. To win this race, the rate of economic golden rule was merely one of a host of inherited growth would have to fall. People in the devel- maxims: blacks should know their place, my oped world would have to lower their standard of country right or wrong, the obligations of honor living. The present trend of raising people in the (kill my daughter), and the “rights” of the indi- developing world out of poverty would come to a vidual (own my own gun). They might fi nd some tragic end (Flynn 2013b ). of these inherited things more attractive than oth- The Kyoto talks are going nowhere. What ers, but that did not mean they had the new habits American President is going to accept targets that 484 J.R. Flynn would have him face reelection on a platform of References less prosperity? What Chinese leader is going to tell his rural poor that they are going to stay poor? Armstrong, E. L., & Woodley, M. A. (under review). The How can we stop temperature rise without cut- rule-dependence model explains the commonalities ting the growth rate that is the only hope of the between the Flynn effect and IQ gains via retesting. world’s poor? Stephen Salter of Britain has pro- Blythe, D. (1964). The age of illusion. Boston: Houghton Miffl in. posed by far the least dangerous method. At a Brinton, C. (1959). A history of Western morals . cost negligible compared to the costs of climate New York: Harcourt. change, a fl eet of ships would send sea spray Carey, J. (1992). The intellectuals and the masse: Pride upward to whiten the clouds and refl ect away the and prejudice among the literary intelligentsia, 1880– 1939 . London: Faber. sun’s heat. This would actually lower the earth’s Carrie, A. (2012). Occupation change: 1920–2010. temperatures, and in the meantime, we might Weldon Cooper Center for Public Service: http://stat- develop clean power: using lasers or plasma to chatva.org/2012/04/06/occupation-change- achieve hydrogen fusion. 1920-2010/ Note: The 1900 census did not use the same system of classifi cation. However, Carrie put The territorial commandment that “no one 1920 at 5 percent professionals and 1910 was 4 per- uses force to annex territory” is fundamental to cent: Durand, E. D., & Harris, W. J. (1999), Population banning war over the twenty-fi rst century. The 1910: Occupational statistics (United States Bureau of Middle East is volatile because of the antago- the Census), New York: Norman Ross (Table 14). Therefore, I put 1900 at 3 percent. nism between Sunni Muslims and Shiite Coyle, T. R., & Pillow, D. R. (2008). SAT and ACT pre- Muslims. There is also an ambiguity that creates dict college GPA after removing g. Intelligence, 36 , a far more dangerous situation. Many regard the 719–729. 1967 border between Israel and the occupied ter- Darwin, C. (1871). The descent of man and selection in relation to sex . London: Macmillan. ritories on the West bank of the Jordan as the de la Jara, R. (2012). How to estimate your IQ based on potential border between Israel and a Palestinian your GRE or SAT scores. Google “IQ comparison state. Thus, even moderate Arab opinion sees site”. Israeli expansion of settlements in that area as a Dickens, W. T., & Flynn, J. R. (2001a). Great leap forward: A new theory of intelligence. New Scientist , 21 violation of the territorial commandment. This April: 44–47. gives camoufl age to extremists who preach a Dickens, W. T., & Flynn, J. R. (2001b). Heritability esti- crusade to eliminate Israel. Terrorist groups mates versus large environmental effects: The IQ para- harass Israel with all means of sabotage they can dox resolved. Psychological Review, 108 , 346–369. Dickens, W. T., & Flynn, J. R. (2002). The IQ paradox is command. When they get drones, this may reach still resolved: Reply to Loehlin and Rowe and Rodgers. intolerable levels. Psychological Review, 109 , 764–771. The statement of the problems we face forbids Elon, A. (2003). The pity of it all: A portrait of the optimism. Still, whatever happens to us, we can German-Jewish epoch 1743–1933. New York: Picador. take satisfaction in how far we have come. Living Evans, E. P. (1906, 1987). The criminal prosecution and our lives day by day, we take modernity for capital punishment of animals . London: Faber and granted. The very existence of the modern world Faber . is astonishing. I refer not to the Internet or the air Flynn, J. R. (2008). Where have all the liberals gone: Race, class, and ideals in America . Cambridge, UK: travel or the organ transplants but to the people. Cambridge University Press. No totalitarian regime created a “new man” but Flynn, J. R. (2012). Beyond patriotism: From Truman to without fanfare impersonal social forces have Obama . Exeter: Imprint Academic. begun the task. The upper classes were so confi - Flynn, J. R. (2013a). The “Flynn effect” and Flynn’s paradox. Intelligence , 41, 851–857. dent that the masses could never match their Flynn, J. R. (2013b). Intelligence and human progress: intellectual attainments and social responsibili- The story of what was hidden in our genes . London: ties. They were so confi dent that the London mob Elsevier. could never be pacifi ed. They were wrong. As Fox, M. C., & Mitchum, A. L. (2012). A knowledge based theory of rising scores on “culture-free” tests. Journal Kipling (1996 ) put it, “For the Colonel’s lady and of Experimental Psychology: General . doi: 10.1037/ Judy O’Grady are sisters under their skins.” a0030155 . 29 The March of Reason: What Was Hidden in Our Genes 485

Fox, M. C., & Mitchum, A. L. (2013). A knowledge based Ritchie, S. J., Bates, T. C., & Deary, I. J. (under review). theory of rising scores on “culture-free” tests. Journal Does education boost general intelligence ( g ) or spe- of Experimental Psychology: General , 142 , 979–1000. cifi c cognitive abilities? Genovese, J. E. (2002). Cognitive skills valued by educa- Schooler, C. (1998). Environmental complexity and the tors: Historic content analysis of testing in Ohio. Flynn effect. In U. Neisser (Ed.), The rising curve: Journal of Educational Research, 96 , 101–114. Long-term gains in IQ and related measures (pp. 67–79). German, L., & Rees, J. (2012). A people’s history of Washington, DC: American Psychological Association. London . London: Verso. Scott, S. W. (1817). Rob Roy. Edinburgh: Archibald Grove, A. (2012). College admissions. About.com.guide Constable. (search by each state). Shoemaker, R. B. (2004). The London mob: Violence and Hallpike, C. R. (2008). How we got here: From bows and disorder in eighteenth- century England . London: arrows to the space age. Central Milton Keynes: Hambledon & London. AuthorHouse. United Nations Offi ce on Drugs and Crime. (May 2013). Herrnstein, R. J., & Murray, C. (1994). The bell curve: Homicides 2008. New York: United Nations Publications. Intelligence and class structure in American life . van Bloois, R. M., Geutjes, L. L., te Nijenhuis, J., & de New York: The Free Press. Pater, I. E . (2009). g loadings and their true score cor- Hodson, G., & Busseri, M. A. (2012). Bright minds and relations with heritability coeffi cients, giftedness, and dark attitudes. Psychological Science, 23 , 187–195. mental retardation: three psychometric meta-analyses. Jensen, A. R. (1980). Bias in mental testing . London: Paper presented at the 10th Annual Meeting of the Methuen. International Society for Intelligence Research, Jensen, A. R. (1998). The g factor: The science of mental Madrid, Spain, December. ability . New York: Praeger. Virk, S. H. (2012). The domestication theory. http:// Kipling, R. (1996). The Ladies, Stanza VIII, in Barrack- worldpulse.com/node/49936 room ballads (2nd series). Wallace, A. R. (1890). Human selection. Popular Science Leach, H. M. (2003). Human domestication reconsidered. Monthly, 38 , 90–102. Current Anthropology, 44 , 349–368. Wilson, P. J. (1991). The domestication of the human Luria, A. R. (1976). Cognitive development: Its cultural species . New Haven: Yale University Press. and social foundations. Cambridge, MA: Harvard Woodley, M. A. (2012a). A life history model of the Lynn- University Press. Flynn effect. Personality and Individual Differences, Melton, C. K. W. (2001). Between war and peace: Woodrow 53 , 152–156. Wilson and the American expeditionary force in Woodley, M. A. (2012b). The social and scientifi c tempo- Siberia, 1918–1921 . Macon: Mercer University Press. ral correlates of genotypic intelligence and the Flynn Oesterdiekhoff, G. W. (2009). Mental growth of human- effect. Intelligence, 40 , 189–204. kind in history . Norderstedt: Norderstedt Bod. Woodley, M. A., Figueredo, A. J., Ross, K. C., & Brown, Pinker, S. (2011). The better angels of our nature: The S. D. (in press). Four successful tests of the cognitive decline of violence in history and its causes . London: differentiation-integration effort hypothesis. Penguin. Intelligence, 41, 832–842. Closing Comments: Intelligence and Intelligence Tests – Past, 30 Present, and Future

