Research Report 2005–2006

Home , Chinese classifier, Daniel Wolpert, Gerd Gigerenzer, German Cancer Aid

Umschlag_2007_Ora_070717.fhmx 18.07.2007 14:06 Uhr Seite 1

C M Y CM MY CY CMY K

he Institute was founded in 1963 by Hellmut Becker, who was joined subsequently by Friedrich Edding (1964), Dietrich Goldschmidt (1964), and Saul B. Robinsohn (1964) as the first generation of scientific directors. In the first decade of its existence, the development of educational research and educational policy was emphasized. Research Report Research Report

The appointment of a second generation of directors 2005–2006 2005–2006 (Wolfgang Edelstein, 1973, and Peter M. Roeder, 1973) added to this framework a commitment to basic Research Report 2005-2006 he Max Planck Institute for Human Develop- research in human development and educational ment, founded in 1963, is a multidisciplinary processes. research establishment dedicated to the study of human development and education. Its Since the 1980s and with the appointment of a third inquiries are broadly defined, but concentrate generation of senior fellows and scientific directors Center for Adaptive Behavior and Cognition on the evolutionary, social, historical, and insti- (Paul B. Baltes, 1980; Karl Ulrich Mayer, 1983; Jürgen (Director: Gerd Gigerenzer) tutional contexts of human development. The Baumert, 1996; Gerd Gigerenzer, 1997), research at disciplines of education, psychology, and history the Institute has concentrated more and more on reflect the current directors’ backgrounds, but questions of basic research associated with the nature Center for Educational Research (Director: Jürgen Baumert) the Institute’s scholarly spectrum is enriched by of human development, education, and work in a the work of colleagues from such fields as be- changing society. At the same time, life-span develop- havioral developmental neuroscience, sociology mental and life-course research were added as a of education, mathematics, economics, computer signature profile of the Institute’s research program. Center for the History of Emotions (Director: Ute Frevert) (as of 2007) science, evolutionary biology, and the humanities. Latest developments in the succession of generations The Institute for Human Development is one of were marked by the appointment of Ulman Linden- about 80 research facilities financed by the Max berger as new director of the Center for Lifespan Center for Lifespan Psychology Planck Society for the Advancement of Science (Director: Ulman Lindenberger) Development (2004), adding an emphasis on neural (Max-Planck-Gesellschaft zur Förderung der correlates of human behavior and cognitive plastici- Wissenschaften e.V.), the core support for which ty, and by the appointment of Ute Frevert as director of the newly established Center for the History of Independent Junior Research Group is provided by the Federal Republic of Germany and its 16 states. Emotions (2007), adding perspectives from cultural Neurocognition of Decision Making

history to the Institute's research agenda on human (Head: Hauke Heekeren) Max Planck Institute for Human Development development. Continuity and change is also involved Max-Planck-Institut für Bildungsforschung in the establishment of a Max Planck International Research Network on Aging (MaxnetAging) directed International Max Planck Research School LIFE by Paul B. Baltes (✝ 2006). (Co-chairs: Jacquelynne S. Eccles – University of Michigan, Ulman Lindenberger – Berlin, John R. Nesselroade – University of Virginia)

Max-Planck-Institut für Bildungsforschung Max Planck Institute for Human Development Max Planck International Research Network on Aging (Director: Paul B. Baltes ✝) Lentzeallee 94 · 14195 Berlin, Germany Fon: +49-30/824 06-0 · Fax: +49-30/824 99 39 Max-Planck-Institut für Bildungsforschung http://www.mpib-berlin.mpg.de Max Planck Institute for Human Development

Probedruck 01_Seite01_001_001.qxd 18.07.2007 11:14 Uhr Seite 1

Research Report 2005–2006

Max-Planck-Institut für Bildungsforschung Max Planck Institute for Human Development 02_Board_Contents_002_005.qxd 18.07.2007 11:17 Uhr Seite 2

Impressum © 2007 Max Planck Institute for Human Development, Berlin Design: Grafisches Atelier Rudolf J. Schmitt, Berlin Realization: Jürgen Baumgarten, Dagmar Gülow, Marianne Hauser, Renate Hoffmann, Jürgen Rossbach of the Max Planck Institute for Human Development Printed 2007 by Mareis Druck GmbH, Weißenhorn/Bayern, Germany 02_Board_Contents_002_005.qxd 18.07.2007 11:17 Uhr Seite 3

Board of Directors Paul B. Baltes († November 7, 2006) Jürgen Baumert (Managing Director, until June 2005) Ute Frevert (as of June 2007) Gerd Gigerenzer (Managing Director, July 2005–August 2006) Ulman Lindenberger (Managing Director, September 2006–) Karl Ulrich Mayer (as of July 2005: Yale University)

Board of International Scientific Advisors Marlis Buchmann Jacobs Center for Productive Youth Development, Zurich Christian Büchel University Hospital, Hamburg-Eppendorf (2007–) Laura L. Carstensen Stanford University (chair, until 2005) Leda Cosmides University of California, Santa Barbara (until 2005) Hartmut Ditton Ludwig Maximilians University, Munich (2007–) Jacquelynne S. Eccles University of Michigan, Ann Arbor Klaus Fiedler University of Heidelberg Eric Johnson Columbia University, New York (2007–) Andreas Krapp Bundeswehr University, Munich (until 2005) Herbert W. Marsh Oxford University Walter Müller University of Mannheim (until 2005) Denise Park University of Illinois, Urbana-Champaign (chair) Manfred Prenzel Institute for Science Education at the University of Kiel Patricia A. Reuter-Lorenz University of Michigan, Ann Arbor (2007–) Anik de Ribaupierre-Bobillier University of Geneva (until 2005) Frank Rösler University of Marburg Wolfgang Schneider University of Würzburg 02_Board_Contents_002_005.qxd 18.07.2007 11:17 Uhr Seite 4

Contents 02_Board_Contents_002_005.qxd 18.07.2007 11:17 Uhr Seite 5

Introduction 6

Highlights 14

Cooperation with Universities 19 LIFE 21

MaxnetAging 28

Center for Adaptive Behavior and Cognition 31

Center for Educational Research 67

Center for Lifespan Psychology 127

Independent Junior Research Group Neurocognition of Decision Making 181

Service Units 201

Appendix 208 03_Introduction_006_013.qxd 19.07.2007 15:17 Uhr Seite 6

Introduction 03_Introduction_006_013.qxd 19.07.2007 15:17 Uhr Seite 7

Introduction | 7 03_Introduction_006_013.qxd 19.07.2007 15:17 Uhr Seite 8

8 | Introduction 03_Introduction_006_013.qxd 19.07.2007 15:17 Uhr Seite 9

Introduction

The Max Planck Institute for Human Development is a multidisciplinary research institution dedicated to the study of human development and education. Its inquiries encompass evolutionary, historical, social, and institutional contexts of individual human development from infancy to old age. The disciplines of education, history, and psychology, which reflect the current directors’ backgrounds, are enriched by the work of colleagues from computer science, developmental behavioral neuroscience, economics, evolutionary biology, mathematics, sociology, and the humanities. The Institute is one of about 80 research facilities financed by the Max Planck Society for the Advancement of Science (Max-Planck-Gesellschaft zur Förderung der Wissenschaften e.V.), the core support for which is provided by the Federal Republic of Germany and its 16 states. The total permanent staff at the Institute is 110, including 35 researchers, supplemented by a varying number of predoctoral, postdoctoral, and affiliate researchers and visiting fellows.

Research Centers from three perspectives: (1) the long-term Research into processes of human develop- consequences of schools’ opportunity struc- ment is conducted primarily from the theo- tures on individual development in terms of retical vantage points offered by models of cognitive competencies as well as motiva- bounded rationality and risk behavior, insti- tional and social resources, (2) international tutionalized learning, lifespan psychology, comparison of the outcomes of schooling in life-course sociology, and, as of 2007, cul- the fields of reading comprehension, mathe- tural history. matics and science literacy, and cross-curric- In the 2005–2006 period, research at the ular competencies, and (3) improvement of Institute was primarily organized in three learning and instruction in terms of the research centers: cognitive activation of students, mainly in The Center for Adaptive Behavior and science and mathematics (pp. 66–125). Cognition (Director: Gerd Gigerenzer) inves- The Center for Lifespan Psychology tigates human rationality, in particular deci- (Director: Ulman Lindenberger) is character- sion making and risk perception in an uncer- ized by a lifespan perspective and a concern tain world. Current research focuses on with the optimization of human potential. (1) bounded rationality, that is, the simple Empirical investigations of infants, children, heuristics—cognitive, emotional, and behav- adolescents, adults, and the elderly concen- ioral—that laypeople and experts use to make trate on the domains of cognition, sensori- decisions under constraints of limited time motor functions, affect, and motivation, the and knowledge, (2) social intelligence in co- interplay among these domains within indi- operation and competition, and (3) risk un- viduals of different ages, and on inter- derstanding and uncertainty management in personal action coordination and codevelop- everyday life, including applications in medi- ment. The plasticity of human behavior and cine, law, and education. Each of these re- the potential for successful development, search areas emphasizes the evolutionary including their neural and societal causes foundations of behavior and cognition, in and effects, play a central role in theory, particular their domain specificity and func- research design, and methodology tional adaptiveness (pp. 31–65). (pp. 126–179). The Center for Educational Research In 2007, the Center for the History of (Director: Jürgen Baumert) examines learning Emotions (Director: Ute Frevert) will be and development from an institutional point added to the research profile of the Institute. of view. Educational settings, such as schools, Karl Ulrich Mayer, the Director of the former offer a variety of developmental opportuni- Center for Sociology and the Study of the ties, but, at the same time, exclude others. Life Course, left the Institute in July 2005. He The impact of such settings is investigated is now at Yale University, New Haven, USA,

Introduction | 9 03_Introduction_006_013.qxd 19.07.2007 15:17 Uhr Seite 10

where he directs the Center for Research on Paul B. Baltes (1939–2006) Inequalities and the Life Course. The Institute parts in gratitude from Paul B. The Institute also houses an Independent Baltes, who passed away on November 7, Junior Research Group on the Neuro- 2006, at the age of 67 years, after a fierce cognition of Decision Making (Head: Hauke battle against cancer. An Academic Memorial Heekeren). Using a combination of psycho- in his honor took place in January 2007 (see physical methods, functional and structural Program, pp. 11–12). Paul Baltes directed the neuroimaging, modeling, and pharmacologi- Center for Lifespan Development from cal intervention, this group investigates 1980–2004, and became the Founding Direc- mechanisms of decision making in the human tor of MaxnetAging in 2005. Paul Baltes was brain (pp. 180–199). an innovator who reshaped the fields of life- An important collaborative effort involving span psychology and gerontology. His empiri- all Centers at the Institute as well as univer- cal and theoretical contributions have sities in Berlin and the United States is the opened up new perspectives and pathways International Max Planck Research School for science and society, and his contributions “The Life Course: Evolutionary and Onto- to the welfare of the Institute are immense. genetic Dynamics (LIFE),“ which is currently co-chaired by Ulman Lindenberger, Berlin, Jacquelynne S. Eccles, University of Michigan, and John R. Nesselroade, University of Virginia. This interdisciplinary graduate program brings together doctoral students from the United States and Germany but also from many other countries (see pp. 21–26). Another international project organized by the Max Planck Society and housed at the In- stitute is the International Max Planck Re- search Network on Aging (MaxnetAging). This network was directed by the late Paul B. Baltes (Deputy Directors: Shu-Chen Li and Jacqui Smith). The general focus of the network is on research, theory, and method of the behavioral and social science of aging. Specific objectives include the development of new interdisciplinary research initiatives, and providing a forum for promoting the careers of young scientists interested in the study of aging. The institutional collaboration involves several Max Planck Institutes, the Karolinska Institute, Stockholm, and the Uni- versity of Virginia (see pp. 28–29). Continued p. 13

10 | Introduction 03_Introduction_006_013.qxd 19.07.2007 15:17 Uhr Seite 11

Introduction | 11 03_Introduction_006_013.qxd 19.07.2007 15:17 Uhr Seite 12

12 | Introduction 03_Introduction_006_013.qxd 19.07.2007 15:17 Uhr Seite 13

Governance of the Institute Organization of the Annual Report The Institute is governed by a Board of This research report is organized in the fol- Directors, currently consisting of the four lowing manner: members of the Institute who are Fellows • The presentation of each research center (Wissenschaftliche Mitglieder) of the Max begins with an introductory overview sum- Planck Society (Jürgen Baumert, Ute Frevert, marizing its program. Gerd Gigerenzer, Ulman Lindenberger). • The introduction is followed by descrip- The Board is augmented by one member from tions of the center’s research areas and the Institute’s research staff (Ulrich selected projects along with a list of scien- Trautwein) and the Head of Administration tific publications. (Olaf von Maydell). On a rotational basis, one • The supportive activities of the service of the Directors is elected by the Board to units—library and computing services—are serve as Managing Director, usually for a described in a special section at the end of period of two years. the report (pp. 203–206). Several in-house committees composed of • The appendix provides information on the representatives elected by the entire scien- research colloquia held at the Institute, the tific staff or by appointment advise the Board visiting scholars, and the cooperation of of Directors on matters of scientific research the Institute’s scientific staff with projects and policy. One of the major institute-wide outside the Institute. It also includes committees is the Scientific Staff Committee an index of the scientific staff and their (Mitarbeiterausschuss), which is elected by research interests (pp. 208–227). all scientists. The International Board of Scientific Advisors offers an important source of external review and advice to both the Directors and the scientific staff on matters of research at the Institute. Members are selected from an international circle of distinguished researchers and appointed by the President of the Max Planck Society. They meet biannually to discuss completed, ongoing, and future research projects at the Institute. A list of the current members can be found on the front matter of this report.

Berlin, June 2007 For the Board of Directors: Ulman Lindenberger

Introduction | 13 04_Highlights_014_017.qxd 19.07.2007 15:18 Uhr Seite 14 (Schwarz/Process Black Auszug)

Highlights 04_Highlights_014_017.qxd 19.07.2007 15:18 Uhr Seite 15 (Schwarz/Process Black Auszug)

Honors and Awards 2005–2006

Paul B. Baltes Award for Distinguished Career Contributions to the Study of Human Development across the Life Span, Society for the Study of Human Development Jürgen Baumert Vice-President, Max Planck Society for the Advancement of Science Jürgen Baumert EARLI Oeuvre Award, European Association for Research on Learning and Instruction Jürgen Baumert National Leadership Award, Economic Forum Germany Natalie C. Ebner Heinz Heckhausen Prize for Young Scientists, German Psychological Association Wolfgang Edelstein Order of Merit of the Federal Republic of Germany Wolfgang Gaissmaier New Investigator Award, Brunswik Society Gerd Gigerenzer Honorary Professor of Psychology, Humboldt University Berlin Monika Keller Honorary Professor of Psychology, Free University Berlin Ulman Lindenberger Member, German Academy of Sciences Leopoldina, Section "Empirical Psychology and Cognitive Sciences" Martin Lövdén Sofja Kovalevskaja Prize, Alexander von Humboldt Foundation Jacqui Smith New South Wales Return Award for an Expatriate Researcher Elsbeth Stern Invited contribution to the yearbook "Highlights of German Science 2005", edited by the German Universities Association

Students Awards Nathalie C. Ebner Best Poster Prize, German Psychological Association, Section for Developmental Psychology Jutta Mata (Wittig) Springer Prize for Best Diploma Thesis, Humboldt University Berlin, Dept. of Psychology Gabriel Nagy Junior Researcher Award, German Psychological Association, Section for Educational Psychology Susanne Scheibe Otto Hahn Medal for Young Researchers, Max Planck Society Andreas Wilke Owen F. Aldis Award, International Society for Human Ethology 04_Highlights_014_017.qxd 19.07.2007 15:18 Uhr Seite 16 (Schwarz/Process Black Auszug)

Where Have Our Researchers Gone? New Positions 2005–2006

Research Scientists Cordula Artelt 2005, University of Bamberg, Full Professor of Pre- and Elementary School Research Arndt Bröder 2005, Max Planck Institute for Research on Collective Goods Martin Brunner 2006, University of Luxemburg, Assistant Professor Anja Dieckmann 2005, Institute for Research in Consumer Goods, Nuremberg Stefan Krauss 2007, University of Kassel, Assistant Professor Martin Lövdén 2006, Lund University, Associate Professor Britta Matthes 2005, Institute for Labor Market and Employment Research (IAB), Nuremberg, Research Scientist Antje Mertens 2005, Berlin School of Economics, Professor of Economics Maike Reimer 2005, Bavarian State Institute for Research in Higher Education, Munich, Research Scientist Florian Schmiedek 2006, Humboldt University Berlin, Junior Professor for Cognitive Developmental Psychology Holger Seibert 2005, Institute for Labor Market and Employment Research, Berlin- Brandenburg, Research Scientist Jacqui Smith 2006, University of Michigan, Ann Arbor, Full Professor of Psychology, and Institute for Social Research (ISR), Research Professor Heike Solga 2005, University of Göttingen, Full Professor of Sociology Petra Stanat 2005, University of Erlangen-Nuremberg, Full Professor of Empirical Research in Teaching and Instruction Elsbeth Stern 2006, Swiss Federal Institute of Technology (ETH), Zurich, Full Professor for Empirical Research in Teaching and Instruction Masanori Takezawa 2006, Tilburg University, Faculty of Social and Behavioural Sciences, Lecturer Peter Todd 2005, Indiana University, School of Informatics, Professor of Informatics, Cognitive Science, and Psychology Heike Trappe 2005, German Institute for Economic Research (DIW), Managing Director Council of Socio-economic Data Rainer Watermann 2005, University of Göttingen, Full Professor of Educational Sciences

Postdoctoral Research Fellows Will Bennis 2005, Northwestern University, Postdoctoral Researcher Natalie C. Ebner 2007, Yale University, Postdoctoral Researcher Rocio Garcia-Retamero 2006, University of Granada, Department of Experimental Psychology, Associated Professor Denis Gerstorf 2005, University of Virginia, Assistant Professor Shenghua Luan 2006, Singapore Management University, Social Sciences and Humanities, Assistant Professor Rui Mata 2006, University of Michigan, Ann Arbor, Department of Psychology, Postdoctoral Researcher Thorsten Pachur 2006, University of Basel, Department of Psychology, Assistant Professor Justin J. W. Powell 2005, University of Göttingen, Assistant Professor Torsten Reimer 2006, University of Maryland, Department of Communication, Assistant Professor Susanne Scheibe 2007, Stanford University, Postdoctoral Researcher Andreas Wilke 2006, University of California, Los Angeles, Postdoctoral Researcher

16 | Highlights 04_Highlights_014_017.qxd 19.07.2007 15:18 Uhr Seite 17 (Schwarz/Process Black Auszug)

Predoctoral Research Fellows Yvonne Brehmer 2006, Karolinska Institute Stockholm, Postdoctoral Researcher Katarzyna Buzcek 2006, University of Warszaw Zoe Daniels 2005, Rehabilitation Hospital Garder-See, Lohmen Daniel Grühn 2006, University of Geneva, Postdoctoral Researcher Michaela Gummerum 2005, University of Vancouver, Postdoctoral Researcher Stefanie Gundert 2006, Yale University, Postdoctoral Researcher Anke Höhne 2005, University of Halle-Wittenberg, Assistant Professor Tim Johnson 2006, Stanford University Christina K. Limbird 2006, Berlin-Brandenburg International School Andrea G. Müller 2007, German Youth Institute (DJI), Munich Lisa Pfahl 2005, University of Göttingen, Assistant Professor Christina Röcke 2006, Brandeis University, Boston, Postdoctoral Researcher Henrik Saalbach 2006, Swiss Federal Institute of Technology (ETH), Zurich, Postdoctoral Researcher Michael Schneider 2006, Swiss Federal Institute of Technology (ETH), Zurich, Postdoctoral Researcher

Highlights | 17 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 18 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 19

Cooperation with Universities 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 20

Teaching and Academic Degrees

The Institute has always considered its cooperation with universities as very important, especially by participating in teaching activities. Researchers from our Institute teach courses at three universities in Berlin and at the University of Potsdam as well as at many other universities in Germany and abroad. In the years 2005–2006, more than 100 courses were taught by scientific staff members —directors, research scientists, postdoctoral as well as predoctoral fellows—of the Institute. In addition, Institute members were supported in completing their academic degrees in cooperation with the universities in Berlin and elsewhere. In the years 2005–2006, 6 habilitations and 24 doctoral dissertations were completed by scientific staff members of the Insti- tute. All degrees are listed in the Appendix.

Exhibition in the Institute’s entrance hall. Sculptures by Özgür Emeklier

20 | Cooperation with Universities 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 21

International Max Planck Research School The Life Course: Evolutionary and Ontogenetic Dynamics (LIFE)

This graduate program on the Life Course (LIFE) is part of the Max Planck Society's framework of International Max Planck Research Schools. It was established in 2002 as a collaboration between the Max Planck Institute for Human Development, Berlin, the Humboldt University Berlin, the Free University Berlin, and the University of Michigan, Ann Arbor. The University of Virginia, Charlottesville, joined in October 2004.

The goal of the Research School is advanced The strong interlocking components are two Co-chairs research training in the study of human annual weeklong academies in which fellows Paul B. Baltes † behavior and institutional systems over evo- and faculty from each university participate. (MPI for Human lutionary and ontogenetic (life cycle) time. During the report period, four such acade- Development, until LIFE takes an integrative and interdisciplinary mies took place. The LIFE Spring Academy 11/2006), Ulman approach to understanding human develop- 2005 was held at the University of Virginia, Lindenberger (MPI ment in a changing world, connecting evolu- the LIFE Fall Academy 2005 took place at the for Human Develop- tionary, ontogenetic, historical, and institu- University of Michigan. In 2006, the Spring ment, since 09/2006) tional perspectives. Academy was held at the MPI for Human Jacquelynne S. Eccles The target groups of the Research School are Development in Berlin, and the Fall Academy (University of post-diploma or post-master's graduate stu- took place at the University of Virginia. Gen- Michigan), & John R. dents who intend to pursue a doctorate in erally, the average number of participants Nesselroade (Univer- one of the relevant disciplines (biology, psy- including faculty was about 50. The teaching sity of Virginia) chology, neuroscience, sociology, anthro- faculty consisted of faculty members from pology, educational science). As a collabora- Ann Arbor, Berlin, and Charlottesville as well Coordinators tive Research School, LIFE offers students a as several guests from other institutions. The Julia Delius (MPI for unique education experience: Discipline- next Academy is scheduled to take place in Human Develop- based training in the study of the life Berlin in May 2007. ment, until 09/2006), course/life cycle that is enriched by inter- In addition, each participating university Sandra Fronz und disciplinary and international perspectives. offers special courses reflecting the special Christina Röcke (MPI The training program involves seminars at the profile of its graduate programs and selects a for Human Develop- participating institutions, a series of Fall and subset of fellows for the added specialization ment, interim), Spring Academies, and collaborative super- provided by LIFE. The Berlin approach is to Imke Kruse (MPI for vision of research training. It also includes offer weekly seminars at the MPI for Human Human Develop- opportunities for research abroad at a Development taught throughout the aca- ment, since cooperating institution. Five fellows from demic year by a varying team of faculty from 10/2006), Janice Berlin have made use of this in 2005 and the three Berlin institutions. Templeton (Univer- 2006. sity of Michigan), Juanita L. Geer (Uni- versity of Virginia) In total, 69 faculty (Berlin: 24; Ann Arbor: 33; Charlottesville: 12), 51 fellows (Berlin: 27; Ann Arbor: 13; Charlottesville: 11) have been involved as of December 2006. In addition, 20 fellows (Berlin: 14; Ann Arbor: 4; Charlottesville: 2) completed their dissertation during the reporting period. Fellows joined the program on the following annual schedule: www.imprs-life.mpg.de • 2002: 18 (Berlin: 14; Ann Arbor: 4) • 2003: 12 (Berlin: 7; Ann Arbor: 5) • 2004: 19 (Berlin: 10; Ann Arbor: 4; Charlottesville: 5) • 2005: 3 (Ann Arbor) • 2006: 23 (Berlin: 11; Ann Arbor: 4; Charlottesville: 8)

Cooperation with Universities | 21 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 22

During the Spring Academy 2006, an inde- Wolfgang Prinz, MPI for Human Cognitive pendent Scientific Committee evaluated the and Brain Science, and Herbert Festl, Unit for LIFE Research School very positively. The Research Strategy at the Max Planck Society, Committee consisted of the following mem- participated in the evaluation. The reviewers bers: Herbert Marsh, University of Western enthusiastically recommended continuing the Sydney, Denise Park, University of Illinois at IMPRS for another six-year period until 2013. Urbana-Champaign, Wolfgang Schneider, For more details, see the LIFE Evaluation University of Würzburg, and Hannelore Report 2006. Weber, University of Greifswald. In addition,

Topics of the Berlin LIFE seminars have included: • The Biology of Life History, Peter Hammerstein (HU) & guests • Contexts and Dynamics of Behavior and Cognition in the Life Course, Ralf Krampe (MPI), Shu-Chen Li (MPI), Peter Todd (MPI), & guests • The Life Course: Structures and Institutions, Martin Kohli (FU), Felix Büchel (MPI), Heike Solga (MPI), & guests • Developmental Methodology, Ulman Lindenberger (MPI-FU-HU) & guests • Norms and Development: Interdisciplinary Approaches, Monika Keller (MPI), Masanori Takezawa (MPI), & Gerd Gigerenzer (MPI-FU) • Biological, Psychological, and Social Constraints on the Development of Learning across the Life Course, Jens B. Asendorpf (HU), John M. C. Hutchinson (MPI), Frieder R. Lang (Universität Halle), Shu-Chen Li (MPI), Justin Powell (MPI/University of Göttingen), Alessandra Rusconi (HEMPAS/University of Bremen), Petra Stanat (MPI), and Elsbeth Stern (MPI) • Theoretical and Methodological Approaches to the Study of the Life Cycle: Inter- disciplinary Perspectives, Paul B. Baltes (MPI-FU), John M.C. Hutchinson (MPI), Ulman Lindenberger (MPI-FU-HU), Martin Kohli (FU), Ulrich Trautwein (MPI), & guests • The Social Psychology and Sociology of Adult Development and Aging, Jacqui Smith (MPI-FU) & guests • Research Methods, Ulman Lindenberger (MPI-FU-HU) & guests

LIFE Spring Academy 2006 at Max Planck Institute for Human Development

22 | Cooperation with Universities 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 23

MPI for Human Development Jeff Hutsler, Psychology Professorial Paul B. Baltes †, Psychology & Gerontology James S. Jackson, Social Psychology Faculty 2005/2006 (until 11/2006) Daniel Keating, Psychology Jürgen Baumert, Educational Science Shinobu Kitayama, Psychology Gerd Gigerenzer, Psychology Daniel Kruger, Psychology & Public Health Shu-Chen Li, Psychology Bobbi S. Low, Evolutionary & Behavioral Ulman Lindenberger, Psychology Ecology Karl Ulrich Mayer, Sociology (until summer Betsy Lozoff, Pediatrics 2005) Ram Mahalingam, Psychology Lael J. Schooler, Psychology Kevin F. Miller, Psychology & Education Jacqui Smith, Psychology (until summer Maria Muzik, Psychiatry 2005) Randolph M. Nesse, Psychology & Psychiatry Ulrich Trautwein, Educational Science Richard Nisbett, Psychology Peter Todd, Psychology (until summer 2005) Sheryl Olson, Psychology Thad Polk, Psychology Free University Berlin Patricia A. Reuter-Lorenz, Cognitive Michael Eid, Psychology (since 2006) Psychology & Neuroscience Arthur Jacobs, Psychology (since 2006) Arnold Sameroff, Psychology Martin Kohli, Sociology (currently European John Schulenberg, Psychology University Institute, Florence) Rachael Seidler, Psychology & Kinesiology Hans Merkens, Educational Science Jacqui Smith, Psychology (since 2006) Herbert Scheithauer, Psychology Abigail Stewart, Psychology Ralf Schwarzer, Psychology & Health Beverly Strassman, Anthropology Clemens Tesch-Römer, Psychology & Twila Tardif, Psychology Gerontology Arland Thorton, Sociology Henry M. Wellman, Psychology Humboldt University Berlin Robert Zucker, Psychology & Psychiatry Jens B. Asendorpf, Psychology Hans Bertram, Sociology University of Virginia Peter Frensch, Psychology Steven M. Boker, Cognitive & Quantitative Peter Hammerstein, Biology Psychology Gerd Kempermann, Neuroscience Gerald L. Clore, Social Psychology Olaf Köller, Psychology & Education (since Judy DeLoache, Developmental Psychology 2006) Chad Dodson, Cognitive Psychology Rainer H. Lehmann, Educational Science David L. Hill, Psychobiology Florian Schmiedek, Psychology (since 2006) Angeline Lillard, Developmental Psychology Arno Villringer, Psychology John J. McArdle, Quantitative & Developmen- Oliver Wilhelm, Psychology & Education tal Psychology (until summer 2005) John R. Nesselroade, Quantitative & University of Michigan Developmental Psychology Toni C. Antonucci, Psychology Robert C. Pianta, Education Don Brown, Psychology Timothy Salthouse, Cognitive Psychology Cleopatra Howard Caldwell, Health & PRBA Eric Turkheimer, Quantitative & Clinical Kai S. Cortina, Psychology & Education Psychology Jacquelynne S. Eccles, Psychology & Timothy D. Wilson, Social Psychology Education David L. Featherman, Sociology & Social Psychology Richard Gonzalez, Psychology L. Rowell Huesmann, Psychology & Communication Studies

Cooperation with Universities | 23 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 24

LIFE Doctoral Max Planck Institute for Human Humboldt University Berlin Fellows Development Tobias Bothe, Psychology 2005/2006 Michael Becker (External LIFE Fellow), Daniel Caro Vasquez, Educational Science Psychology Robert Gaschler (External LIFE Fellow), Annette Brose, Psychology Psychology Cornelia Gresch (External LIFE Fellow), Fanny Jimenez, Psychology Sociology Judith Lehnart (External LIFE Fellow), Bettina von Helversen, Psychology Psychology Oliver Huxhold, Psychology Lars Penke (External LIFE Fellow), Psychology Kathrin Jonkmann (External LIFE Fellow), Dennis Rünger, Psychology Psychology Dana Kotter-Grühn (External LIFE Fellow), University of Michigan Psychology Joaquin Anguera, Psychology Ana Sofia Alves Conte de Morais, Psychology Marie Burrage, Education & Psychology Irene Nagel (External LIFE Fellow), Psychology Heather Fuller, Psychology Lisa Pfahl, Sociology Jessica Garrett, Educational Science Yee Lee Shing, Psychology Ashley Hazel, Anthropology Yi-Miau Tsai, Educational Science Justin Jager, Psychology Imke Vollmar (Max Delbrück Center), Xuezhao Lan, Education & Psychology Molecular Biotechnology Cristine Legare, Psychology Julius Verrel, Cognitive Neuroscience Emily Messersmith, Psychology Markus Werkle-Bergner (External LIFE Nicky Newton, Psychology Fellow), Psychology Georges Potworowski, Psychology Besangie Sellars, Psychology Free University Berlin Christy Watkins, Psychology Poldi Kuhl (External LIFE Fellow), Educational Science University of Virginia Jutta Mata, Psychology Sarah K. Galloway, Psychology Tabea Reuter, Psychology Jesse C. Graham, Psychology Kathryn Paige Harden, Psychology Selin Kesebir, Psychology Vanessa LoBue, Psychology Joshua C. Magee, Psychology Jamie Mangold, Psychology Felicity F. Miao, Psychology Elizabeth R. Tenney, Psychology Elliot Tucker-Drob, Psychology Lijuan (Peggy) Wang, Psychology

LIFE fellows and faculty members in a poster session.

24 | Cooperation with Universities 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 25

Table 1 Dissertations Completed (2005–May 2007)

Fellow Dissertation title Institution Discipline Current position

Guido Biele No Man is an Island: Cooperation in Groups MPI/FU Psychology MPI for Human Development and Social Learning Yvonne Brehmer Episodic Memory Plasticity Across the Lifespan MPI/HU Psychology Karolinska Institute Stockholm Jaap J.A. Denissen Understanding and Being Understood: The HU Psychology Utrecht University Impact of Intelligence and Dispositional Valuations on Social Relationships Sean Duffy Seeing Through Cultures UM Psychology Rutgers University-Camden Katherine Leigh Fiori Social Relations and Health Among Older UM Psychology Long Island University Adults: A Cross-Cultural and Pattern-Centered Approach Jessica Garrett Motivation in the Transition to Adulthood UM Psychology University of Michigan Michaela Gummerum The Development of Prosocial Behavior: MPI/FU Psychology University of British Columbia Integrating Psychological, Economic, and Evolutionary Perspectives Stefanie Gundert Short-Term Contracts of Entrants and Older MPI/FU Sociology Yale University Employees Jung-Hwa Ha Determinants and Consequences of Changing UM Sociology Waisman Center Wisconsin Social Support Following Late Life Widowhood Nicole Hess Informational Warfare: Female Friendship and HU Biological Humboldt-University Berlin the Coalitional Manipulation of Reputation Anthropology Rainer Heuer Policy Within the Family—Intergenerational FU Sociology Social Science Research Center Power at Middle and Older Age. The Example Berlin of Parents' Physical Losses Oliver Huxhold Processing Fluctuations in Postural Control: FU Psychology German Center of Gerontology Relations to Adult Age and Fluctuation in Cognition Helen Krumme Parent-Adult Relationship in Later Life: A FU Sociology Social Science Research Center Comparative Study on Migrant and Non- Berlin Migrant Families in Germany Christina Karlyn Limbird Phonological Processing, Verbal Abilities, and FU Educational Berlin-Brandenburg Second Language Literacy Development Among Science International School Bilingual Turkish Children in Germany Rui Mata The Aging Decision Maker: Cognitive Aging MPI/FU Psychology University of Michigan and the Use of Decision Strategies Emily Messersmith Longitudinal Correlates of Changing UM Psychology University of Michigan Educational Expectations During the Transition to Adulthood Andrea Müller Barriers to Reaching Proficiency in School- FU Psychology German Youth Institute Related Language Among Second Language Learners Gabriel Nagy The Relevance of Professional Interests, MPI/FU Psychology MPI for Human Development Cognitive and Field-Specific Competences for the Choice of Field of Study and the Success of Study Pablo A. Nepomnaschy Stress and Female Reproduction in a Rural UM Anthropology National Institute of Mayan Population Environmental Health Sciences Thorsten Pachur Ecological Rationality: Do Samples in Memory MPI/FU Psychology University of Basel Reflect the World? Nilam Ram Emphasizing Individuality in Developmental UVa Psychology Pennsylvania State University Models of Change Christina Röcke Intraindividual Variability in Positive and MPI/FU Psychology Brandeis University Negative Affect: Age-Related and Individual Differences in Magnitude and Coupling with Cognitive Performance

Cooperation with Universities | 25 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 26

Table 1 (continued) Dissertations Completed (2005–May 2007)

Fellow Dissertation title Institution Discipline Current position

Sabine Schäfer Concurrent Cognitive and Sensorimotor MPI/FU Psychology MPI for Human Development Performance: A Comparison of Children and Young Adults Karen Siedlecki Episodic Memory and the Retrieval of UVa Psychology Columbia University Information Across the Adult Lifespan Andreas Wilke Evolved Responses to an Uncertain World MPI/FU Psychology University of California, Los Angeles Nicole Zarrett The Dynamic Relation Between Out-of-School UM Psychology Tufts University Activities and Adolescent Development

Presentation of LIFE Certificates at the first LIFE Commencement Event.

26 | Cooperation with Universities 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 27

Cooperation with Universities | 27 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 28

MaxnetAging

28 | Cooperation with Universities 05_Cooperation_018_029.qxd 18.07.2007 13:26 Uhr Seite 29

Based on deliberations in the Max Planck The four profile topics of the Network are: Director: Presidential Committee on Aging, the Max (1) Images and regulatory systems of the life Paul B. Baltes Planck International Research Network on course and aging: Historicial, legal, social, († 2006) Aging was established in the Fall of 2004. and psychological The initial time window for this Network is (2) Law and aging: General and specific Collaboration: five years (2005–2009). During this period, (3) Plasticity in old and very old age: Neu- Human Sciences the substantive task of the Network is to fo- ronal, behavioral, and societal perspec- Section of the Max cus on the behavioral and social sciences of tives Planck Society, aging, broadly defined, with topical emphases (4) Dynamics of mortality and morbidity: Karolinska Institute, on research about aging reflected in such Biological, psychological, and social vital- Stockholm, fields as cognitive and social neuroscience, ity in tthe oldest ages University of demography, psychology, law, history of sci- Semi-annual meetings, focusing on topics of Virginia ence, cultural studies, sociology, economics, human aging ranging across the full spec- and the humanities. The Network consists of trum of the behavioral and social sciences a core permanent group of Senior and Junior and the humanities, provide the primary basis www.maxnetaging.mpg.de Fellows. Junior Fellows can apply for seed for fostering cross-disciplinary discussion and funds to support new research initiatives. collaboration. The main Network office is MaxnetAging is conceived as a forum in located at the Max Planck Institute for which cross-disciplinary communication and Human Development in Berlin. Shu-Chen Li is cooperation in the study of aging is nurtured the Behavioral Sciences Deputy Director and within the Human Sciences Section of the Jacqui Smith (University of Michigan) is the Max Planck Society and as a platform for Social Sciences Deputy Director. international collaboration. Nine Max Planck Institutes have joined the network. Inter- national connections have been established with the University of Virginia and the Karolinska Institute, Stockholm.

Cooperation with Universities | 29 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 30 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 31

Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 32

Contents

Introductory Overview ...... 33

Bounded Rationality ...... 35

Ecological Rationality ...... 39

Social Rationality ...... 45

Evolutionary Psychology ...... 50

Decision Making in the Wild ...... 53

Publications 2005–2006 ...... 59

Scientific Staff 2005–2006

Nathan Berg, Henry J. Brighton, Uwe Czienskowski, Anja Dieckmann (as of 2005: GFK, Insti- tute for Research in Consumer Goods), Gerd Gigerenzer, John M. C. Hutchinson, Konstantinos V. Katsikopoulos, Monika Keller, Henrik Olsson, Jörg Rieskamp, Lael J. Schooler, Jeffrey Stevens, Peter M. Todd (as of 2005: Indiana University)

Postdoctoral Research Fellows Will M. Bennis (as of 2005: Northwestern University), Juliet Conlin, Mirta Galesic, Rocio García-Retamero (as of 2006: University of Granada), Shenhua Luan (as of 2006: Singapore Management University), Rui Mata (as of 2006: University of Michigan, Ann Arbor), Thorsten Pachur (as of 2006: University of Basel), Magnus Persson, Jing Qian, Torsten Reimer (as of 2006: University of Maryland), Julia Schooler, Masanori Takezawa (as of 2006: Tilburg Univer- sity), Andreas Wilke (as of 2006: University of California, Los Angeles), Odette Wegwarth

Predoctoral Research Fellows Ana Sofia Alves Conte de Morais (LIFE), Adrien Barton, Guido Biele (as of 2006: Independent Junior Research Group Neurocognition of Decision Making at MPI), Katarzyna Buzcek (as of 2006: University of Warszaw), Linan Diao, Wolfgang Gaissmaier, Michaela Gummerum (as of 2005: University of Vancouver), Bettina von Helversen (LIFE), Tim Johnson (as of 2006: Stanford University) Julian N. Marewski, Jutta Mata (Wittig, LIFE), Benjamin Scheibehenne, Nils Straubinger

32 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 33

Introductory Overview

The Center for Adaptive Behavior and Cognition (ABC) investigates reasoning and decision making under uncertainty at the levels of both individuals and social groups. The research group consists of psychologists, mathematicians, computer scientists, evolutionary biologists, economists, and researchers from other fields. With different methodological abilities, such as experimental methods, computer simulation, and mathematical analysis, they cooperate in solving the same problems. The ABC program combines a strong theoretical focus with practical applications, that is, the research group both develops specific models and explores their applications. Applications range from helping physicians and patients understand the statistical evidence arising from medical research, helping courts, administrators, and legislators understand the importance of heuristic thinking in the law, and improving teaching practices in statistical education by introducing transparent representation formats. The theoretical focus is on rationality and can be, albeit artificially, divided into three aspects: bounded, ecological, and social rationality.