Jack A. Naglieri and Sam Goldstein

For nearly 3,000 years, philosophers, educators, however, is about solving problems in better and physicians, scientists, and psychologists have effective ways. We have chosen to address this discussed, debated, and written about the appar- closing chapter not to the thousands of years of ent singular force of intelligence that has evolved discussion about intelligence as a philosophical in all species but found its greatest progression or scientifi c concept but rather to the last 100 in humans. Yet, it has been a little more than 100 years and the myth of IQ. years that the concept has dramatically evolved The history of IQ and intelligence testing has as a powerful phenomenon used to make critical, been mired in controversy, acrimony, and mis- often life-changing decisions about people. Often trust leading some to characterize the fi eld in lost in the debate over “what is intelligence and the most negative of terms. For example, Ridley how should it be measured” is an appreciation of writes “few debates in the history of science have the evolutionary basis of this concept. In this vol- been conducted with such stupidity as the one ume, we have sought to push the historical roots about intelligence” (200x, p. 77). Ridley’s criti- of this concept in ourselves and all species on this cism may be considered an overstatement, but it planet to the forefront. We argue and debate the is clear that intelligence as a measurable phenom- role of intelligence in education, yet it appears to enon was poorly defi ned at the outset. The lack of us that few of the participants in this often heated a fi rm theoretical basis for IQ tests was noted by controversy have taken the time to understand and Pintner (1923 ) when he wrote that test develop- appreciate this concept on an evolutionary basis. ers “borrowed from every-day life a vague term As the joke goes that our nose evolved to hold our implying all-round ability and knowledge” and glasses, so too can we apply this blind logic to are still “attempting to defi ne it more sharply and conclude that intelligence evolved to make aca- endow it with a stricter scientifi c connotation” demic decisions about children. Intelligence, as (p. 53). Despite the vague origins of the concept the many contributors to this volume can attest, of general intelligence, today, more than a century after the seminal work of Alfred Binet and the US Army, verbal, nonverbal, and quantitative J. A. Naglieri , Ph.D. (*) tests are used to measure intelligence and widely University of Virginia accepted among psychologists and the general e-mail: [email protected] public as well and often considered a gold stan- S. Goldstein dard for assessment of intelligence (Matarazzo University of Utah , 230 South 500 East, Suite 100 , Salt Lake City , UT 84102 , USA 1992 ; Stano 2004). As this volume attests, the e-mail: [email protected] time has come for a reexamination of the concept

S. Goldstein et al. (eds.), Handbook of Intelligence: Evolutionary Theory, 487 Historical Perspective, and Current Concepts, DOI 10.1007/978-1-4939-1562-0_30, © Springer Science+Business Media New York 2015 488 J.A. Naglieri and S. Goldstein of intelligence. The time has come to recognize it was known that a person could fail Alpha how early notions and historical precedence have (verbal and quantitative) tests because of limited infl uenced our thinking about intelligence. educational training and particularly poor skills Despite the 3,000-year debate about intelli- in reading and writing in English. These persons gence, traditional IQ as we know it today was were then tested with the nonverbal tests to avoid formulated just over 100 years ago in 1917 as “injustice by reason of relative unfamiliarity described by Yoakum and Yerkes (1920 ) and with English” (Yoakum and Yerkes 1920 , p. 19). solidifi ed in 1939 with the publication of the Thus, the original intent of verbal and nonverbal Wechsler-Bellevue scales. The importance of tests refl ects the practical need that existed 100 these tests to our society is substantial because years ago and still has relevance today (see they shaped how we defi ne intelligence (Anastasi Naglieri and Ford 2003 ) to measure general and Urbine 1997). Results from these tests have intelligence. infl uenced the lives of countless children and It is well documented that there is