Bounded Rationality the adaptation of mental and social strate- Models of bounded rationality attempt to an- gies to real-world environments rather than swer the question of how people with limited compare strategies to the laws of logic and time, knowledge, money, and other scarce probability theory. resources make decisions. This program is an alternative to the dominant optimization Social Rationality paradigm in cognitive science, economics, Social rationality is a variant of ecological ra- and behavioral biology that poses the ques- tionality, one for which the environment is tion of how Laplacean superintelligences or social rather than physical or technical. Mod- near omniscient beings would behave. We els of social rationality describe the structure study the proximal mechanisms of bounded of social environments and their match with rationality, that is, the adaptive heuristics boundedly rational strategies that people that enable quick and frugal decisions under might use. There is a variety of goals and uncertainty. This collection of heuristics and heuristics unique to social environments. their building blocks is what we call the That is, in addition to the goals that define adaptive toolbox. ecological rationality—to make fast, frugal, and fairly accurate decisions—social rational- Ecological Rationality ity is concerned with goals, such as choosing Models of ecological rationality describe the an option that one can defend with argu- structure and representation of information ment or moral justification, or that can cre- in actual environments and their match with ate a consensus. To a much higher degree mental strategies, such as boundedly rational than for the purely cognitive focus of most heuristics. To the degree that such a match research on bounded rationality, socially exists, heuristics need not trade accuracy for adaptive heuristics include emotions and so- speed and frugality. The simultaneous focus cial norms that can act as heuristic principles on the mind and its environment, past and for decision making. present, puts research on decision making under uncertainty into an evolutionary and ecological framework, a framework that is missing in most theories of reasoning, both descriptive and normative. In short, we study

Center for Adaptive Behavior and Cognition | 33 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 34

These three notions of rationality, according to which the present text is largely structured, converge on the same central issue: to understand human behavior and cognition as adaptations to specific environments, ecological and social, and to discover the heuristics that guide adaptive behavior. In a fourth section, we report on work that directly relates to evolutionary psychology, which, as a metatheoretical framework, lies behind the "adaptive" in our Center's name. The research reported in the last section focuses on the applications of our research, and how it can impact on “decision making in the wild” in domains such as the law, medicine, and education. The ABC program explores an exciting and very different perspective on rationality. Our perspective questions the familiar one we know from the many stories in cognitive science and economics, which often implicitly assume that humans and animals have unlimited time, knowledge, and operate according to the laws of logic and probability theory. The land of rationality we set out to explore is, in contrast, inhabited by people endowed with only limited time, knowledge, and computational capacities with which to make inferences about the world. The notions of bounded, ecological, and social rationality are our guides to understanding how humble humans survive without following the heavenly rules of rational choice theory.

34 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 35

Bounded Rationality

Humans and animals must make inferences about unknown features of their world under constraints of limited time, knowledge, and computational capacities. We do not conceive bounded rationality as optimization under constraints nor do we think of bounded rationality as the study of how people fail to meet normative ideals. Rather, bounded rationality is the key to understanding how people make decisions without utilities and probabilities. Bounded rationality consists of simple step-by-step rules that function well under the constraints of limited search, knowledge, and time—whether an optimal procedure is available or not. Just as a mechanic will pull out specific wrenches, pliers, and gap gauges to maintain an engine rather than just hit everything with a hammer, different domains of thought require different specialized tools. The notion of a toolbox full of unique single-function devices lacks the beauty of Leibniz’s dream of a single all-purpose inferential power tool. Instead, it evokes the abilities of a craftsman, who can provide serviceable solutions to almost any problem with just what is at hand.

The Adaptive Toolbox cognitive psychology, and have applied them This repertoire of specialized cognitive mech- in the areas of consumer behavior, medicine, anisms, which include fast and frugal heuris- and the law. tics, were shaped by evolution, learning, and culture for specific domains of inference and The Priority Heuristic Key Reference reasoning. We call this collection of mecha- The priority heuristic is a simple and precise Brandstätter, E., nisms the “adaptive toolbox.” We clarify the model for choices among risky options Gigerenzer, G., & Hertwig, R. (2006). The concept of an adaptive toolbox as follows: (Brandstätter, Gigerenzer, & Hertwig, 2006). priority heuristic: Making – It refers to a specific group of rules or When making a choice between gambles, the choices without trade- heuristics rather than to a general-purpose priority heuristic assumes that each gamble offs. Psychological Review, 113, 409–432. decision-making algorithm. can be described by several aspects: its mini- – These heuristics are fast, frugal, and com- mum outcome, its maximum outcome, and putationally cheap rather than consistent, the probabilities of each particular outcome coherent, and general. actually occurring. Contrary to the expected – These heuristics are adapted to particular utility perspective, this heuristic does not environments, past or present, physical or assume that people form an overall evalua- social. tion of gambles by weighting outcomes – The heuristics in the adaptive toolbox are according to their probabilities of occurring. orchestrated by some mechanism reflect- Instead, it assumes that aspects of the ing the importance of conflicting motiva- gambles are compared sequentially, and that tions and goals. each of these comparisons can lead to a decision. Fast and Frugal Heuristics When making a choice between two gam- Fast and frugal heuristics generally consist of bles, their “minimum outcomes” are com- three building blocks: simple rules for guiding pared first. If the difference between two search for information (in memory or in the outcomes exceeds 10 % of the maximum environment), for stopping search, and for payoff that can be obtained from either of decision making. They are effective when the two, the gamble with the larger mini- they exploit the structure of the information mum outcome is selected. For gambles with in the environment. That is, their rationality only positive payoffs, the minimum outcome is a form of “ecological rationality” rather refers to the minimum gain (i. e., the lowest than one of consistency and coherence. We payoff). If the difference does not exceed the have continued to explore fast and frugal threshold, the second aspect is examined: heuristics, and their relationship to diverse the probabilities of the minimum outcomes disciplines, such as biology, economics, and occurring. If the probabilities differ by more

Center for Adaptive Behavior and Cognition | 35 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 36

Key References than 10 %, the gamble with the lower proba- Suspending the Recognition Heuristic Pachur, T., & Hertwig, R. bility of the minimum outcome is selected. The assumption underlying the recognition (2006). On the psychol- Otherwise, the last aspect—the maximum heuristic—one of the central components of ogy of the recognition outcomes of the two gambles—is considered, the adaptive toolbox—is that recognition in- heuristic: Retrieval pri- macy as a key determi- and the gamble with the larger maximum formation serves as a central cue governing nant of its use. Journal outcome is selected. Brandstätter et al. inferences about the environment. But why of Experimental Psycol- (2006) compared the priority heuristic to sev- should recognition have such a prominence ogy: Learning, Memory, and Cognition, 32, eral variants of expected utility theory on in decision making? Pachur and Hertwig 983–1002. gambles from four studies. (2006) tested and found support for the Volz, K. G., Schooler, L., The priority heuristic was better able to pre- thesis that, due to the architecture of the J., Schubotz, R., Raab, dict the majority choice of participants for a cognitive system, recognition information is M., Gigerenzer, G., & von Crammon, D. Y. wide range of gambles. Brandstätter et al. provided quickly and at little cognitive cost, (2006). Why you think (2006) demonstrated that the systematic and therefore precedes the retrieval of other Milan is larger than study of adaptive heuristics, such as the pri- information from memory that can be used Modena: Neural correlates of the recognition ority heuristic, can provide a psychologically to make inferences. In addition to these heuristic. Journal of Cog- informed alternative to the expected utility demonstrations, they examined how people nitive Neuroscience, 18, framework, which has been taken as the deal with recognition information in environ- 1924–1936. foundation (and gold standard) for much of ments where the recognition heuristic often the work in economics and the psychology of leads to incorrect decisions. How might peo- decision making. ple constrain their use of recognition in such environments? A suspension hypothesis re- Recognition Heuristic ceived the strongest support. According to A strategy that uses recognition to make in- this hypothesis, the decision not to follow ferences about the environment is what recognition is made selectively and based on Goldstein and Gigerenzer (1999, 2002) called item-specific knowledge (rather than item- the recognition heuristic. For two-alternative general knowledge, such as the recognition choice tasks, where one has to decide which validity in the given environment). For exam- of two objects scores higher on a criterion, ple, even though you may have heard of lep- the heuristic can be stated as follows: rosy, but not shigellosis, you may still infer If you recognize one object but not the other, that shigellosis could be more prevalent in then infer that the recognized object has a Germany than leprosy, because you know higher value on the criterion. that leprosy is rare in the modern world. In Feeding on recognition, the recognition sum, these results provide evidence that, heuristic piggybacks on a highly efficient cog- among the different cues a person could use nitive ability, and also exploits an environ- to make an inference, recognition has a spe- mental regularity, namely, that recognition in cial status. Moreover, these results reveal natural environments is often systematic more about the psychology of the recognition rather than random. In these environments, heuristic and the mechanisms which underlie the heuristic is ecologically rational and its use as an adaptive tool. exemplifies Herbert Simon’s vision of ration- This suspension hypothesis also received sup- ality as being shaped by two blades, one be- port from the neuroimaging study of Volz ing the mind, the other being the environ- et. al. (2006). Understanding the neural cor- ment (Simon, 1990). There has been contin- relates which underlie decisions consistent ued progress in demonstrating the predictive with the recognition heuristic could help de- power of the recognition heuristic in, for in- termine whether people simply make a stance, soccer (Pachur & Biele, 2007) and ten- recognition judgment if one alternative is nis (Scheibehenne & Bröder, under review). In recognized but the other not. Alternatively, addition, we have further explored its cogni- there may be an additional evaluation tive and neural foundations, focusing on process which inhibits application. After when people choose the recognized alterna- checking the heuristic’s ecological validity for tive and when they do not. a given situation, the authors measured brain

36 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 37

activity by using Figure 1. In the fMRI functional Magnetic study by Volz et al. Neural correlates with the use of the recognition heuristic (2006), decisions in ac- Resonance Imaging cordance with the recog- (fMRI) while partici- Retrosplanial cortex nition heuristic correlate pants indicated with higher activation within the anterior fron- which of two cities tomedian cortex (aFMC), they thought was and the precuneus bilat- larger (Experiment 1), erally extending into the retrosplenial cortex. The or which city they aFMC mean percentage signal recognized (Experi- changes with error bars ment 2). Experi- for the aFMC for trials in which the recognition ment 2 revealed that heuristic was applied the decisional (red line) and was not processes cannot be applied (blue line) is x = –1 z = 0 shown in the accompa- reduced simply to a nying plot. judgment based on recognition memory. In addition, a deacti- aFMC vation was observed within the anterior FrontoMedian Cortex 0.2 (aFMC) when people did not follow the 0.0 recognition heuristic, but rather indicated that the unrecog- –0.2 nized alternative was % signal change larger. As the aFMC –0.4 has previously been associated with self- 0123456789101112131415 referential judg- t (in sec) ments, the authors concluded that the processes underlying the recognition heuristic involve an assess- The Benefits of Cognitive Limits ment about the applicability of the heuristic. The premise that the human capacity for A related interpretation of this deactivation information processing is limited is usually in aFMC is that acting against recognition accompanied by another assumption, namely, demands more cognitive effort than simply that these limitations pose a liability: They Key References choosing the recognized alternative. It seems constrain our cognitive potential. These limi- Gaissmaier, W., that the recognition heuristic is the default, tations bar us from performing feats, such as Schooler, L. J., & so to speak, but that contradicting source reciting the Iliad from memory or, for many Rieskamp, J. (2006). Simple predictions fueled knowledge and criterion knowledge can in- of us, remembering the three things we were by capacity limitations: hibit its use. to pick up at the store. Clearly, forgetfulness When are they success- Taken together the results of Pachur and is amongst our most troublesome cognitive ful? Journal of Experi- mental Psychology: Hertwig (2006) and Volz et al. (2006) suggest limitations over both the long-term (“what is Learning, Memory, and that processes underlying recognition-based his name?”) and short-term (“what tempera- Cognition, 32, 966–982. decisions go beyond automatically choosing ture did the recipe say to set the oven?”). In Schooler, L. J., & the recognized alternative, but include an contrast, we have examined the adaptive Hertwig, R. (2005). How forgetting aids heuristic additional judgment about whether such a nature of forgetting. inference. Psychological recognition-based decision is appropriate. Review, 112, 610–628.

Center for Adaptive Behavior and Cognition | 37 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 38

When is Long-Term Forgetting Beneficial? theory dictates a maximizing strategy; that Why don’t we have the perfect recall of a is, always choosing the green envelopes, or at computer memory chip? A few theorists have least, doing so after establishing that the argued that forgetting should not be seen as green envelopes are the better bet. Such a a limitation, but as key to the effective strategy maximizes the expected gain, which working of human memory. Essentially, for- is $ 70 in this example. However, people nor- getting prevents outdated information from mally adopt a strategy known as probability interfering with more recent information matching, which, in this example, means that is likely to be more relevant. Schooler choosing an envelope in proportion to its and Hertwig (2005) proposed that forgetting chances of being the right bet. Here, this may, in addition, prove beneficial for infer- would mean picking the green envelope in ence heuristics that exploit mnemonic infor- 70 % of the trials and the red in 30 %. On mation, such as recognition and retrieval average, this will yield only $ 58 = 0.7 x 70 fluency. To explore the mechanisms that link + 0.3 x 30. loss of information and heuristic perform- In such an envelope task, Kareev, Lieberman, ance, they implemented the recognition and Lev (1997) found that people with heuristic (Goldstein & Gigerenzer, 2002) and smaller short-term memory capacity were the fluency heuristic (e. g., Jacoby & Dallas, more likely to adopt a maximizing strategy 1981) in ACT-R (Anderson & Lebiere, 1998). than people with larger capacities. Kareev The ACT-R research program seeks to de- et al.’s original explanation was that those velop a coherent theory of cognition, speci- with low digit spans perceived the correla- fied to such a degree that phenomena from tions between the color of the envelope and perceptual search to the learning of algebra the probability of winning as more extreme can be modeled within the same framework. because their limited short-term memory ca- Using computer simulations, the authors pacity limited the size of the samples they demonstrated that forgetting boosts the ac- could draw inferences from. Such small sam- curacy of the recognition heuristic, which ples tend to overestimate correlations, and relies on systematic failures of recognition this overestimation can be advantageous in to infer which of two objects scores higher correlation detection. Instead of assuming on a criterion value. Similarly, simulations of that people differ in how they perceive the the fluency heuristic, which arrives at the correlations, Gaissmaier, Schooler, and same inference on the basis of the speed Rieskamp (2006) hypothesized that people with which the two objects are recognized, differ in the degree to which they explore indicate that forgetting helps maintain the their environment. Their modeling efforts discriminability of recognition speeds. Thus, showed that the low levels of exploratory the ignorance that forgetting brings can, behavior associated with having a smaller paradoxically, enhance inferences about ob- working memory capacity leads to a maxi- jects in the world. mizing strategy, whereas the more exploratory behavior associated with greater When is a Limited Short-Term Memory capacity leads to probability matching. Con- Capacity Beneficial? gruent with this hypothesis, two experiments Would we be better off if we could reason revealed a low-capacity advantage in a more logically, and remember more? The fol- stable environment, but a high-capacity lowing scenario illustrates one such apparent advantage when the correlations were sub- failure in human reasoning. A person must ject to change. Our findings suggest that the choose repeatedly between red and green exploratory behavior that leads to suboptimal envelopes, one of which has a crisp dollar bill probability matching in psychological labora- tucked inside. In a particular sequence of tories may be the right strategy to use in the 100 trials, say, 70 % of the green envelopes world, where many sequences of events hold a dollar, while in the remainder the dol- follow specific patterns that are well worth lar is in the red envelope. Rational choice trying to figure out.

38 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 39

Ecological Rationality

Fast and frugal heuristics can not only perform as well as more complex algorithms, they can also perform better. Even if humans and other animals had the computational resources to use such complex algorithms, following them might be inefficient and result in poor performance. The surprisingly high performance of heuristics results from their ecological rationality. A heuristic is ecologically rational when it performs well by exploiting the statistical structure of the environment. We study ecological rationality from several perspectives, three of which we will focus on in this section. First, we use computer simulation to test fast and frugal heuristics in different environments. Second, we derive analytic results which specify the conditions under which a simple heuristic can match or outperform more complex strategies. Finally, we explore whether people use heuristics, and how they adapt to different environments by selecting different heuristics from the adaptive toolbox.

Simple Heuristics for Inductive Inference connectionist models, exemplar models, and Key Reference Humans and other animals need to make decision tree models. Brighton, H. J. (2006). inductive inferences. This is the task of mak- How well does Take The Best compare to the Robust inference with more resource intensive connectionist, exem- simple cognitive models. ing an informed guess about the future using In C. Lebiere & B. Wray observations of the past. An influential trend plar, and rule-based methods? In particular, (Eds.) Between a rock and in contemporary psychology is to examine by performing less processing, will perform- a hard place: Cognitive ance suffer? For a number of environmental science principles meet human performance from the perspective of AI-hard problems. Papers rational principles of induction, such as settings, we performed a model comparison from the AAAI Spring Bayesian statistics. This poses a puzzle. to assess the predictive ability of Take The Symposium (AAAI Tech. Best and several nonlinear models. In a learn- Rep. No. SS–06–03, Although humans appear to be extremely pp. 17–22). Menlo Park, effective at making inductive inferences, and ing phase, each model is first presented with CA: AAAI Press. sometimes act as if they followed rational partial experience of the environment. We principles, decades of research into machine then examine how well each model predicts learning and pattern recognition have shown future events in the same environment. that classically rational induction becomes Examples of environments include the popu- intractable, from a mechanistic perspective, lation of German cities where the task is to for anything but trivial problems (Brighton & infer which of two novel cities has a greater Todd, 2006). How can humans perform so population, or the salaries of professors, effectively despite constraints on speed, where the task is to infer which of two computational resources, and the availability professors earns more. Brighton (2006) of information? demonstrated that Take The Best, even We examine how simple cognitive mecha- though it performs less processing, can out- nisms adapted to natural contexts can shed perform five well-known connectionist, light on this puzzle. Previously, we have exemplar, and rule-based models of inductive shown that the simple heuristic Take The inference. Best often makes more accurate predictions For three environments, the impressive per- than sophisticated linear regression models formance of Take The Best is shown in (Czerlinski et al., 1999). Many found these Figure 3, which plots the predictive accuracy results surprising, but others proposed of Take The Best and three competing models putting our models to a far tougher test. as a function of degree of exposure to the Addressing this challenge, we compared Take environment. Take The Best outperforms the The Best with several resource intensive and other models by a significant margin. These sophisticated nonlinear models. These results strengthen our previous findings models are commonplace in research into which suggested that inductive inference artificial intelligence as well as cognitive does not demand intensive and domain- modeling. Figure 2 outlines three classic general processing methods in order to approaches to nonlinear processing: achieve high performance. Simple heuristics

Center for Adaptive Behavior and Cognition | 39 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 40

(A) Exemplar Models

Z2 X1 X2 Exemplar models use the metaphor of the memory system, and simply store past observations. Z1 These observations are represented as points in a multidimensional feature space. When a novel problem is encountered, the most relevant past observations are retrieved and used to make a prediction.

Y Z X

(B) Neural Network Models

Outputs Neural network models use the metaphor of the brain to inform the processing problems by implementing networks of artificial neurons. During learning, the strength of the connections between neurons is adjusted to fit past experience. Neural activations resulting from novel problems are then used to derive a prediction.

Hidden layer

Age Decision tree models construct hierarchies of if-then rules to describe and categorize past experiences. These rules define the patterns that are used to make inferences about < 18 18–35 36–55 > 55 novel problems. 2 will buy/ 0 will buy/ 5 will buy/ 5 will buy/ 1 won’t buy 4 won’t buy 3 won’t buy 0 won’t buy

Income Won’t buy Marital Will buy status

High Low Married Single 0 will buy/ 2 will buy/ 0 will buy/ 5 will buy/ 1 won’t buy 0 won’t buy 3 won’t buy 0 won’t buy

Won’t buy Will buy Won’t buy Will buy

Figure 2. Learning algorithms simulate the task of making inductive inferences. They process sequences of observations in an attempt to uncover predictive patterns. Once found, these patterns are then used to predict future events. We evaluated the simple heuristic Take The Best by examining how well it predicts future events in comparison to a number of resource-intensive processing models drawn from three well- known paradigms: (A) exemplar/nearest neighbor models, which store past experiences verbatim, and then retrieve relevant ones them when a confronted with a new problem; (B) feed-forward neural network models, which encode past experiences by learning activation strengths between simulated neurons that map inputs to outputs; (C) decision tree induction models which construct hierarchies of IF-THEN rules to categorize past observations.

40 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 41

Figure 3. These plots City population Mammals lifespan show how the predictive 75 75 accuracy of the simple heuristic Take The Best compares to three stan- 70 70 dard machine learning algorithms, in three environments. The rival 65 65 algorithms include a decision tree induction model (CART), an exem- 60 60 plar model (Exemplar), and a neural network Predictive accuracy Predictive accuracy model (BackProp). As a 55 55 result of performing less processing, Take The Best can often significantly 50 50 outperform these stan- 0102030405060708090 0102030 40 5060 dard models. Sample size Sample size

Professor salary 85

75 Take The Best 70 BackProp CART 65 Exemplar

Predictive accuracy 60

50 010155202530 35 30 45 50 55 Sample size

selected from the mind’s adaptive toolbox, criterion is known. This independence as- Key Reference which fit the problem at hand, do just as well sumption has been made by many authors Katsikopoulos, K. V., & or better. This is a clear and striking example when examining Bayesian methods for in- Martignon, L. (2006). Naive heuristics for of how simple processes, by being ecologi- ductive inference (Domingos & Pazzani, paired comparisons: cally rational, can lead to adaptive behavior. 1997; Krauss & Martignon, 2003). Some results on their Katsikopoulos and Martignon (2006) have relative accuracy. Journal of Mathematical Psy- When Do Simple Heuristics Perform As studied under which conditions simple chology, 50, 488–494. Well As a Naive Bayesian? heuristics perform as well as a Naive Simple heuristics can perform as well or bet- Bayesian model, which assumes independent ter than more resource-intensive models cues. In particular, they focus on Take The when observations are limited, but can sim- Best, which searches through cues in order of ple heuristics also compete with these mod- their validity, and makes a decision on the els when the decision maker has had com- basis of a single cue. They also examined plete exposure to the environment? A deci- Tally, which sums up the positive evidence for sion maker who ranks two objects on a each object and selects the object with the criterion (e. g., price) by using several proba- largest sum. Katsikopoulos and Martignon bilistic cues is naive if he or she assumes that (2006) showed that Tally is as accurate as cues are independent when the value of the Naive Bayes when all cues have the same

Center for Adaptive Behavior and Cognition | 41 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 42

Key Reference validity. Furthermore, they show that Take the adaptive toolbox. The idea that people Rieskamp, J., & Otto, P. The Best performs as well as Naive Bayes are equipped with a repertoire of cognitive E. (2006). SSL: A theory when the odds ratio of the cues validities fol- strategies can be found in many areas of psy- of how people learn to select strategies. Journal low a specific structure. The odds ratio is the chology, and this makes the strategy selec- of Experimental Psychol- validity of a cue divided by 1 minus the valid- tion problem a pressing issue for many re- ogy: General, 135, ity. If this ratio for each cue is larger than the searchers. The traditional view follows a 207–236. product of the odds ratio for all other cues cost-benefit approach, where individuals are with lower validities, the odds condition, then viewed as trading off the cost of a strategy Take The Best performs as well as Naive against its benefits when performing strategy Bayes. selection. The cost of a strategy is related to These results extend our previous analytic the cognitive effort required to execute it, work, studying under which conditions Take and the benefits are related to its accuracy. The Best is more or equally accurate as Tally, According to this view, people anticipate and the converse (Martignon & Hoffrage, both the benefits and costs associated with 1999, 2002). Among 20 natural environments different strategies, and choose the most cost tested, Katsikopoulos and Martignon (2006) effective strategy for the problem at hand. found that the odds condition held in three This approach has been criticized as not be- of them, and speculated that the human ing sufficiently well-specified, making it nec- mind could be wired to detect the presence essary to advance the theoretical approach of this condition, which, in turn, might trig- by providing a computational model that de- ger the use of Take The Best. In previous scribes the strategy selection process more work, Krauss and Martignon (2003) found precisely. that the proportion of participants that used Rieskamp and Otto (2006) have argued that Take The Best jumped from 20 % to 75 % in people do not apply a metastrategy which these cases. This work provides a valuable in- trades off costs and benefits. If they did, how sight into what features of the environment would they select the metastrategy? To avoid might trigger the selection of one strategy the problem of infinite regress, Rieskamp and over another. Otto (2006) suggested that people select strategies through learning. They aimed to Selecting Strategies From the Adaptive shed light on three fundamental questions: Toolbox First, do people select different strategies in The adaptive toolbox is the repertoire of spe- different environments? Second, do people cialized cognitive mechanisms, which in- learn to select the strategy of the adaptive cludes fast and frugal heuristics. One open toolbox that performs best in a particular question is how heuristics are selected from environment? Finally, how can the learning

Figure 4. Rieskamp and Otto (2006) consider the How do people select strategies? Strategy Selection Theory (SSL) problem of how the cognitive system selects between the heuristics in Select strategy proportional to the adaptive toolbox. initial expectancies They propose that these heuristics are undergoing reinforcement. Rein- Apply strategy forcement occurs as a result of feedback reflecting how well each Use strategy’s success or failure heuristic performs in the to update strategy‘s expectancy given context.

”… the question is when to Select strategy proportional to use which tool!“ updated expectancies Duncan Luce, 2000

42 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 43

process underlying strategy selection best be of reinforcement learning (Figure 4). Accord- described? Rieskamp and Otto (2006) focused ing to SSL, people use unobservable cognitive on the problem of choosing which of two strategies, rather than stimulus response as- alternatives, described by several probabilistic sociations, and these cognitive strategies are cues, has a higher criterion value. For in- reinforced though feedback. The “expectan- stance, which of two companies described by cies” of the strategies are updated through several cues is most creditworthy? When reinforcement learning, which relies on feed- making such an inference, a number of infer- back on past performance. SSL predicts that ence strategies could be applied, for example, the strategy that performs best will be se- Take The Best, Tally, or a Weighted Additive lected, given sufficient learning opportunity. strategy. Recent studies have shown that Rieskamp and Otto (2006) tested SSL against noncompensatory strategies like Take The four alternative models, including three gen- Best perform well at predicting inferences eral learning models and one exemplar when participants are under time pressure, model. They performed a number of experi- when there are high costs in information mental studies in which participants made acquisition, or when information about cues repeated inferences in different environ- has to be retrieved from memory. In contrast, ments, and were given feedback on their per- compensatory strategies like the Weighted formance. In one study, participants had to Additive strategy were more suitable for select which of two unnamed companies was making inferences in situations of low time more creditworthy. This experiment consid- pressure, low information acquisition costs, ered two environment conditions. In the first or when the information was provided simul- condition, featuring a noncompensatory en- taneously via a computer screen. vironment, Take The Best yielded the highest Rieskamp and Otto (2006) proposed a new performance. In the second condition, featur- Strategy Selection Learning (SSL) theory, ing a compensatory environment, a Weighted which views strategy selection as a process Additive strategy achieved the highest per-

Adapting to the environment Theory comparison 0.9 1.00

0.8

0.7 0.75

0.6

0.5 0.50 TTB’s fit—friendly environment 0.4 TTB’s fit—unfriendly environment SSL theory 0.3 Porportion of participants Porportion 0.25 0.2

0.1 Fit of TTB/SSL’s predicted probability of selecting TTB Fit of TTB/SSL’s

0.0 0.00 1234567 SSL theory Exemplar model Trial blocks

Figure 5. The proportion of choices predicted by the simple heuristic Take The Best increases over time in the Take The Best-friendly environment (left). This adaptation process is predicted by the SSL Theory. The SSL Theory also predicts the behavior of the majority of participants better than a competing exemplar model (right).

Center for Adaptive Behavior and Cognition | 43 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 44

Key References formance. This study demonstrated that being carried out in behavioral ecology. The Rieskamp, J. (2006a) when people repeatedly make probabilistic hope was that both disciplines could learn Perspectives of proba- inferences, their performance improves. Fig- from each other. bilistic inference: Rein- forcement learning and ure 5 shows that the participants initially had One advantage that biologists have over an adaptive network a preference to integrate the provided infor- psychologists is that animals can be studied compared. Journal of Ex- mation, which is not surprising, given that all in the natural environments to which they perimental Psychology: Learning, Memory, and information was provided and information have adapted, although in both disciplines it Cognition, 32, 1355– search was not required. This initial prefer- is often maladaptive behavior in atypical 1370. ence changed very quickly depending on the environments that spurs research. Interest- Hutchinson, J. M. C., & structure of the environment. When the envi- ingly, the range of animal behavior described Gigerenzer, G. (2005b). ronment structure favored the integration of using rules of thumb is far broader than the Simple heuristics and rules of thumb: Where the available information, people selected the range of human behavior explored by ABC. psychologists and behav- Weighted Additive strategy more frequently. The rules of thumb used by animals have ioural biologists might However, when the structure of the environ- been discovered largely through careful em- meet. Behavioural Processes, 69, 97–124. ment favored the selection of Take The Best, pirical observation and experiments, and people selected Take The Best more fre- there are real prospects for understanding quently. the neurological and physiological processes At the end of the experiment, the strategy underlying them. Mirroring ABC’s approach, that performed best in the environment was there are also simulation studies in biology also the strategy that predicted participants’ that assess the performance of rules of inferences most accurately. This learning thumb. process was also accurately described by the Can biologists learn from ABC’s work? The SSL theory (see Figure 5). Interestingly, the idea that simple rules of thumb can outper- more complex learning models and the ex- form more complex processes has not been emplar model failed to describe the learning thoroughly explored, or widely appreciated, process more accurately. This finding was in biology. From ABC’s perspective, a rule of consistent across several inference situations, thumb is likely to exist as an adaptive re- including a situation involving costly infor- sponse to the ecological structure of the task mation search, and a situation where a environment, rather than existing merely as a choice between three rather than two alter- result of a simple brain. Other aspects lacking natives was required. Furthermore, Rieskamp in biology are (a) a theoretical perspective (2006a) compared the SSL theory with a capable of explaining the enormous diversity well-established connectionist model, and of methods of cue integration biologists have found further evidence for the adaptive se- documented and (b) the development of pre- lection of strategies from an adaptive tool- cise models specifying how multiple cues are box. processed. Although ABC also has much progress to make in explaining why the sta- Do Animals Use Simple Heuristics? tistical structure of environments favors par- For biologists, the concept of ecological ra- ticular methods of cue integration, biologists tionality is uncontroversial. The assumption could benefit from our attempts to model that organisms are often well-adapted to and understand this aspect of cognition. Ulti- their environment is ubiquitous in biology. mately, the conceptual vocabulary needed to Consequently, biologists see no reason why address this fundamental problem is likely to this assumption should not be extended to be the same for both disciplines. the study of cognition. Furthermore, there is a long tradition in biology of explaining adaptive behavior using rules of thumb, which are similar in spirit to the simple heuristics studied by ABC. Hutchinson and Gigerenzer (2005b) explored the similarities and differences between ABC’s work and that

44 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 45

Social Rationality

Social rationality is a specific form of ecological rationality, one capturing the fact that social species need to make decisions in an environment which may be constructed by the actions of others. In studying social rationality, we attempt to understand the cues and heuristics that underlie cooperation and group decision making, and to uncover the potential role of emotional states in the functioning of social heuristics. These social heuristics could represent adaptive solutions to recurring social problems faced by humans during their phylogenetic and ontogenetic development.

Cross-Cultural and Developmental victimizer” tradition, where some of the rule Key Reference Perspectives on Decision Making violations presented to children may have Keller, M., Brandt, A., & In a longitudinal project, we examined the been unfamiliar. The dilemma approach Sigurdardottir, G. (in allows us to study both moral emotions of press). Happy victimizers development of sociomoral reasoning, and, in and unhappy moralists: particular, we consider reasoning about close guilt due to violating an obligation, or posi- The development of relationships and moral norms in both China tive moral feelings of pride due to acting in moral emotions in a accordance with obligations. cross-cultural perspec- and various Western countries. Children and tive. In W. Koops & A. adolescents of different age groups, and in Four age groups (7-, 9-, 12-, and 15-year- Sanders (Eds.), The devel- different cultures, focus on specific defining olds) of Icelandic and Chinese participants opment and structure of were studied. The results revealed that the conscience. Hove, UK: properties (or cues) of relationships which Psychology Press. can lead to different behavioral responses “happy victimizer” phenomenon was very when making decisions and moral evalua- infrequent, even in the 7-year-olds. Few tions. Here, culture determines the specific (mostly Icelandic) children did not under- meaning of obligations and responsibilities in stand the moral aspect of the situation at close relationships as well as the conception all. Nearly all younger Icelandic children of self (Keller, 2006). Some of these differ- were “unhappy violators,” feeling guilty ences can be explained by the well-known when they gave precedence to self-interest distinction of (Western) individualism and over friendship responsibilities. On the other (Asian) collectivism. However, our research hand, there were some younger children indicates that cultural transformations, like (again mostly Icelandic) who were “unhappy the rapid process of modernization in China, moralists,” that is, they decided to keep the can influence decision making (Keller, 2006; promise to the friend, but felt bad because Keller & Krettenauer, 2007). they missed out on a good opportunity. In general, the Icelandic participants interpret- Moral Emotions ed the dilemma as a conflict between friend- Moral emotions, such as guilt and shame, ship obligation and self-interest. In contrast, which are associated with the consequences for Chinese children, self-interest rarely of moral transgressions, are important cues played a role in their decision making and for the motivational acceptance of moral emotions. Independent of which decision norms. In our previous research, we had they made, either for a friend or a newcomer, demonstrated that the “happy victimizer” they would always feel bad about leaving phenomenon in young children (attributing out one person. Thus, they can be seen as positive feelings to a moral-rule violator in “unhappy moralists” who are stuck in a spite of moral knowledge that the violation moral dilemma of conflicting obligations of is not right) was, in part, a result of the friendship and altruism across all four age experimental procedure failing to distinguish groups. This research demonstrates that between other and self. Recently, the topic decision making in people of different ages of moral emotions was followed up in a and different cultures will depend on differ- cross-cultural context (Keller, Brandt, & ent cues and emotions. Sigurdardottir, in press). This study avoids the problem of previous studies in the “happy

Center for Adaptive Behavior and Cognition | 45 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 46

Cognition and Emotion in Cooperative regarded as prosocial. In contrast, in the ulti- Decision Making matum game, the responder group can reject In a series of studies, we explored children’s the offer of the proposer. In this case, no and adolescents’ sharing of resources in the player receives any money. If the responder context of bargaining games (Gummerum, accepts the offer, however, the money is dis- 2005; Gummerum, Keller, Takezawa, & Mata, tributed according to the suggestion of the in press; Keller & Canz, 2007; Takezawa, proposer. Hence, a positive offer of the pro- Gummerum, & Keller, 2006). These studies poser can be motivated both by strategic and relate developmental, social, and cognitive prosocial considerations. Three same sex, psychological perspectives to research in same age members of the groups had to first economic game theory and evolutionary make an individual decision, and then nego- anthropology and biology. This integration tiate a common decision within ten minutes. leads to a comprehensive, multidisciplinary, In general, individual offers in the dictator and multimethod perspective on the devel- game were more prosocial than those of opment of cooperation. Economic game the- adults in previous studies, and the modal ory allows us to conduct an experimental offer in all age groups was the equal split. investigation into unresolved questions in Individual offers demonstrated no age effect moral development theory, such as the con- in either game, and altruistic preferences nection between moral reasoning and action, predicted individual offers in both the dicta- ontogenetic changes in the manifestation of tor and ultimatum games. Furthermore, selfish and moral behavior, and the ability to individual offers were, on average, more negotiate a common decision. prosocial than group offers. Analyses of the Previously, economic studies show that group decision-making process showed dif- adults do not only act according to the prin- ferences in the process used at the level of ciple of utility maximization, but reveal fair- the group, which might either be explained ness and prosocial preferences. For the dicta- by a majority or averaging model or the tor and ultimatum game, we studied the influence of prosocial and selfish group problem of sharing in groups of children and members during the group discussion in the adolescents of ages 9, 12, 14, and 17 years dictator game. (Figure 6). In both games, one group (pro- A qualitative analysis of the video-taped poser) decides whether, and how, to share a group discussions in the dictator game sum of money (20 coins of different value) revealed different types of argument. These with another anonymous group (responder). arguments can be seen as based on social In the dictator game, the proposer group heuristics that serve to increase or decrease unilaterally decides how much money each group offers, such as reciprocity, hierarchy, player receives. Therefore, from a game theo- and group membership. In particular, the retical point of view, each distribution that negative use of these arguments tends to allocates money to the responder can be diminish the offers made to the other group. Thus, the argument that the other group Figure 6. To share or not would also give little (negative tit-for-tat) to share? Three 9-year- could convince others to make a more selfish olds discussing how to share a sum of money offer. The youngest age group usually pro- with another (anony- posed an offer and stated factual (nonnor- mous) group of children. mative) preferences, the two middle groups deal with facts and norms while the oldest age group initiates discussion about norm conflicts and the justification of norms.

46 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 47

Information Processing and Decision Figure 7. Five adults Making in Groups arriving at a group decision. How do groups of people make joint decisions? Using group experiments, computer simulations, and mathematical analyses, ABC examines efficient and effective decision strategies that can be used by committees and groups who have to come to a joint decision (Reimer & Hoffrage, 2005, 2006, in press; Reimer, Hoffrage, & Katsikopoulos, 2007; Reimer & Katsikopoulos, 2004; Reimer, Kuendig, Hoffrage, Park, & Hinsz, 2007). The predominant view is that a good group decision requires intensive information exchange, and a pooling of resources. But one frequent The Information Processing Cube Key References complaint by group members is that such a We distinguish between combination- and Reimer, T., Kuendig, S., strategy leads to excessive amounts of time communication-based heuristics (Reimer & Hoffrage, U., Park, E., & being spent when reaching group decisions. Hoffrage, 2005, 2006). These two classes of Hinsz, V. B. (2007). Ef- fects of the information Furthermore, ABC’s findings on decision group decision strategies can be framed in environment on group making in individuals, where less information terms of the information processing cube discussions and decisions processing can be shown to be beneficial, displayed in Figure 8 (Adamowicz, in the hidden-profile paradigm. Communica- suggests that extending heuristics used in Hanemann, Swait, Johnson, Layton, tion Monographs, 74, individual decision making to the context of Regenwetter, Reimer, & Sorkin, 2005; Reimer 1–28. groups and teams may be fruitful. In particu- & Hoffrage, in press). The information- Reimer, T., & Hoffrage, lar, we examine the ecological rationality of processing cube has three dimensions: the U. (2006). The ecological rationality of simple these heuristics when used in the context members of the group, the choice alter- group heuristics: Effects of a group, and test their ability to describe natives, and the cues used to describe the of group member strate- group behavior. alternatives. A cell in this cube refers to the gies on decision accuracy. Theory and Deci- knowledge a group member has about a sion, 60, 403–438. certain choice alternative on a certain cue.