consider- adults around the world. It is clear that intelli- able empirical support for the concept of general gence tests represent one of the most infl uential intelligence as measured by tests such as the contributions made by psychology to society in Wechsler and Binet (see Jensen 1998 for a general. The fact that these tests have had con- review). Perhaps one of the most important siderable impact on our fi eld is indisputable, as sources of validity evidence for traditional IQ is the fact that they have infl uenced what we tests is the fact that IQ scores are a good predictor think intelligence is and how it should be of school achievement (Naglieri and Bornstein measured. 2003; Ramsey and Reynolds 2004 ). But there is The Binet and Wechsler scales have histori- circularity in this logic; the verbal tests used to cally measured IQ using tests of similar content. measure intelligence are remarkably similar to Wechsler’s tests were based on Binet’s work the tests used to measure achievement. This prob- (Kaufman and Lichtenberger 2000 ) and the lem was true when the tests were fi rst described methods used by the US military in the early by Yoakum and Yerkes ( 1920), and it is still true 1900s (Yoakum and Yerkes 1920). Wechsler today for individual as well as group tests of included verbal and performance, or nonverbal, intelligence. tests but not because he intended to measure ver- The lack of a clear distinction between ability bal and nonverbal types of intelligence but and achievement has corrupted the very concept rather because it was clear that nonverbal tests of ability in such a way that any child who does helped to “minimize the over-diagnosing of not have an adequately enriched educational feeble- mindness that was, he believed, caused by experience will be at disadvantage when assessed intelligence tests that were too verbal in content … with a so-called “ability” test that involves verbal [in fact] he viewed verbal and performance and quantitative test questions. tests as equally valid measures of intelligence The practical importance of separating the and criticized the labeling of performance [non- content in tests of ability and achievement is par- verbal] tests as measures of special abilities” ticularly salient for children with limited native (Boake 2002, p. 396). This view is similar to that language skills or those from lower socioeco- described by Yoakum and Yerkes ( 1920) perhaps nomic levels where the degree of enrichment in best illustrated by the administration procedures the home is limited. It is well known that poverty for the Army Mental Tests. The general proce- or low SES negatively affects students’ test per- dure involved assignment of literate draftees formance; high poverty is correlated with low to one room, called the Alpha room, and the illit- test scores because of issues associated with erates to another room, called the Beta room educational enrichment at home and at school. (p. 18). Beta (nonverbal) tests were used because Thus, many students receive low test scores 30 Closing Comments: Intelligence and Intelligence Tests – Past, Present, and Future 489 because of unequal opportunity to learn. Too References many of these students – from all racial and cultural backgrounds – are penalized on traditional Anastasi, A., & Urbine, S. (1997). Psychological testing tests of intelligence and achievement and, subse- (7th ed.). Upper Saddle River: Prentice Hall. quently, denied access to gifted education pro- Boake, C. (2002). From the Binet-Simon to the Weschsler- grams and services. Such denial of access is Bellevue: Tracing the history of intelligence testing. Journal of Clinical and Experimental Neuropsychology, common when tests are highly verbal and highly 24 (3), 383–405. achievement oriented, as just discussed. Bracken, B. A., & Naglieri, J. A. (2003). Assessing diverse In recent years, however, researchers have populations with nonverbal tests of general intelligence. begun to examine how effective the general intel- In C. R. Reynolds & R. W. Kamphaus (Eds.), Handbook of psychological and educational assessment of chil- ligence approach is and indeed to wonder about dren (2nd ed., pp. 243–273). New York: Guilford. the limitations of this approach (Naglieri 1999 ; Ford, D. Y. (1996). Reversing underachievement among Sternberg 1988 ). The verbal/nonverbal approach gifted black students: Promising practices and pro- to conceptualizing intelligence