Figure 8. The information processing cube high- V11k … … … V1jk lights three dimensions that frame the group decision making problem: V V 21k … … … 2jk Decision strategy the members of the group, the choice alter- … natives, and the cues A V V … … … V used to describe these 1 111 121 … 1j1 alternatives. A cell in the … Group A V … … V cube represents an indi- 2 211 2j1 decision vidual’s knowledge about … a cue for one of the alternatives. … Vi1k Vi2k … … … Vijk Alternatives

… k … … …

… … M Ai Vi11 Vi21 … … … Vij1 2 M M 1 Group members C C C 1 2 … … … j Cues

Center for Adaptive Behavior and Cognition | 47 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 48

Key Reference In the literature, two types of aggregation that integrates individual preferences on the Reimer, T., Park, E., & mechanisms have been distinguished. Social basis of a majority rule may be strongly Hinsz, V. B. (2006). combination rules, like majority and plurality affected by the decision strategies used by Shared and coordinated rules, assume that each group member its individual members. Furthermore, and in cognition in competitive and dynamic task envi- aggregates across the cues and alternatives, line with the notion of ecological rationality, ronments: An informa- first by forming an individual decision, and, the accuracy of the decision strategies tion-processing perspec- in a second step, the group aggregates depended on the structure of the environ- tive for team sports. In- ternational Journal of across the preferences or opinions of the ment. Sport and Exercise Psy- individual group members. In contrast, social chology, 4, 376–400. communication rules are those where the Social Communication Rules: The Hidden members of a group pool their knowledge on Profile Task the decision alternatives. For example, group The literature on the Condorcet theorem members may infer in a discussion which shows that the majority rule tends to accen- candidate is best suited for the job vacancy tuate differences between individuals. by forming an impression on each candidate Specifically, the majority rule will enhance and choosing the candidate with the best accuracy when members tend to be correct; overall evaluation, or a group may compare however, when a group consists of members the alternatives cue-wise by reaching a con- that tend to favor a wrong decision, the sensus on which the most important cues majority rule will yield even worse decisions are, and by choosing the alternative that than those arrived at by averaging the deci- scores highest on the most important cues sions of individual group members (see (Reimer & Hoffrage, 2005). Reimer, Bornstein, & Opwis, 2005; Reimer, Park, & Hinsz, 2006). One situation in which Social Combination Rules: The Majority Rule the majority rule systematically fails because One of the most popular decision rules, group members tend to fail are hidden pro- especially when groups fail to reach a con- file tasks (Reimer & Hoffrage, 2005; Reimer sensus, is the majority rule. The majority rule et al., in press). Consider the following exam- requires that group members have formed ple of a hidden profile task: individual opinions which are then (either implicitly or explicitly) integrated by the group. Very few studies have examined how Two candidates, A and B, apply for a posi- the strategies used by individual group mem- tion, and a four-member committee has bers influence the outcome of the majority to select one of them. Overall, most argu- rule. In a simulation study, we tested the ments are in favor of candidate A. How- sensitivity of the majority rule to four differ- ever, no single group member is aware of ent decision strategies (Reimer & Hoffrage, this because information is distributed 2006). We considered two compensatory among the committee members in a bi- decision strategies, and two noncompen- ased way, such that each group member satory heuristics. In addition, to assess the has more arguments in favor of candidate ecological rationality of the strategies, we B. As a consequence, when groups inte- varied the distribution of cue validities, the grate their members’ individual opinions quantity, and the validity of shared informa- on the basis of a majority rule, they will tion. Group performance strongly depended not detect the hidden profile. Experimen- on the distribution of cue validities. When tal studies indicate that groups rarely de- validities were linearly distributed, groups tect hidden profiles. using a compensatory strategy achieved the highest accuracy. Conversely, when cue validities followed a J-shaped distribution, The prominent explanation for the hidden groups using a simple lexicographic heuristic profile effect is that groups fail to pool and achieved the highest accuracy. These studies integrate all available pieces of information. suggest that the performance of a group But the question of how the information

48 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 49

should be processed by the group has rarely In contrast to unique cues, common cues Key Reference been considered. In several of our simulation provide information on each and every Gigerenzer, G. (2007b). studies, a group version of Take The Best option of the choice set. As expected, groups Gut feelings: The intelligence of the uncon- performed very effectively at identifying in the unique cue condition chose the hidden scious. New York: Viking concealed alternatives in the hidden profile profile alternative less often than groups in Press. task, thereby demonstrating that the detec- the common cue condition (Reimer et al., in UK edition: Short cuts: tion of a hidden profile does not necessarily press). However, it turned out that it is not The intelligence of the require exhaustive information processing sufficient to provide groups with environ- unconscious. London: Penguin Books, 2007c. (Reimer & Hoffrage, 2005). This strategy ments that include common cues for groups German translation: requires that groups apply a cue-based strat- to detect hidden profiles. Common cue envi- Bauchentscheidungen: egy by identifying cues that discriminate ronments facilitated hidden profile detec- Die Intelligenz des Unbe- among alternatives (e. g., a search committee tions, but even in these environments, most wussten und die Macht der Intuition. München: might compare applicants with regard to groups failed to detect the hidden profile. Bertelsmann, 2007a. their publication record, and choose the can- This seems to be due to the fact that group Spanish translation: didate with the most publications). members entered group discussions with Arial. Previously, studies on the hidden profile preconceived opinions. A follow-up study Dutch translation: effect typically assume that all choice alter- revealed that all groups detected hidden Kosmos. natives are described on the basis of unique profiles when group members entered group Korean translation: cues describing only one of the choice alter- discussions without a preconceived pre- Chugrim. natives. Because this kind of environment ference. Italian translation: Raffaelo Cortina. structure may hinder groups from applying a cue-based strategy, we conducted a study which included common cues.

In his recent book Gut Feelings, Gerd Gigerenzer explains the science behind Malcolm Gladwell’s best- selling Blink. By examining various decisions we make—how we choose a spouse, a stock, a medical procedure, or the answer to a million-dollar game show question—Gigerenzer shows how gut feelings not only lead to good practical decision but also underlie the moral choices that make our society function.

Center for Adaptive Behavior and Cognition | 49 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 50

Evolutionary Psychology

Evolutionary psychology lies at the heart of many research projects undertaken by the ABC Research Group. It motivates the notion of bounded rationality, supports the significance of the environment in the notion of ecological rationality, and emphasizes the importance of social interactions in social rationality. At the same time, evolutionary psychology is grounded in ecological rationality: It assumes that our minds were designed by natural selection to solve practical problems in an efficient and effective manner. Recently, Todd, Hertwig, and Hoffrage (2005) argued in an influential handbook of evolutionary psychology, a set of broad forces operating on multiple domains that impact on the design of specific cognitive systems. They discussed how the costs of gathering information, and of using too much information, can be reduced by decision mechanisms that rely on as little information as possible to come to their choices. They also explored how the pressures to use small amounts of appropriate information may have produced particular patterns of forgetting in long-term memory and particular limits of capacity in short-term memory. ABC’s interest and research into evolutionary psychology continues, and, in this section, we focus on empirical studies of mate selection, and simulation studies focusing on the evolution of cooperation.

Key References Mate Choice danger to a high-quality male than to a low- Wilke, A., Hutchinson, One of the most evolutionarily important de- quality male, high-quality males can afford J. M. C., Todd, P. M., & cisions is mate choice. By definition, sexual to take such risks more often, and thus rivals Kruger, D. J. (2006). Is risk taking used as a cue reproduction requires combining your own and potential mates should use risk taking as in mate choice? Evolu- genes with another individual’s genes in or- a cue to quality. tionary Psychology, 4, der to produce offspring. In mate choice, de- Wilke, Hutchinson, Todd, and Kruger (2006) 367–393. cisions made on the basis of perceived cues investigated the possible signalling functions Todd, P. M., Billari, F. C., can influence the quality of the genes passed of risk taking by measuring how attractive & Simão, J. (2005). Ag- gregate age-at-marriage on to their offspring, and the quality of the each sex finds risk taking in the opposite sex. patterns from individual parental care their offspring will receive. Pre- If risk taking is used as a cue for mate choice mate-search heuristics. vious work has emphasized the importance of in humans, evolutionary theory would predict Demography, 42, 559– 574. mechanisms of mate choice by examining the that different types of risk provide different construction of building blocks for mate information about the condition of a poten- choice strategies and the resulting popula- tial partner. For instance, risk taking in sexual tion level pattern of mate choice behavior, behavior may signal high probability for sex- such as age at marriage (Todd, Billari, & ually transmitted disease, whereas participat- Simão, 2005). Recent work has explored posing in a risky or dangerous sport may signal sible cues used in mate choice as well as the physical prowess. Wilke et al. (2006) surveyed primary evolutionary outcome of mate how attractive both men and women found choice: offspring production. risky behaviors in the opposite sex across a For humans, the number of females in the number of domains. Surprisingly, the sexes population limits the ability of males to pro- differed very little in what risky behaviors duce offspring; therefore, males often experi- they found more attractive or unattractive, ence more variance in their reproductive suc- but these ratings differed across domains cess (some males will produce a number of (see Figure 9). In particular, both men and offspring, while others will produce none). women reported social and recreational risk This variance in male mating success may fa- taking as attractive, but other domains of risk vor risk taking because high potential gains taking as unattractive (ethical, gambling, and (e. g., in resources promoting partner acquisi- health) or neutral (investment). Thus, risk tion) outweigh the high risks. Additionally, taking does not act as a general cue for mate males may take risks as a form of advertise- choice; instead, the risk environment ment of their quality to both females and determines the value of information in mate rival males. If risky behaviors are less of a choice.

50 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 51

Figure 9. Risk taking is 5 Attractive German not uniformly evaluated as attractive, but depends on the domain in question. This figure shows mean domain 4 scores (and standard deviations across items) for ratings by women of the attractiveness of male 3 risk taking (blue) and by men of the attractiveness of female risk taking (white).ˇ A rating of 2 5 refers to attractive, and a rating of 1 refers to unattractive.

1 Unattractive

5 Attractive U.S.

1 Unattractive Recreation Ethics Gambling Investment Health Social

These results emphasize the necessity of tak- ety (the Shuar community of Ecuador) face Key Reference ing a domain-specific approach to studying this quantity-quality tradeoff by measuring Hagen, E. H., Barrett, H. the functions of risk taking (see also Hanoch, proxies of quantity and quality of children in C., & Price, M. E. (2006). Johnson, & Wilke, 2006). Although we still do the Shuar families. As predicted, children in Do human parents face a quantity-quality trade- not know if human mechanisms for decision families with more children per adult had off? Evidence from a making concerning risk evolved for particular lower weights, shorter heights, thinner skin- Shuar community. Amer- domains (e. g., foraging) separately or for the fold thicknesses, and narrower arm and leg ican Journal of Physical Anthropology, 130, 405– more general problem of choice under uncer- circumferences. This suggests a real cost to 418. tainty, posing these questions leads to new increasing family size: More mouths to feed testable predictions and a more differenti- results in less food per mouth. ated understanding of risk taking. Why invest so much effort into choosing an appropriate Cooperation mate? Evolutionary theory suggests that bet- Cooperation poses an interesting puzzle to ter mates yield better offspring and higher evolutionary theory. Why would an individual probability of passing on genes. But given help another? Many mechanisms of coopera- limited resources, is one constrained to invest tion have been proposed, but most of this heavily in few offspring or sparsely in many work has been dominated by a focus on sym- offspring? Hagen, Barrett, and Price (2006) metric games (typically, the prisoner’s asked whether parents in a small-scale soci- dilemma game) in which both individuals

Center for Adaptive Behavior and Cognition | 51 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 52

Figure 10. The investment game. The payoff of player A is defined as Player A Player B his endowment minus his investment plus player B’s return. Player B’s payoff is defined as her endowment plus the trebled investment of player A minus player B’s return. Investment: Increase: Return: x % of 10 3 • 10 • x y % of 3 • 10 • x

Payoff A =10 –10 • x+y• 3 • 10 • x Payoff B =10 + 3 • 10 • x–y• 3 • 10 • x

Key Reference face the same strategic situation. However, receives a greater payoff when both players Rieskamp, J., & Todd, P. social relationships are often imbalanced. cooperate than when they both defect. M. (2006). The evolution Consider, for instance, an employee that The difference in symmetry between these of cooperative strategies for asymmetric social might have frequently worked overtime two games yielded the evolution of different interactions. Theory and trusting that his effort will be reciprocated, strategies. The asymmetrical investment Decision, 60, 69–111. but the employer may, or may not, ultimately game resulted in the unforgiving minimum- reward this extra effort. Whereas the two in- grim and minimum-return strategies in dividuals in a symmetric interaction face the which the partner’s investments and returns same decision alternatives, in an asymmetric must exceed a minimum threshold. If they interaction the decision alternatives differ for do not meet the threshold, the player never the two parties. Do these commonplace invests again. In contrast, the prisoner’s asymmetric interactions support or inhibit dilemma game resulted in a more forgiving the emergence of cooperation? cautious tit-for-tat strategy, which initially Rieskamp and Todd (2006) explored whether defects, but then copies its opponent’s previ- decision strategies exist that can reliably lead ous behavior. Unlike the grim strategies, it to ongoing cooperation in an asymmetric in- will forgive a partner’s defection if the part- teraction, and determined whether such ner responds with cooperation. Therefore, the strategies differ from those that are appro- strategies that enable cooperation for the priate for cooperation in a symmetric inter- asymmetric investment game react more action. They tested the evolution of strategies sensitively to exploitation, meaning that co- for the investment game (an asymmetric operation can more easily break down. Fur- game) and compared it with results for the thermore, once cooperation has stopped, it is prisoner’s dilemma game (a symmetric game) much more difficult to reestablish than in using evolutionary simulations. In the invest- symmetric situations. If asymmetry influ- ment game, player A can invest any amount ences decision strategies as suggested here, of an endowment, which is then tripled, be- then the predominant use of the symmetric fore it is delivered to player B. Next, player B prisoner’s dilemma game as a general model decides how much of the tripled amount he for social interactions is not justified. or she wishes to return to player A (see Fig- ure 10). Thus, the players face asymmetrical payoffs. In contrast, players in the prisoner’s dilemma game experience symmetrical payoffs by either cooperating or defecting. For both players, defecting always leads to a greater payoff regardless of whether the opponent defects or cooperates; thus, defecting is a dominant strategy. However, each player

52 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 53

Decision Making in the Wild

According to an ancient piece of wisdom, attributed to many luminaries, every good theory leads to good applications. In this section, we review the main applications of our basic research in the last 2 years.

Heuristics and the Law (emphasizing the rational actor perspective), Key Reference What turns an issue into a public issue, and and behavioral law and economics (empha- Gigerenzer, G., & Engel, then into a law? How do jurors and judges sizing the heuristics and biases perspective). C. (Eds.). (2006). Heuris- tics and the law. Report decide whether a witness is trustworthy or a The Dahlem Conference and the book are of the 94th Dahlem defendant guilty? Insofar as one believes that based on a year-long collaboration between Workshop on heuristics the reasoning of people involved in creating the Max Planck Institute for Research on and law, Berlin, June 6–11, 2004. Cambridge, and practicing law is guided solely by logic Collective Goods in Bonn and the Max Planck MA: MIT Press. and legal texts, heuristic thinking plays a Institute for Human Development in Berlin. minor role. And in fact, many lawyers would This collaboration will be extended to a joint posit “heuristics and the law” to be a non- International Max Planck Research School, issue. Similarly, most psychologists would not together with the Max Planck Institute for think of the law when they think of heuris- Economics in Jena. The Research School will tics. Upon closer inspection, however, one begin in 2007. finds courts cutting through complex cases A conference at the George Mason University by relying on rules of thumb, administrators School of Law dealt with police hunches, and making decisions based on one good reason, how the courts deal with them (Gigerenzer & and legislators responding to scandal. Should Brighton, in press). Like other experts—from this be seen as irrational and irresponsible, or chicken sexers to chess masters to soccer can simplicity, transparency, limited search, players—police officers have intuitions that aspiration levels, and the other conceptual they cannot always verbalize. However, “I think the publication of Heuristics and the Law tools in the “adaptive toolbox” actually help? American courts tend to discount the is a (rare) truly signifi- For instance, should a tax law be simple and hunches of police officers, requiring them to cant event in legal transparent in order to create trust and com- articulate specific facts that justify their de- adacemic life.” Mark Kelman, William pliance, or rather as complex as possible to cision to conduct a search, an interrogation, Nelson Cromwell Profes- satisfy many special situations and interests? or to make an arrest. At issue are civil liber- sor of Law, Stanford A recent book entitled Heuristics and the Law ties. When an officer had a hunch, stopped a University. (Gigerenzer & Engel, 2006) explores the car, found illegal drugs or guns, and reported intersection between legal and heuristic rea- exactly this, judges repeatedly rejected “mere soning for both legal norms and practice. The hunches” as insufficient cause for a search. book is based on a Dahlem Conference (at Yet insisting on after-the-fact justification which 40 scholars met for 5 days; no talks ignores that good expert judgment is gener- were given; papers were written, distributed, ally of intuitive nature. As a consequence, and commented in advance; and the four when police officers testify before a judge, group reports were written during the Con- they have learned not to use terms such as ference) that centered on four questions: hunch or gut instinct, but to produce “objec- (1) Are heuristics a problem or a solution? tive” reasons after the fact. (2) What is the role of heuristics in making In contrast to the dismissal of police law? hunches, Gigerenzer and Brighton (in press) (3) What is the role of heuristics in litiga- argued that (1) there is evidence that these tion? hunches are based on smart heuristics, and (4) How do heuristics mediate the impact of that these heuristics can actually outperform law on behavior? sophisticated reason-based judgments; The participants included scholars from law, (2) training of police officers and judges psychology, and economics, including the should be geared at improving these hunches interdisplinary fields of law and economics rather than discarding them; and (3) in order

Center for Adaptive Behavior and Cognition | 53 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 54

to avoid discrimination, the legal system depending on auxiliary assumptions that are In cooperation with the George Mason needs to survey the quality of police officers’ external to this distinction. School of Law and hunches, that is, a detective’s actual success for the second year in spotting criminals, rather than his or her Risk Communication in Medicine in a row, Gerd Gigerenzer taught ability to provide verbal after-the-fact justi- The term “risk” appears in the title of more decision making and fications (see also Gigerenzer, 2007). than 10,000 medical articles published every risk communication The fact that people often violate rationality year (2 % of the total). It should be alarming to American judges at the Institute on axioms has provided a new rationale for that, at the same time, most physicians and the Logic and Limits paternalism to “de-bias” individuals who patients do not understand the crucial statis- of Contract Law for exhibit errors, cognitive illusions, or other tical numbers which with medical research Judges, Tucson, AZ. allegedly pathological psychological irregular- provides them. Collective innumeracy im- ities. Legal scholars, such as Cass Sunstein, pedes the efficacy of evidence-based medi- and behavioral economists, such as Richard cine and the ideal of shared decision making. Key References Thaler, have used psychological experiments From our studies on HIV and cancer screen- Gigerenzer, G., & as the fundament for their concepts of anti- ing, summarized in Gigerenzer (2002), we Brighton, H. J. (in anti-paternalism and libertarian paternalism estimate that some 80 % to 90 % of doctors press). Can hunches be rational? In C. S. Lerner (e. g., Sunstein & Thaler, 2003). The argument are innumerate, that is, do not understand & D. Polsby (Eds.), Mere that bounded rationality implies paternalism the outcomes of standard tests, and are con- hunches: Policing in the has direct implications for the degree to fused about the meaning of basic concepts, age of terror. which individual choice (risk taking, sex, such as sensitivity and false positive rate. To Berg, N., & Gigerenzer, drugs, and retirement savings) is regulated by illustrate this problem, consider the two ex- G. (2007). Psychology implies paternalism? the government and restricted by institutions amples given in the box. Bounded rationality may in order to “protect” us against our mental If patients knew about this degree of vari- reduce the rationale to weaknesses. Berg and Gigerenzer (2007) ability and statistical innumeracy, they regulate risk-taking. So- cial Choice and Welfare, asked whether this argument is actually cor- would be justly alarmed. Physicians’ igno- 28, 337–359. rect, and define unbounded and bounded rance is enhanced by the attitude of many Engel, C., & Gigerenzer, rationality in a context of societal risk taking. patients not to ask questions, but instead to G. (2006). Law and They show analytically that the argument is rely on the rule of thumb “follow the white heuristics: An interdisciplinary venture. In G. flawed, and that in fact both utility maximiz- coat.” With the rise of evidence-based Gigerenzer & C. Engel ers (who are unboundedly rational) and medicine in the last two decades, and the (Eds.), Heuristics and the satisficers (who are boundedly rational) can decline of the traditional view of the physi- law (pp. 1–16). Cam- bridge, MA: MIT Press. exhibit excessive or normal risk taking, cian as an artist who does not bother with

The misunderstanding of risks and uncertainties by German and Swiss physicians. Example 1: German physicians with an average of 14 years of professional experience were asked to imagine using the Haemoccult test to screen for colorectal cancer. The prevalence of cancer was 0.3 %, the sensitivity of the test was 50 %, and the false positive rate was 3 %. The doctors were asked: What is the probability that someone who tests positive actually has colorectal cancer? The correct answer is about 5 %. However, the physicians’ answers ranged from 1 % to 99 %, with about half of them estimating the probability as 50 % (the sensitivity) or 47 % (the sensitivity minus the false positive rate) (Gigerenzer, 2002; Hoffrage & Gigerenzer, 1998). Example 2: After reading our publications, the head of the Department of Gynecology at the Lucerne Cantonal Hospital, Switzerland, asked his 15 gynecologists what they think the following widely publicized figure means: “Mammography screening reduces the risk of dying from breast cancer by 25 %.” They were then asked: How many fewer out of 1,000 women will die of breast cancer? The physicians’ answers varied between 1 in 1,000 and 750 in 1,000 (Schüssler, 2005).

54 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 55

numbers, this collective innumeracy is now the public or physicians, but to a large Key References recognized as a fundamental problem in degree in the external presentation of Gigerenzer, G. (2005). health care. However, decision researchers’ information (e. g., Gigerenzer & Hoffrage, Wie versteht man Risiken proposals to send physicians into statistics 1995; Gigerenzer & Edwards, 2003). und Unsicherheiten? HNO Informationen, 4, courses has been unsuccessful, given the Numerical information cannot be communi- 287–294. time pressure under which doctors are cated without a form, and some forms are Elmore, J. G., & under, and their common self-perception as easily digested by the human brain, while Gigerenzer, G. (2005a). being mathematically illiterate. In earlier others tend to produce (typically uncon- Benign breast disease— the risks of communicat- research, we have shown that the problem scious) misunderstandings. Thus, teaching ing risk. New England of innumeracy is not simply in the minds of physicians about different representations Journal of Medicine, 353, 297–299. Mata, J., Dieckmann, A., & Gigerenzer, G. (2005). Verständliche ABC’s contribution to improving the understanding of medical risks and Risikokommunikation, uncertainties leicht gemacht – oder: Wie man verwirrende Publications in scientific medical journals with high impact factors. For instance, in Wahrscheinlichkeits- the New England Journal of Medicine, Elmore and Gigerenzer (2005a) explained simple angaben vermeidet. techniques for visually representing relative risk increases for benign breast biopsies in Zeitschrift für Allgemein- medizin, 81, 537–541. transparent ways. The problem with publications in scien- Gigerenzer, G., & Publications in widely read medical journals. Edwards, A. (2003). Sim- tific medical journals is that few physicians actually read these. Thus, we also sum- ple tools for understand- marize our findings in journals that are widely distributed by national medical organi- ing risks: From innumer- zations. For instance, in 2005, we wrote about transparent risk communication in the acy to insight. British Medical Journal, 327, Zeitschrift für Allgemeinmedizin and in Hals-Nasen-Ohren Informationen. 741–744. Further education of physicians. The German government’s health system reform (GKV-Modernisierungsgesetz; effective since 2004) no longer allows the pharmaceutical industry to give presents to physicians or invite them to participate in “further training” at some nice vacation hotel, all costs paid, spouses included. Pharmaceutical companies nonetheless continue to organize further training, but are no longer allowed to promote their own products. In 2006 and 2007, Gerd Gigerenzer and other members of ABC took an active part in physicians’ further education, teaching risk communication to some 1,000 German gynecologists, among others. Keynotes to medical organizations and conferences. In the last 2 years, Gerd Gigerenzer has given some 20 keynotes and lectures to medical organizations in various countries. Wolfgang Gaissmaier has co-organized a conference on “communicating risk and benefits in medicine,” together with the Evangelische Akademie (Protestant Academy), Berlin, and the Institut Mensch, Ethik und Wissenschaft (Research Institute on Medical Ethics), Berlin. The work on risk communication in medicine is a collaboration between the Max Planck Institute for Human Development and the Berliner Ärztekammer (Berlin Medical Council), the Ärztliches Zentrum für Qualitätssicherung Berlin (Center for Quality Control in Medicine), and the University of Washington School of Medicine. A new project with the Deutsche Krebshilfe e.V. (German Cancer Aid) is currently underway. It addresses the problem that the information concerning cancer screening in its widely distributed brochures relies on exactly those representations of potential benefits and harms that are misunderstood by the public, and in its attempt to overstate the benefits of cancer screening and downplay its harms to patients also conflicts with the evidence from medical research. The Deutsche Krebshilfe has reacted positively to ABC’s input on this issue, and we are now helping in the design of new brochures which will present evidence-based recommendations in a transparent manner.

Center for Adaptive Behavior and Cognition | 55 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 56

of the same information and introducing Statistics Education proper representations in conversation Statistical thinking—learning to live with and written communication can provide uncertainty—is the most important part of an effective remedy for physicians’ and mathematics in everyday life after school. patients’ lack of understanding. Thinking means calling certainties into ques- Yet at present, physicians and patients alike tion. Yet the curricula in Western countries know little about how to represent crucial are almost entirely preoccupied with the statistical numbers comprehensibly. For mathematics of certainty, from algebra to instance, the colorectal screening example geometry to trigonometry. The central role illustrates that conditional probabilities (such statistical thinking plays for educated citi- as sensitivity and false positive rate) tend to zens in a modern technological world has cloud physicians’ minds, and we teach physi- not yet been recognized in elementary cians to translate these into natural frequen- school, high school, or in the university edu- cies: think of 10,000 patients. We expect 30 cation of physicians and lawyers. If it is to have cancer, of which 15 will test positive. taught, statistics is often presented as a dry Of those without cancer, some 300 will also subject, interest is killed by boring examples, test positive. Thus, we expect that out of a and understanding made difficult by repre- total of 330 patients who test positive, 15 sentations that are hard to digest by young actually have cancer (about 5 %). Similarly, minds. We proudly report that there is now relative risks, such as “a 25 % reduction of the first textbook on stochastics used in breast cancer,” cloud physicians’ minds, German high schools (Jahnke & Wuttke, mostly without them noticing, whereas 2005), with a section on decision making absolute risks or stating the number of that begins with a motivating real-world patients needed to treat turn innumeracy topic (HIV tests) as well as transparent rep- into insight. The reason for the confusion, we resentation (natural frequencies)—both taken hypothesize, is the reference class. from Gigerenzer (2002)—in order to intro- Sensitivities and false positive rates condi- duce probabilistic inference. tionalize numerical information with respect Zhu and Gigerenzer (2006) have shown, for to different reference classes, which makes the first time, that by being exposed to nat- Bayesian inference computationally more ural frequencies (as opposed to relative fre- complex than inference based on natural quencies or conditional probabilities), 4th, frequencies, which all relate to the same ref- 5th, and 6th graders can solve Bayesian erence class. Similarly, relative risks refer to problems without any teaching. Note that a reference class which is different from the until recently, psychological research claimed one patients assume. that even college students are unable to In our studies, most patients assume that solve Bayesian problems. The older the chil- “25 %” refers to the class of women who dren were, the better they performed, with participate in screening, and conclude that 19 %, 39 %, and 53 % of the Bayesian prob- 25 % of those are saved from dying from lems being solved, respectively. At the indi- breast cancer. Yet this number refers to the vidual level, however, the transition was not class of women who die from breast cancer as gradual, but indicated a qualitative without screening within 10 years. In this insight. Most children either solved all prob- class, 4 out of every 1,000 women die from lems or none. In 6th graders, the perform- breast cancer, whereas the corresponding ance with natural frequencies was as good number is 3 out of 1,000 among those who as that of MBA students with conditional participate in screening. Thus, the absolute probabilities. An analysis of the process of risk reduction is 1 in 1,000, while the relative thinking revealed several pre-Bayesian rules risk reduction is 25 %. To promote insight of thumb whose frequency distribution into crucial numbers in health care, we have changed over developmental course. This been active at various levels, as shown in result indicates that complex statistical Box, p. 55. thinking can be taught at a much earlier age

56 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 57

than previously assumed, provided one uses September 11 did not strike once, but twice— Key References a proper representation of numerical infor- the second time through the minds of many Gigerenzer, G. (2006f). mation. In collaboration with the Pädago- Americans who avoided flying, took their Out of the frying pan into the fire: Behavioral gische Hochschule Ludwigsburg, we design cars instead, and died in fatal traffic acci- reactions to terrorist at- and test visual and tactile representations of dents (see Figure 11). tacks. Risk Analysis, 26, numbers that can be used already in elemen- Terrorist attacks cause direct damage (the 347–351. tary school. immediate consequences of terrorist action) Zhu, L., & Gigerenzer, G. A beneficial amount of statistical thinking as well as indirect damage, which is not (2006). Children can solve Bayesian prob- could save lives. Low-probability, high-dam- under the terrorists’ control, but due to pub- lems: The role of repre- age events in which many people are killed lic and institutional reactions. This initial sentation in mental at one point in time are called dread risks. As analysis was restricted to the first 3 months computation. Cognition, 98, 287–308. opposed to situations in which a similar after the attack. In a full analysis of the 18 Krämer, W., & number of people or more are killed over a months after the attack, Gigerenzer (2006f) Gigerenzer, G. (2005). longer period of time, people tend to react to found that, in response to the decline in air How to confuse with dread risks with avoidance behavior. The travel, car travel increased up to 5.2 % for a statistics or: The use and misuse of conditional crash of the four planes in the terrorist period of 12 months, specifically on inter- probabilities. Statistical attack on September 11, 2001, exemplifies state highways, and then returned to normal. Science, 20, 223–230. such a catastrophic event. In contrast, the During these 12 months, an estimated 1,500 estimated 44,000 to 98,000 patients who die Americans lost their lives on the roads in the every year in U. S. hospitals because of pre- attempt to avoid the risk of flying. ventable and documented medical errors do Does this dread risk avoidance generalize to not constitute a dread risk, and public atten- other cultures? On March 11, 2004, exactly tion to these preventable death tolls is 2 1/2 years after September 11, the bombings minor. In an earlier publication, Gigerenzer of four commuter trains during the Madrid (2004) showed that the terrorists of rush hour killed about 200 people and

300

200

100

–100 Number of fatal crashes

above/below monthly average –200

–300 Jul Jul Feb Feb Feb Apr Jun Oct Apr Jun Oct Sep Sep Dec Dec Aug Nov Aug Nov Mar Mar Mar May May 2001 Jan 2001 2002 Jan 2002 2003 Jan

Figure 11. The number of fatal traffic accidents in the United States increased after the terrorist attacks on Sep- tember 11, 2001, for a period of 12 months. Numbers are expressed as deviations from the five-year base line 1996-2000 (the zero line). The error bars (shown for the 12 months following the terrorist attacks) specify the maximum and the minimum numbers for each month of the base line. Before September 11, the average of the monthly numbers of fatal traffic accidents for 2001 was close to the zero line, and the monthly values were always within the maximum and minimum of the previous five years. Yet in the 12 months following the terrorist attacks (October 2001 to September 2002, shown in red), the number of fatal traffic accidents every month was higher than the zero line, and in most cases exceeded the maximum of the previous years. Data are taken from the U.S. Department of Transportation, Federal Highway Administration.

Center for Adaptive Behavior and Cognition | 57 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 58

wounded 1,460. Like Americans avoided known to most citizens. How many miles planes, Spaniards tended to avoid trains, but does one have to drive so that the risk of the effect was smaller and shorter—only dying in a car accident is the same as in a 2 months. Yet Spaniards did not increase the nonstop flight, say from Boston to Frankfurt? intensity of their car travel, and, consequent- Few are aware that the answer is about ly, no additional loss of lives on the roads 12 miles. That is, if one arrives safely by car was observed. We can only speculate why; at the airport, the most dangerous part of there is less of a car culture in Spain than in the trip may be over. In collaboration with the United States, and a better developed the Max Planck Institute for Foreign and public transport system. International Criminal Law, we are applying It is not always possible to overcome an presently for EU funds to study the indirect emotional fear reaction with statistical rea- harms due to reactions of individuals and soning. Yet, the very elementary facts that institutions to terrorist attacks and threats would empower reasoning are not even on a larger scale.

The ABC Research Group in February 2007

Left to right: Henry Brighton, Nils Straubinger, Juliet Conlin, Henrik Olsson, Magnus Persson, David Funder, Jing Qian, Gerd Gigerenzer, Wolfgang Gaissmaier, Shabnam Mousavi, Jörg Rieskamp, Jeffrey Stevens, Monika Keller, Lael Schooler, Ana Sofia Alves Conte de Morais, Odette Wegwarth, Mirta Galesic, Julian Marewski, Nathan Berg, Benjamin Scheibehenne.

58 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 59

Publications 2005–2006 (last update: May 2007)

Adamowicz, W., Hanemann, M., Publishing House (Proceedings of Berg, N., & Maital, S. (in press). Bröder, A. (2006). Experiment. In Swait, J., Johnson, R., Layton, D., the 6th Annual Conference of the Tailoring globalization to national J. Funke & P. A. Frensch (Eds.), Regenwetter, M., Reimer, T., & State University—Higher School of needs and wellbeing: One size Handbuch der Allgemeinen Psy- Sorkin, R. (2005). Decision strat- Economics, Moscow). never fits all. Global Business and chologie: Vol. 5. Kognition egy and structure in households: Economics Review. (pp. 717–725). Göttingen: A "groups" perspective. Market- Berg, N. (2005c). Non-response Hogrefe. ing Letters, 16, 387–399. bias. In K. Kempf-Leonard (Ed.), Boom, J., Wouters, H., & Keller, Encyclopedia of social measure- M. (2007). A cross-cultural vali- Bröder, A. (2005). Entscheiden Bachmann, L. M., Gutzwiller, F. S., ment (Vol. 2, pp. 865–873). dation of stage development: A mit der "adaptiven Werkzeug- Puhan, M. A., Steurer, J., Steurer- London: Academic Press. Rasch re-analysis of longitudinal kiste": Ein empirisches For- Stey, C., & Gigerenzer, G. (in socio-moral reasonig data. schungsprogramm. Lengerich: press). Health knowledge in the Berg, N. (2005d). [Review of the Cognitive Development, 22, Pabst Science Publishers. community—A survey. BMC book Conflict and Cooperation: 213–229. Medicine Journal. Institutional and Behavioral Bröder, A., & Eichler, A. (2006). Economics]. Journal of Socio- Brand, S., Reimer, T., & Opwis, K. The use of recognition and addi- Barrett, H. C., & Behne, T. (2005). Economics, 34, 866–868. (2007). How do we learn in a neg- tional cues in inferences from Children's understanding of death ative mood? Effects of a negative memory. Acta Psychologica, 121, as the cessation of agency: A test Berg, N., & Gigerenzer, G. mood on transfer and learning. 275–284. using sleep versus death. Cogni- (2007). Psychology implies pater- Learning and Instruction, 17, tion, 96, 93–108. nalism? Bounded rationality may 1–16. Bröder, A., & Gaissmaier, W. (in reduce the rationale to regulate press). Sequential processing of Barrett, H. C., Todd, P. M., risk-taking. Social Choice and Brandstätter, E., Gigerenzer, G., cues in memory-based multi- Miller, G. F., & Blythe, P. W. Welfare, 28, 337–359. & Hertwig, R. (2006). The priority attribute decisions. Psychonomic (2005). Accurate judgments of heuristic: Making choices without Bulletin and Review. intention from motion cues Berg, N., & Gigerenzer, G. trade-offs. Psychological Review, alone: A cross-cultural study. (2006). Peacemaking among in- 113, 409–432. Bröder, A., & Schiffer, S. (2006a). Evolution and Human Behavior, consistent rationalities? Comment Adaptive flexibility and maladap- 26, 313–331. on Alex Kacelnik et al. In C. Engel Brighton, H. J. (2006). Robust tive routines in selecting fast and & L. Daston (Eds.), Is there value inference with simple cognitive frugal decision strategies. Journal Bennis, W. M., & Pachur, T. in inconsistency? (pp. 423–433). models. In C. Lebiere & B. Wray of Experimental Psychology: (2006). Fast and frugal heuristics Baden-Baden: Nomos (Common (Eds.), Between a rock and a Learning, Memory, and Cogni- in sports. Psychology of Sport goods: Law, politics and eco- hard place: Cognitive science tion, 32, 904–918. and Exercise, 7, 611–629. nomics 15). principles meet AI-hard problems. Papers from the Bröder, A., & Schiffer, S. (2006b). Berg, N. (in press-a). Behavioral Berg, N., Gu, A., & Lien, D. (in AAAI Spring Symposium Stimulus format and working labor economics. In M. Altman press). Dynamic correlation: A (AAAI Tech. Rep. No. SS–06–03, memory in fast and frugal strat- (Ed.), Handbook of behavioral tool for hedging house-price risk? pp. 17–22). Menlo Park, egy selection. Journal of Behav- economics. Armonk, NY: Sharpe. Journal of Real Estate Portfolio CA: AAAI Press. ioral Decision Making, 19, Management. 361–380. Berg, N. (in press-b). Finance, psy- Brighton, H. J., & Kirby, S. chology, economics and the design Berg, N., & Hoffrage, U. (2005). (2006). Understanding linguistic Brunner, T., Reimer, T., & Opwis, of successful institutions. In Environmental determinants of evolution by visualizing the emer- K. (2005). Cancellation and focus: Shanghai Forum 2005 Anthology. simple decision rules: No cogni- gence of topographic mappings. The impact of feature attractive- Shanghai: Fudan University Press. tive limitations needed. In K. Artificial Life, 12, 229–242. ness on recall. In K. Opwis & I.-K. Opwis & I.-K. Penner (Eds.), Penner (Eds.), Proceedings of Berg, N. (2006). A simple Proceedings of KogWis05: The Brighton, H. J., Mata, R., & KogWis05: The German Cogni- Bayesian procedure for sample German Cognitive Science Wilke, A. (2006). Reconciling tive Science Conference 2005 size determination in an audit of Conference 2005 (pp. 9–14). vague and formal models of lan- (pp. 27–32). Basel: Schwabe. property value appraisals. Real Basel: Schwabe. guage evolution. Behavioral and Estate Economics, 34, 133–155. Brain Sciences, 29, 282. Dawes, C. T., Fowler, J. H., Johnson, Berg, N., & Lien, D. (2006). T., McElreath, R., & Smirnov, O. Berg, N. (2005a). Decision- Same-sex sexual behavior: U.S. Brighton, H. J., Smith, K., & (2007). Egalitarian motives in hu- making environments in which frequency estimates from survey Kirby, S. (2005). Language as an mans. Nature, 446, 794–796. unboundedly rational decision data with simultaneous mis- evolutionary system. Physics of makers choose to ignore relevant reporting and non-response. Life Reviews, 2 (3), 177–226. Dieckmann, A., & Gigerenzer, information. Global Business and Applied Economics, 38 (7), G. (2005). Macht Halbwissen Economics Review, 7, 59–73. 757–769. Brighton, H. J., & Todd, P. M. klug: Ist mehr Information immer (in press). Situating rationality: besser? Berliner Ärzte, 42, 16–19. Berg, N. (2005b). Psychological Berg, N., & Lien, D. (2005). Does Ecologically rational decision economics and the size of govern- society benefit from investor making with simple heuristics. In Dieckmann, A., & Krauss, S. ment: On anti-, anti-anti-, ... , overconfidence in the ability of P. Robbins & M. Aydede (Eds.), (2005). Wenn weniger Wissen (anti-)n. In Y. Yasin (Ed.), Economic financial market experts? Journal Cambridge handbook of situated mehr sein kann: Einfache Heuris- modernization and cultivation of of Economic Behavior and Orga- cognition. Cambridge, UK: tiken zur psychologischen Ent- institutions (pp. 367–375). HSE nization, 58, 95–116. Cambridge University Press. scheidungsfindung. Zeitschrift für

Center for Adaptive Behavior and Cognition | 59 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 60