has considerable grams . New York: Teachers College Press. Greenfi eld, P. M. (1997). You can’t take it with you: Why limitations, especially for culturally and linguisti- ability assessments don’t cross cultures. American cally diverse populations, those with limited Psychologist, 52 , 1115–1124. English language skills, and children who are Hoover, H. D., Dunbar, S. B., & Frisbie, D. A. (2001). experiencing academic problems, like a learning Iowa tests of basic skills . Itasca: Riverside. Jensen, A. R. (1998). The g factor: The science of mental disability. The limited utility of the verbal/nonver- ability . Westport: Praeger. bal model for evaluation of specifi c intellectual Kaufman, A. S., & Lichtenberger, E. O. (2000). Essentials of problems associated with learning disabled (LD) WISC-III and SPPSI-R assessment . New York: Wiley. children’s academic failure has led some to argue Kavale, K. A., & Forness, S. R. (1984). A meta-analysis of the validity of Wechsler scale profi les and recatego- that intelligence tests are irrelevant to the diagno- rizations: Patterns or parodies? Learning Disability sis of learning disabilities (Siegle, 1989). Quarterly, 7 , 136–156. Similarly, Kaufman and Lichtenberger (2000 ) Lohman, D. F., & Hagen, E. P. (2001). Cognitive abilities concluded that WISC-III subtest profi les “do not test . Itasca: Riverside. Matarazzo, J. D. (1992). Psychological testing in the 21st have adequate power on which to base differential century. American Psychologist, 47, 1007–1018. diagnosis” (p. 205) for LD or attention-defi cit/ Naglieri, J. A. (1997). Naglieri nonverbal ability test . San hyperactivity disorder (ADHD). The recognition Antonio: The Psychological Corporation. that scores on a verbal/nonverbal test of intelli- Naglieri, J. A. (1999). Essentials of CAS assessment . Hoboken: Wiley. gence have not been especially helpful for diag- Naglieri, J. A., & Bornstein, B. T. (2003). Intelligence and nosis of LD or ADHD (Kavale and Forness 1984 ; achievement: Just how correlated are they? Journal of Kaufman and Lichtenberger 2000) should moti- Psychoeducational Assessment, 21 , 244–260. vate professionals to look elsewhere to understand Naglieri, J. A., & Ford, D. (2003). Addressing under- representation of gifted minority children using the these conditions. This was one of the goals of this Naglieri nonverbal ability test (NNAT). Gifted Child book – to stimulate thought about where we are in Quarterly, 47 , 155–160. the fi eld of intelligence and intelligence testing Naglieri, J. A., & Ronning, M. E. (2000). Comparison of and what are the most promising suggestions for white, African-American, Hispanic, and Asian children on the Naglieri nonverbal ability test. Psychological where we should go from here. We applaud the Assessment, 12 , 328–334. work of our colleagues in zoology, evolutionary Naglieri, J. A., Booth, A. L., & Winsler, A. (2004). science, psychology, and education to appreciate Comparison of Hispanic children with and without the genetic and evolutionary roots of intelligence limited english profi ciency on the Naglieri nonverbal ability test. Psychological Assessment, 16 , 81–84. and to move forward to defi ne intelligence. We Pintner, R. (1923). Intelligence testing . New York: Henry are confi dent that the next 50 years of intelligence Holt. research will usher a new age in our understand- Ramsey, M. C., & Reynolds, C. R. (2004). Relations between ing, evaluating, and enhancing intellectual intelligence and achievement tests. In G. Goldstein & S. Beers (Eds.), Comprehensive handbook of psychologi- development. cal assessment (pp. 25–50). New York: Wiley. 490 J.A. Naglieri and S. Goldstein

Ramirez, R. R., & Patricia de la Cruz, G. (2002). The Sternberg, R. (1988). The triarchic mind: A new theory of Hispanic population in the United States: March 2002 , intelligence . New York: Viking Press. Current population reports, P20-545, U.S. Census Whitmore, J. R. (1980). Giftedness, conﬂ ict, and under- Bureau, Washington, DC. achievement . Boston: Allyn & Bacon. Roid, G. (2003). Stanford-Binet (5th ed.). Itasca: Woodcock, R. W., Mc Grew, K. S., & Mather, N. (2001). Riverside. Woodcock-Johnson III tests of achievement . Itasca: Stano, J. F. (2004). Test review: Wechsler abbreviated Riverside Publishing. scale of intelligence. Rehabilitation Counseling Yoakum, C. S., & Yerkes, R. M. (1920). Army mental Bulletin, 48 , 56–57. tests . New York: Henry Holt.