Erziehungswissenschaft, 8, chologie (pp. 31–49). Lengerich: García-Retamero, R., & Hoffrage, Gigerenzer, G. (in press-g). Why 187–201. Pabst Science Publishers. U. (2006). How causal knowledge heuristics work. Perspectives on simplifies decision-making. Minds Psychological Science. Dieckmann, A., & Rieskamp, J. Gaissmaier, W., Schooler, L. J., & Machines, 16, 365–380. (in press.) The influence of in- & Rieskamp, J. (2006). Simple Gigerenzer, G. (in press-h). formation redundancy on prob- predictions fueled by capacity García-Retamero, R., & López- Wissenschaftler an einem Max- abilistic inferences. Memory & limitations: When are they suc- Zafra, E. (2006a). Congruencia Planck-Institut. In K. Weis & Cognition. cessful? Journal of Experimental de rol de género y liderazgo: M. Amelang (Eds.), Psychologen Psychology: Learning, Memory, El papel de las atribuciones im Beruf: Anforderungen, Chan- Dieckmann, A., & Todd, P. M. and Cognition, 32, 966–982. causales sobre el éxito y el cen, Perspektiven. Stuttgart: (2005). Simple ways to construct fracaso. Revista Latino- Kohlhammer. search order. In K. Forbus, D. Gaissmaier, W., Schooler, L. J., americana de Psicologia, 38, Genter, & T. Reqier (Eds.), Pro- & Gigerenzer, G. (2005). Re- 245–257. Gigerenzer, G., Hoffrage, U., & ceedings of the 26th Annual ceptive memory in judgment Goldstein, D. G. (in press). Fast Conference of the Cognitive and decision making. In M.-L. García-Retamero, R., & López- and frugal heuristics are plausi- Science Society (pp. 309–314). Käsermann & A. Altorfer (Eds.), Zafra, E. (2006b). Prejudice ble models of cognition: Reply Mahwah, NJ: Erlbaum. Über Lernen: Ein Gedanken- against women in male-congenial to Dougherty, Franco-Watkins, austausch (pp. 67–79). Bern: environments: Perceptions of and Thomas. Psychological Elmore, J. G., & Gigerenzer, G. EditionSolo. gender role congruity in leader- Review. (2005a). Benign breast disease— ship. Sex and Roles, 55, 51–61. the risks of communicating risk. Galesic, M., & Tourangeau, R. Gigerenzer, G. (2007a). Bauch- New England Journal of Medi- (2007). What is sexual harass- García-Retamero, R., entscheidungen: Die Intelligenz cine, 353, 297–299. ment? It depends on who asks! Takezawa, M., & Gigerenzer, G. des Unbewussten und die Macht Framing effects of sponsorship (2006). How to learn good cue der Intuition. München: Bertels- Elmore, J. G., & Gigerenzer, G. on survey responses. Applied orders: When social learning mann. (2005b). The editorialists reply. Cognitive Psychology, 21, 189– benefits simple heuristics. New England Journal of Medi- 202. Proceedings of the 28th Annual Gigerenzer, G. (2007b). Gut feel- cine, 353, 1857–1858. Conference of the Cognitive ings: The intelligence of the García-Retamero, R. (2007). The Science Society, 1352–1357. unconscious. New York: Viking Elwin, E., Juslin, P., Olsson, H., & influence of knowledge about Press. Enkvist, T. (2006). Constructivist causal mechanisms on compound García-Retamero, R., Wallin, A., encoding: Learning from selective processing. The Psychological & Dieckmann, A. (in press). Does feedback. Psychological Science, Record, 57, 295–306). causal knowledge help us be 18, 105–110. faster and more frugal in our García-Retamero, R. (2006). decisions? Memory & Cognition. Engel, C., & Gigerenzer, G. Identidad de género y nivel de (2006). Law and heuristics: An aspiraciones profesionales en Gigerenzer, G. (in press-a). interdisciplinary venture. In G. alumnos universitarios [Gender Bounded rationality. In D. Sander Gigerenzer & C. Engel (Eds.), identitiy and level of profes- & K. Scherer (Eds.), Oxford com- Heuristics and the law (pp. 1– sional aspirations in undergrad- panion to the affective sciences. 16). Cambridge, MA: MIT Press. uates]. Revista Mexicana de New York: Oxford University Psicología, 23, 217–224. Press. Fasolo, B., McClelland, G. H., & Todd, P. M. (2007). Escaping the García-Retamero, R., & Gigerenzer, G. (in press-b). tyranny of choice: When fewer Dieckmann, A. (2006). Una Moral intuition = fast and frugal attributes make choice easier. visión crítica del enfoque de los heuristics? In W. Sinnott- Marketing Theory, 7, 13–26. heurísticos rápidos y frugales Armstrong (Ed.), The cognitive [A critical view of the approach science of morality. Fiddick, L., & Rutherford, M. D. of fast and frugal heuristics]. (2006). Looking for loss in all the Revista Latinoamericana de Gigerenzer, G. (in press-c). Ra- wrong places: Loss avoidance Psicología, 38, 509–522. tionality for mortals. New York: Gigerenzer, G. (2007c). Short does not explain cheater detec- Oxford University Press. cuts: The intelligence of the tion. Evolution and Human García-Retamero, R., Hoffrage, unconscious. London: Penguin Behavior, 27, 417–432. U., & Dieckmann, A. (in press). Gigerenzer, G. (in press-d). Reply Books. When one cue is not enough: to comments. In W. Sinnott- Fowler, J. H., Johnson, T., & Combining fast and frugal heuris- Armstrong (Ed.), The cognitive Gigerenzer, G. (2006a). Bounded Smirnov, O. (2005). Egalitarian tics with compound cue process- science of morality. and rational. In R. J. Stainton motive and altruistic punishment. ing. The Quarterly Journal of Ex- (Ed.), Contemporary debates in Nature, 433, E1. perimental Psychology. Gigerenzer, G. (in press-e). The cognitive science (pp. 115–133). role of representation in Oxford, UK: Blackwell (Contempo- Gaissmaier, W., & Gigerenzer, García-Retamero, R., Hoffrage, Bayesian reasoning: Correcting rary Debates in Philosophy 7). G. (2006). Wie funktioniert Intui- U., Dieckmann, A., & Ramos, M. common misconceptions. Be- tion? In E. H. Witte (Ed.), Evolu- M. (2007). Compound cue pro- havioral and Brain Sciences. Gigerenzer, G. (2006b). Einfache tionäre Sozialpsychologie und cessing within the fast and frugal Heuristiken für komplexe automatische Prozesse: Beiträge heuristics approach in nonlinearly Gigerenzer, G. (in press-f). What Entscheidungen. In Berlin- des 21. Hamburger Symposions separable environments. Learning is bounded rationality? Supreme Brandenburgische Akademie der zur Methodologie der Sozialpsy- and Motivation, 38, 16–34. Court Economic Review. Wissenschaften (Ed.), Mathema-

60 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 61

tisierung der Natur (pp. 37–44). Manfred Wettler (pp. 19–29). common misconceptions. Behav- ern and Asian societies. Interna- Berlin (Debatte No. 4). Lengerich: Pabst Science ioral and Brain Sciences. tional Journal of Behavioral Publishers. Development. Gigerenzer, G. (2006c). Einfache Gigerenzer, G., & Kurzenhäuser, Heuristiken für komplexe Ent- Gigerenzer, G., & Engel, C. S. (2005). Fast and frugal heuris- Gummerum, M., Keller, M., scheidungen. Jahrbuch 2005 der (Eds.). (2006). Heuristics and the tics in medical decision making. Takezawa, M., & Mata, J. (in Deutschen Akademie der Natur- law: Report of the 94th Dahlem In R. Bibace, J. D. Laird, K. L. press). To give or not to give: forscher Leopoldina, 51, 337– Workshop on heuristics and law, Noller, & J. Valsiner (Eds.), Science Children’s and adolescents’ 343. Berlin, June 6–11, 2004. Cam- and medicine in dialogue: Think- sharing and negotiations in eco- bridge, MA: MIT Press. ing through particulars and uni- nomic decision situations. Child Gigerenzer, G. (2006d). Follow versals (pp. 3–15). Westport, CT: Development. the leader. Harvard Business Praeger. Review, 84, 58–59. Hagen, E. H., Barrett, H. C., & Gigerenzer, G., Martignon, L., Price, M. E. (2006). Do human Gigerenzer, G. (2006e). Heuris- Hoffrage, U., Rieskamp, J., parents face a quantity-quality tics. In G. Gigerenzer & C. Engel Czerlinski, J., & Goldstein, D. G. tradeoff? Evidence from a Shuar (Eds.), Heuristics and the law (in press). One-reason decision community. American Journal of (pp. 17–44). Cambridge, MA: MIT making. In C. R. Plott & V. L. Physical Anthropology, 130, Press. Smith (Eds.), Handbook of results 405–418. in experimental economics. Gigerenzer, G. (2006f). Out of Amsterdam: Elsevier. Hanoch, Y. (2005). One theory to the frying pan into the fire: Be- fit them all: The search hypothesis havioral reactions to terrorist at- Gigerenzer, G., & Sturm, T. of emotion revisited. The British tacks. Risk Analysis, 26, 347–351. (2007). Tools = theories = data? Journal for the Philosophy of On some circular dynamics in Science, 56, 135–145. Gigerenzer, G. (2006g). What's cognitive science. In M. G. Ash & in a sample? A manual for build- T. Sturm (Eds.), Psychology's ter- Hanoch, Y., Johnson, J. G., & ing cognitive theories. In K. ritories: Historical and contem- Wilke, A. (2006). Domain speci- Fiedler & P. Juslin (Eds.), Informa- porary perspectives form differ- ficity in experimental measures tion sampling and adaptive cog- Gigerenzer, G., & Gaissmaier, ent disciplines (pp. 305–342). and participant recruitment: An nition (pp. 239–260). New York: W. (2006a). Denken und Urteilen Mahwah, NJ: Erlbaum. application to risk-taking behav- Cambridge University Press. unter Unsicherheit: Kognitive ior. Psychological Science, 17, Heuristiken. In J. Funke (Ed.), Gigerenzer, G., & Todd, P. M. 300–304. Gigerenzer, G. (2005a). Calcular Enzyklopädie der Psychologie: (in press). Rationality the fast and o risco: Aprender a lidar com a Vol. C, II, 8. Denken und Problem- frugal way: Introduction. In C. R. Hanoch, Y., Katsikopoulos, K. V., incerteza [Calculated risks: How lösen (pp. 329–374). Göttingen: Plott & V. L. Smith (Eds.), Hand- Gummerum, M., & Brass, E. P. to know when numbers deceive Hogrefe. book of results in experimental (in press). German and American you]. Lisbon: Gradiva. economics. Amsterdam: Elsevier. students' knowledge, percepetion, Gigerenzer, G., & Gaissmaier, and behavior with respect to Gigerenzer, G. (2005b). Is the W. (2006b). Ironie des Terrors. Goldstein, D. G., & Gigerenzer, G. over-the-counter pain relievers. mind irrational or ecologically Gehirn & Geist, 9, 14–16. (in press). Ignorance-based deci- Health Psychology. rational? In F. Parisi & V. L. Smith sion making and the less-is-more (Eds.), The law and economics of Gigerenzer, G., & Gigerenzer, T. paradox. In C. R. Plott & V. L. von Helversen, B., & Rieskamp, irrational behavior (pp. 37–67). (2005). Is the ultimatum game a Smith (Eds.), Handbooks of ex- J. (in press). The mapping Stanford, CA: Stanford University three-body affair? Behavioral perimental economics results. model: A cognitive theory of Press. and Brain Sciences, 28, 823–824. Amsterdam: Elsevier/North quantitative estimation. Journal Holland. of Experimental Psychology: Gigerenzer, G. (2005c). I think, Gigerenzer G., Hertwig R., General. therefore I err. Social Research, Hoffrage U., & Sedlmeier P. (in González-Vallejo, C., & Bohnham, 72, 195–218. press). Cognitive illusions recon- A. (2007). Aligning confidence Hertwig, R., Pachur, T., & Kurzen- sidered. In C. R. Plott & V. L. Smith with accuracy: Revisiting the role häuser, S. (2005). Judgments of Gigerenzer, G. (2005d). Wie ver- (Eds.), Handbook of results in of feedback. Acta Psychologica, risk frequencies: Tests of possible steht man Risiken und Unsicher- experimental economics. 125, 221–239. cognitive mechanisms. Journal of heiten? HNO Informationen, 4, Amsterdam: Elsevier/North- Experimental Psychology: Learn- 287–294. Holland. González-Vallejo, C., & Reid, A. A. ing, Memory, and Cognition, 31, (2006). Quantifying persuasion 621–642. Gigerenzer, G., & Brighton, H. Gigerenzer, G., Hertwig, R., van effects on choice with the deci- J. (in press). Can hunches be ra- den Broek, E., Fasolo, B., sion threshold of the stochastic Hess, N. H., & Hagen, E. H. (2006). tional? In C. S. Lerner & D. Polsby Katsikopoulos, K. V. (2005). A choice model. Organizational Psychological adaptations for as- (Eds.), Mere hunches: Policing in 30 % chance of rain tomorrow: Behavior and Human Decision sessing gossip veracity. Human the age of terror. How does the public understand Processes, 100, 250–267. Nature, 17, 337–354. probabilistic weather forecasts? Gigerenzer, G., & Dieckmann, Risk Analysis, 25, 623–629. Gummerum, M., & Keller, M. Hill, H., & Raab, M. (2005). Ana- A. (2006). Oblivious confusion. In (in press). Affection, virtue, lyzing a complex visuomotor R. Rapp, P. Sedlmeier, & G. Gigerenzer, G., & Hoffrage, U. (in pleasure, and profit: Developing tracking task with brain-electrical Zunker-Rapp (Eds.), Perspectives press). The role of representation an understanding of friendship event related potentials. Human on cognition: A Festschrift for in Bayesian reasoning: Correcting closeness and intimacy in West- Movement Science, 24, 1–30.

Center for Adaptive Behavior and Cognition | 61 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 62

Hoffrage, U. (2006). Achtung Sta- Hoffrage, U., & Vitouch, O. (in Katsikopoulos, K. V., & Fasolo, Keller, M., & Canz, T. (2007) tistik! Wissenschaftsjournalist, press). Evolutionäre Psychologie B. (2006). New tools for decision Zwischen Fairness und Inte- 18–19. des Denkens und Problemlösens. analysts. IEEE Transactions on resse: Eine entwicklungspsy- In J. Müsseler (Ed.), Allgemeine Systems, Man, and Cybernetics, chologische Studie zu Verhand- Hoffrage, U., García-Retamero, Psychologie. Heidelberg: Part A: Systems and Humans, 36, lungen in Gruppen. In A. A. R., & Czienskowski, U. (in press). Spektrum. 960–967. Bucher (Ed.), Moral, Religion, Compound cue preocessing in lin- Politik: Psychologisch-päda- early and nonlinearly separable Hutchinson, J. M. C., & Katsikopoulos, K. V., & Hanoch, gogische Zugänge: Festschrift environments. The Psychological Griffith, S. C. (in press). Extra- Y. (2005). European Journal of für Fritz Oser (pp. 113–139). Record. pair paternity in the skylark Cognitive Psychology, 17, 155– Wien: LIT Verlag. alauda arvensis. Ibis. 158 [Review of the book Prefer- Hoffrage, U., García-Retamero, ence, belief, and similarity: Se- Keller, M., & Krettenauer, T. R., & Czienskowski, U. (2005a). Hutchinson, J. M. C. (2005). Is lected writings of Amos Tversky]. (2007). Moralentwicklung im The robustness of the Take the more choice always desirable? Kulturvergleich. In G. Trommsdorf Best configural heuristic in lin- Evidence and arguments from Katsikopoulos, K. V., & & H. J. Kornadt (Eds.), Enzyk- early and nonlinearly separable leks, food selection, and environ- Martignon, L. (2006). Naive lopädie der Psychologie: Vol. C, environments. In B. G. Bara, L. mental enrichment. Biological heuristics for paired comparisons: VII, 2. Erleben und Handeln im Barsalou, & M. Bucciarelli (Eds.), Reviews, 80, 73–92. Some results on their relative ac- kulturellen Kontext (pp. 521– Proceedings of the 27th Annual curacy. Journal of Mathematical 555). Göttingen: Hogrefe. Conference of the Cognitive Hutchinson, J. M. C., & Psychology, 50, 488–494. Science Society (pp. 971–976). Gigerenzer, G. (2005a). Connect- Keller, M. (2006). The develop- Mahwah, NJ: Erlbaum. ing behavioural biologists and Katsikopoulos, K. V., & Simsek, ment of obligations and responsi- psychologists: Clarifying distinc- Ö. (2005). Optimal doubling strat- bilities in cultural context. In L. Hoffrage, U., García-Retamero, tions and suggestions for further egy against a suboptimal oppo- Smith & J. Vonèche (Eds.), Norms R., & Czienskowski, U. (2005b). work. Behavioural Processes, 69, nent. Journal of Applied Proba- in human development (pp. 169– The robustness of the Take the 159–163. bility, 42, 867–872. 188). Cambridge, UK: Cambridge Best configural heuristic in lin- University Press. early and nonlinearly separable Hutchinson, J. M. C., & Keller, M., & Becker, G. (in environments. In K. Opwis & I.-K. Gigerenzer, G. (2005b). Simple press). Ein handlungstheore- Keller, M. (2005). Moralentwick- Penner (Eds.), Proceedings of heuristics and rules of thumb: tischer Ansatz zur Entwicklung lung und moralische Sozialisation. KogWis05: The German Cogni- Where psychologists and behav- und Förderung sozio-morali- In D. Horster & J. Oelkers (Eds.), tive Science Conference 2005 ioural biologists might meet. scher Kompetenz. In T. Malti & Pädagogik und Ethik (pp. 149– (pp. 83–88). Basel: Schwabe. Behavioural Processes, 69, S. Perren (Eds.), Entwicklung und 172). Wiesbaden: VS Verlag für 97–124. Förderung sozialer Kompetenzen Sozialwissenschaften. Hoffrage, U., & Hertwig, R. (2006). in Kindheit und Adoleszenz. Which world should be repre- Hutchinson, J. M. C., & Waser, Stuttgart: Kohlhammer. Keller, M., Edelstein, W., sented in representative design? P. (in press). Use, misuse and ex- Krettenauer, T., Fang, F.-X., & In K. Fiedler & P. Juslin (Eds.), In- tensions of "ideal gas" models of Keller, M., Brandt, A. & Fang, G. (2005). Reasoning about formation sampling and adap- animal encounter. Biological Sigurdardottir, G. (in press). Happy moral obligations and interper- tive cognition (pp. 381–408). Reviews. victimizers and unhappy moral- sonal responsibilities in different New York: Cambridge University ists: The development of moral cultural contexts. In W. Edelstein Press. Johnson, T., Myagkov, M., & emotions in a cross-cultural per- & G. Nunner-Winkler (Eds.), Orbell, J. (2005). Sociality as a de- spective. In W. Koops & A. Morality in context (pp. 17–337). Hoffrage, U., Hertwig, R., & fensive response to the threat of Sanders (Eds.), The development Amsterdam: Elsevier (Advances in Gigerenzer, G. (2005). Die ökolo- loss. Politics and the Life and structure of conscience. psychology 137). gische Rationalität einfacher Sciences, 23, 13–19. Hove, UK: Psychology Press. Entscheidungs- und Urteilsheuris- Krämer, W., & Gigerenzer, G. tiken. In H. Siegenthaler (Ed.), Ra- Kameda, T., Takezawa, M., & Keller, M., & Malti, T. (in press). (2005). How to confuse with tionalität im Prozess kultureller Hastie, R. (2005). Where do social Sozialisation sozio-moralischer statistics or: The use and misuse Evolution: Rationalitäts- norms come from? The example Kompetenzen: Eine kognitiv- of conditional probabilities. unterstellungen als eine Bedin- of communal sharing. Current entwicklungspsychologische Statistical Science, 20, 223–230. gung der Möglichkeit sub- Directions in Psychological Perspektive. In K. Hurrelmann, stantieller Rationalität des Han- Science, 14, 331–334. M. Grundmann & S. Walper Krappmann, L. (2005a). Is delns (pp. 65–89). Tübingen: (Eds.), Handbuch der Sozialisa- community compatible with Mohr/Siebeck. Katsikopoulos, K. V. (in press). tionsforschung. Weinheim: autonomy? Some comments to (P, p) retraining policies. IEEE Beltz. Joan Miller's research on differ- Hoffrage, U., Kurzenhäuser, S., & Transactions on Systems, Man & ing moralities in India and the Gigerenzer, G. (2005). Under- Cybernetics: Systems & Humans. Keller, M., Malti, T., & Dravenau, United States. In W. Edelstein & standing the results of medical D. (in press). Bindung, Sozialisa- G. Nunner-Winkler (Eds.), tests: Why the representation of Katsikopoulos, K. V. (2005, tion und die Entwicklung des Morality in context (pp. 313– statistical information matters. In Autumn). [Review of the book sozio-moralischen Denkens von 316). Amsterdam: Elsevier R. Bibace, J. D. Laird, K. L. Noller, Thinking: Psychological perspec- der Kindheit bis zum Jugendalter: (Advances in psychology 137). & J. Valsiner (Eds.), Science and tives on reasoning, judgment, and Befunde einer Längsschnittunter- medicine in dialogue: Thinking decision making]. Retrieved suchung. In C. Hopf & G. Nunner- Krappmann, L. (2005b). Soziolo- through particulars and univer- November 20, 2006, from Winkler (Eds.), Bindung und gische Dimensionen der Iden- sals (pp. 83–98). Westport, CT: http://www.psylone.com/eadm- moralische Entwicklung. tität: Strukturelle Bedingungen Praeger. bulletin.html Weinheim: Juventa. für die Teilnahme an Inter-

62 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 63

aktionsprozessen (10th ed.). Mata, J., Dieckmann, A., & 2005 (pp. 33–37). München: Effects of the information envi- Stuttgart: Klett-Cotta. Gigerenzer, G. (2005). Verständ- Selbstverlag. ronment on group discussions liche Risikokommunikation, leicht and decisions in the hidden- Kurz-Milcke, E., & Gigerenzer, G. gemacht – oder: Wie man ver- Penke, L., Todd, P. M., Lenton, A. profile paradigm. Communication (2007). Heuristic decision making. wirrende Wahrscheinlichkeits- P., & Fasolo, B. (in press). How Monographs, 74, 1–28. Marketing, 3, 48–56. angaben vermeidet. Zeitschrift self-assessments can guide hu- für Allgemeinmedizin, 81, man mating decisions. In G. Geher Reimer, T., Mata, R., Kysar, D. A., Ayton, P., Frank, R. H., 537–541. & G. F. Miller (Eds.), Mating intel- Katsikopoulos, K. V., & Opwis, Frey, B. S., Gigerenzer, G., ligence: Sex, raltionships, and K. (2005). On the interplay be- Glimcher, P. W., et al. (2006). Mata, R., Schooler, L. J., & the mind’s reproductive system. tween heuristics and systematic Group report: Are heuristics a Rieskamp, J. (in press). The Mahwah, NJ: Erlbaum processes in persuasion. In B. G. problem or a solution? In G. aging decision maker: Cognitive Bara, L. Barsalou, & M. Bucciarelli Gigerenzer & C. Engel (Eds.), aging and the adaptive selection Raab, M., & Gigerenzer, G. (Eds.), Proceedings of the 27th Heuristics and the law (pp. 102– of decision strategies. Psychol- (2005). Intelligence as smart Annual Conference of the Cogni- 140). Cambridge, MA: MIT Press. ogy and Aging. heuristics. In R. J. Sternberg & J. E. tive Science Society (pp. 1833– Pretz (Eds.), Cognition and intel- 1838). Mahwah, NJ: Erlbaum. Machery, E. (2005). You don't Mata, R., Wilke, A., & Todd, P. ligence: Identifying the mecha- know how you think: Introspec- M. (2005). Adding the missing nisms of the mind (pp. 188–207). Reimer, T., Park, E., & Hinsz, V. B. tion and language of thought. The link back into mate choice re- Cambridge, UK: Cambridge Uni- (2006). Shared and coordinated British Journal for the Philoso- search. Behavioral and Brain versity Press. cognition in competitive and dy- phy of Science, 56, 469–485. Sciences, 28, 289. namic task environments: An in- Raab, M., & Green, N. (2005). Mo- formation-processing perspective Malti, T., & Keller, M. (in press). Muramatsu, R., & Hanoch, Y. tion as input: A functional expla- for team sports. International Observed aggressive behaviour (2005). Emotions as a mechanism nation of movement effects on Journal of Sport and Exercise in elementaryschool children: Its for boundedly rational agents: cognitive processes. Perceptual Psychology, 4, 376–400. relations with moral cognitions The fast and frugal way. Journal and Motor Skills, 100, 333–348. and moral emotions in close re- of Economic Psychology, 26, Reimer, T., & Rieskamp, J. (in lationships. European Journal of 201–221. Raab, M., & Reimer, T. (2007). press). Fast and frugal heuris- Developmental Psychology. Expertise in Sport: Intuitive und tics. In R. F. Baumeister & K. D. Ortmann, A., Gigerenzer, G., deliberative Entscheidungen im Vohs (Eds.), Encyclopedia of so- Marewski, J. N., Gaissmaier, Borges, B., & Goldstein, D. G. (in Sport. In N. Hagemann, M. Tietjens, cial psychology. Thousand Oaks, W., Dieckmann, A., Schooler, press). Recognition heuristic: A & B. Strauß (Eds.), Psychologie CA: Sage. L. J., & Gigerenzer, G. (in press). fast and frugal way to investment der sportlichen Höchstleistung: Applying the recognition heuris- choice? In C. R. Plott & V. L. Smith Grundlagen und Anwendungen Reise, H., Hutchinson, J. M. C., tic to politics: Antialpha as an (Eds.), Handbook of results in ex- der Expertiseforschung im Forsyth, R. G., & Forsyth, T. J. analytical tool. In Proceedings of perimental economics. Amster- Sport (pp. 93–117). Göttingen: (2005). First records of the terres- the 27th Annual Conference of dam: Elsevier/North Holland. Hogrefe. trial slug Deroceras turcicum the Cognitive Science Society. (Simroth, 1894) in Poland. Folia Mahwah, NJ: Erlbaum. Pachur, T., & Biele, G. (2007). Reimer, T., Bornstein, A.-L., Malacologica, 13, 177–179. Forecasting from ignorance: The & Opwis, K. (2005). Positive Marewski, J. N., Gaissmaier, W., use and usefulness of recognition and negative transfer effects Reise, H., Hutchinson, J. M. C., Dieckmann, A., Schooler, L. J., & in lay predictions of sports events. in groups. In T. Betsch & S. & Robinson, D. G. (2006). Two in- Gigerenzer, G. (2005). Don’t vote Acta Psychologica, 125, 99–116. Haberstroh (Eds.), The routines of troduced pest slugs: Tandonia against the recognition heuristic. decision making (pp. 175–192). budapestensis new to the Ameri- In B. G. Bara, L. Barsalou, & M. Pachur, T., & Hertwig, R. (2006) Mahwah, NJ: Erlbaum. cas, and Deroceras panormitanum Bucciarelli (Eds.), Proceedings of On the psychology of the recog- new to the Eastern USA. The the 27th Annual Conference of nition heuristic: Retrieval pri- Reimer, T., & Hoffrage, U. (2006). Veliger, 48, 110–115. the Cognitive Science Society macy as a key determinant of its The ecological rationality of sim- (p. 2524). Mahwah, NJ: Erlbaum. use. Journal of Experimental ple group heuristics: Effects of Reise, H., Visse, S., & Hutchinson, Psychology: Learning, Memory, group member strategies on deci- J. M. C. (2007). Mating behav- Mata, R. (in press). How the and Cognition, 32, 983–1002. sion accuracy. Theory and iour in the terrestrial slug Dero- central system works? It uses fast Decision, 60, 403–438. ceras gorgonium: Is extreme and frugal heuristics. In K. Forbus, Pachur, T., Rieskamp, J., & morphology associated with ex- D. Gentner, & T. Regier (Eds.), Hertwig, R. (2005). The social cir- Reimer, T., & Hoffrage, U. (2005). treme behaviour? Animal Proceedings of the 26th Annual cle heuristic: Fast and frugal deci- Can simple group heuristics de- Biology, 57, 197–215. Conference of the Cognitive Sci- sions based on small samples. In tect hidden profiles in randomly ence Society (p. 1597). Mahwah, K. Forbus, D. Gentner, & T. Regier generated environments? Swiss Rieskamp, J., & Hoffrage, U. (in NJ: Erlbaum. (Eds.), Proceedings of the 26th Journal of Psychology, 64, 21–37. press). Inferences under time Annual Conference of the Cogni- pressure: How opportunity costs Mata, R. (2005). A neurocompu- tive Science Society (pp. 1077– Reimer, T., Hoffrage, U., & affect strategy selection. Acta tational approach to decision 1082). Mahwah, NJ: Erlbaum. Katsikopoulos, K. V. (2007). Psychologica. making and aging. In W. Dutch, J. Entscheidungsheuristiken in Kacprzyk, E. Oja, & S. Zadrozny Pachur, T., Wittig, J., Dieck- Gruppen. NeuroPsycho- Rieskamp, J., & Reimer, T. (in (Eds.), Proceedings of the 15th mann, A., & Todd, P. (2005). Economics, 2, 7–29. press). Ecological rationality. In R. International Conference on Ar- Kognitive Mechanismen bei der F. Baumeister & K. D. Vohs (Eds.), tificial Neural Networks (Vol. II, Partnerwahl. In Max-Planck- Reimer, T., Kuendig, S., Hoffrage, Encyclopedia of social psychology. pp. 411–416). Gesellschaft (Ed.), Jahrbuch U., Park, E., & Hinsz, V. B. (2007). Thousand Oaks, CA: Sage.

Center for Adaptive Behavior and Cognition | 63 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 64

Rieskamp, J. (2006a). Perspec- bounded rationality: Heuristic make us smart: Ecological ration- Allgemeine Musikpsychologie tives of probalistic inference: decision making in complex ality. Current Directions in Psy- (pp. 717–766). Göttingen: Reinforcement learning and an networks. In D. S. McNamara & chological Science. Hogrefe. adaptive network compared. G. Trafton (Eds.), Proceedings Journal of Experimental Psy- of the 29th Annual Meeting of Todd, P. M., & Gigerenzer, G. Vitouch, O. (2005b). Erwerb chology: Learning, Memory, and the Cognitive Science Society. (in press-b). Social rationality. In musikalischer Expertise. In N. Cognition, 32, 1355–1370. Mahwah, NJ: Erlbaum. C. R. Plott & V. L. Smith (Eds.), Birbaumer, et al. (Series Eds.), T. H. Handbook of results in experi- Stoffer & R. Oerter (Vol. Eds.), Rieskamp, J. (2006b). Positive Sturm, T., & Gigerenzer, G. (2006). mental economics. Amsterdam: Enzyklopädie der Psychologie: and negative recency effects in How can we use the distinction Elsevier/North Holland. Vol. D, VII, 1. Allgemeine Musik- retirement savings decisions. between discovery and justifica- psychologie (pp. 657–715). Journal of Experimental Psychol- tion? On the weaknesses of the Todd, P. M., & Gigerenzer, G. Göttingen: Hogrefe. ogy: Applied, 12, 233–250. strong programme in the sociol- (2007). Mechanisms of ecological ogy of science. In J. Schickore & rationality: Heuristics and envi- Volz, K. G., Schooler, L. J., Rieskamp, J., Busemeyer, J. R., & F. Steinle (Eds.), Revisiting discov- ronments that make us smart. In Schubotz, R., Raab, M., Mellers, B. A. (2006). Extending ery and justification: Historical R. I. M. Dunbar & L. Barrett (Eds.), Gigerenzer, G., & von Crammon, the bounds of rationality: Evi- and philosophical perspectives on The Oxford handbook of evolu- D. Y. (2006). Why you think Milan dence and theories of preferential the context distinction (pp. 133– tionary psychology (pp. 197– is larger than Modena: Neural choice. Journal of Economic 158). Dordrecht: Springer 210). Oxford: Oxford University correlates of the recognition Literature, XLIV, 631–661. (Archimedes 14). Press. heuristic. Journal of Cognitive Neuroscience, 18, 1924–1936. Rieskamp, J., Hertwig, R., & Todd, Takezawa, M., Gummerum, M., Todd, P. M., Hertwig, R., & P. M. (2006). Bounded rationality: & Keller, M. (2006). A stage for Hoffrage, U. (2005). Evolutionary Wallin, A. (2007). Explanation and Two interpretations from psychol- the rational tail of the emotional cognitive psychology. In D. M. environment: The case of psy- ogy. In M. Altman (Ed.), Handbook dog: Roles of moral reasoning in Buss (Ed.), The handbook of evo- chology. In J. Persson & P. Ylikoski of contemporary behavioral eco- group decision making. Journal of lutionary psychology (pp. 776– (Eds.), Rethinking explanation nomics: Foundations and develop- Economic Psychology, 27, 802). Hoboken, NJ: Wiley. (pp. 163–176). Berlin: Springer. ments (pp. 218–236). Armonk, NY: 117–139. Sharpe. Todd, P. M., & Heuvelink, A. Wilke, A. (2005). [Review of the Todd, P. M. (2007a). Coevolved (2006). Shaping social environ- book Human evolution: An illus- Rieskamp, J., & Otto, P. E. cognitive mechanisms in mate ments with simple recognition trated introduction]. Human (2006). SSL: A theory of how search: Making decisions in a heuristics. In P. Carruthers, S. Ethology Bulletin, 20 (3), 6–8. people learn to select strategies. decisionshaped world. In J. P. Laurence, & S. Stich (Eds.) The Journal of Experimental Psychol- Forgas, M. Haselton, & W. von innate mind: Vol. 2. Culture and Wilke, A., Hutchinson, J. M. ogy: General, 135, 207–236 Hippel (Eds.), Evolution and the cognition (pp. 165–180). Oxford: C., & Todd, P. M. (2005). Testing social mind: Evolutionary psy- Oxford University Press. simple rules for human foraging Rieskamp, J., & Todd, P. M. chology and social cognition. in patchy environments. In K. (2006). The evolution of coopera- New York: Psychology Press. Todd, P. M., & Miranda, E. R. Forbus, D. Gentner, & T. Regier tive strategies for asymmetric (2006). Putting some (artificial) (Eds.), Proceedings of the 26th social interactions. Theory and Todd, P. M. (2007b). How much life into models of musical Annual Conference of the Cog- Decision, 60, 69–111. information do we need? Euro- creativity. In I. Deliège & G. A. nitive Science Society (p. 1656). pean Journal of Operational Wiggins (Eds.), Musical creativity: Mahwah, NJ: Erlbaum. Schlegelberger, B., & Hoffrage, U. Research, 177, 1317–1332. Multidisciplinary research in (2005). Implikationen der gene- theory and practice (pp. 376– Wilke, A., Hutchinson, J. M. tischen Beratung bei Hochrisiko- Todd, P. M., Billari, F. C., & 395). Hove: Psychology Press. C., Todd, P. M., & Kruger, D. J. Familien für erblichen Brust- und Simao, J. (2005). Aggregate age- (2006). Is risk taking used as a Eierstockkrebs. In A. Gerhardus, at-marriage patterns from indi- Todd, P. M., Rieskamp, J., & cue in mate choice? Evolution- H. Schleberger, B. Schlegelberger, vidual mate-search heuristics. Gigerenzer, G. (in press). Social ary Psychology, 4, 367–393. & F. W. Schwartz (Eds.), BRCA: Demography, 42, 559–574. heuristics. In C. R. Plott & V. L. Erblicher Brust- und Eierstock- Smith (Eds.), Handbook of results Wilke, A., Todd, P. M., krebs. Beratung – Testverfah- Todd, P. M., & Dieckmann, A. in experimental economics. Hutchinson, J. M. C., & Mata, ren – Kosten (pp. 33–58). Heidel- (2005). Heuristics for ordering cue Amsterdam: Elsevier/North R. (2006). Das adaptive Problem berg: Springer. search in decision making. In L. K. Holland. der Ressourcensuche. In H. Saul, Y. Weiss, & L. Bottou (Eds.), Hecht, S. Berti, G. Meinhardt, & Schooler, L. J., & Hertwig, R. Advances in neural information Todd, P. M., & Schooler, L. J. M. Gamer (Eds.), Beiträge zur (2005). How forgetting Aids processing systems (Vol. 17, (in press). From disintegrated 48. Tagung experimentell arbei- heuristic inference. Psychological pp. 1393–1400). Cambridge, MA: architectures of cogniton to an tender Psychologen (p. 317). Review, 112, 610–628. MIT Press. integrated heuristic toolbox. In Lengerich: Pabst Science Pub- W. D. Gray (Ed.), Integrated lishers. Schooler, L. J., & Shiffrin, R. M. Todd, P. M., Gigerenzer, G., & modles of cognitive systems. (2005). Efficiently measuring the ABC Research Group (in Oxford: Oxford University Press. Zhu, L., & Gigerenzer, G. (2006). recognition performance with press). Ecological rationality: In- Children can solve Bayesian prob- sparse data. Behavior Research telligence in the world. Oxford: Vitouch, O. (2005a). Absolutes lems: The role of representation in Methods, 37, 3–10. Oxford University Press. Gehör. In N. Birbaumer, et al. mental computation. Cognition, (Series Eds.), T. H. Stoffer & R. 98, 287–308. Schwenk, G., & Reimer, T. (in Todd, P. M., & Gigerenzer, G. Oerter (Vol. Eds.), Enzyklopädie press). Social influence and (in press-a). Environments that der Psychologie: Vol. D, VII, 1.

64 | Center for Adaptive Behavior and Cognition 06_ABC_030_065.qxd 19.07.2007 15:19 Uhr Seite 65

Center for Adaptive Behavior and Cognition | 65 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 66 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 67

Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 68

Contents

Introductory Overview ...... 69

Research Area I Opportunity Structures of School and Individual Development in Adolescence and Young Adulthood ...... 71

Research Area II Transitions in the Educational System ...... 81

Research Area III Language Skills and Reading Literacy ...... 90

Research Area IV Cognitive Activation in the Classroom ...... 100

Projects of W. Edelstein, Director Emeritus ...... 116

Publications 2005–2006 ...... 118

Scientific Staff 2005–2006

Cordula Artelt (as of December 2005: University of Bamberg), Jürgen Baumert, Martin Brunner (as of November 2006: Université du Luxembourg), Yvonne Grabbe, Ilonca Hardy, Helmut Köhler (as of May 2005 retired), Stefan Krauss (as of April 2007: University of Kassel), Mareike Kunter, Oliver Lüdtke, Kai Maaz, Nele McElvany, Gabriel Nagy, Detlef Oesterreich (as of November 2006 retired), Gundel Schümer (as of June 2006 retired), Petra Stanat (as of October 2005: University of Erlangen-Nuremberg), Elsbeth Stern (as of October 2006: Swiss Federal Institute of Technology, ETH, Zurich), Ulrich Trautwein, Rainer Watermann (as of April 2005: University of Göttingen)

Predoctoral Research Fellows Michael Becker (LIFE), Thamar Dubberke, Cornelia Gresch (LIFE), Axinja Hachfeld, Nicole Husemann, Kathrin Jonkmann (LIFE), Uta Klusmann, Michaela Köller, Andrea G. Müller (LIFE) (as of January 2007: German Youth Institute, DJI, Munich), Marko Neumann, Swantje Pieper, Dirk Richter, Michael Schneider (as of October 2006: Swiss Federal Institute of Technology, ETH, Zurich), Yi-Miau Tsai (LIFE)

Visiting Scientists Gayle Christensen (Fellow, Alexander von Humboldt Foundation) Erin Furtak (Fellow, Alexander von Humboldt Foundation; “Bundeskanzler-Stipendium”)

68 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 69

Introductory Overview

The specific concern of the Center for Educational Research is the study of development and learning from the perspective of institutionalized education. Educational settings such as schools provide a specific structure of opportunities and constraints for learning and development. This structure offers a variety of developmental opportunities, but, at the same time, ex- cludes others. How do aspects of schooling affect learning within and across subject domains, how do they impact the intra- and interindividual differentiation of personality traits and guide career-forming processes? How strongly do students themselves actively influence their own academic development, for example, by selecting or changing learning environments? Which roles do students’ familial backgrounds play in student development as well as in the selection of learning environments and the optimization of academic outcomes? These and other questions are explored by a multidisciplinary team including educational scientists, psychologists, mathematicians, and sociologists.

Conceptual Orientation: Knowledge researchers to consider several context levels, Key References Acquisition and Psychosocial Development such as countries, schools, classrooms, and Baumert, J., Carstensen, in the Context of Institutional Learning the family. Accordingly, our research is em- C. H., & Siegle, T. (2005). Wirtschaftliche, soziale Settings bedded in a multilevel perspective, both con- und kulturelle Lebensver- Learning in institutional settings is a complex ceptually and methodologically, and ad- hältnisse und regionale and multidetermined process. It is difficult to dresses these different context levels. Disparitäten des Kompe- tenzerwerbs. In M. determine whether a school career and a stu- (3) It is important to analyze the effects of Prenzel, J. Baumert, W. dent’s learning outcomes can be described as various facets of these learning contexts Blum, R. Lehmann, D. successful. It is even more difficult to identify simultaneously. For this reason, we incorpo- Leutner, M. Neubrand et al. (Eds.), PISA 2003: the causes for success or failure. Simple ex- rate conceptually different aspects, such as Der zweite Vergleich der planatory models relying on a single factor as a the curriculum, the quality of instruction, and Länder in Deutschland – cause for successful or unsuccessful learning the composition of the learning group in our Was wissen und können Jugendliche? (pp. 323– processes are usually insufficient, if not en- research models. 365). Münster: Wax- tirely misleading—despite their popularity with (4) Remaining attuned to the historical time mann. the public, the press, and policymakers. in which learning takes place is crucial be- Baumert, J., Blum, W., & Due to the complexity of learning in institu- cause educational systems as well as society Neubrand, M. (2004). tional contexts, multiple perspectives guide as a whole change. We therefore embed our Drawing the lessons from PISA 2000: Long- our Center’s research program. research in historical analyses and conduct term research implica- (1) The interactive nature of the characteris- studies that are able to document the effects tions. Gaining a better tics of an individual student and of institu- of changes in institutional settings. understanding of the relationship between sys- tionalized learning settings has to be taken (5) The way in which educational institutions tem inputs and learning into account. In all of our research, learners have structured content areas into different outcomes by assessing are perceived as the coproducers of their own academic subjects determines the domain- instructional and learning processes as mediat- development. Special emphasis is placed on specificity of knowledge acquisition. Our ing factors. In D. Lenzen, the question of how cognitive activation and research focuses on domains of knowledge J. Baumert, R. Water- self-regulation can be stimulated and sup- such as mathematics, reading, English as a mann, & U. Trautwein (Eds.), PISA und die Kon- ported by instructional environments. How- foreign language, and sciences. These do- sequenzen für die erzie- ever, the role of individuals as coproducers of mains represent basic cultural tools that are hungswissenschaftliche their own development is not restricted to critical for individual development in modern Forschung (pp.143–157). Wiesbaden: VS Verlag für the constructivist sense of active and idio- societies. Sozialwissenschaften. syncratic acquisition of knowledge; rather, (6) Although the acquisition of knowledge in individuals are also assumed to proactively core domains is the most central variable in select and shape their developmental envi- learning settings, it is not the only aspect of ronment. interest. The work of our Center also includes (2) A comprehensive analysis of institutional students’ motivation, personality, personal opportunities and constraints requires goals, and values as both an outcome of

Center for Educational Research | 69 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 70

Key References institutional learning and a predictor of aca- potential at-risk group as far as reading liter- Trautwein, U., Köller, O., demic success and choices. acy is concerned. The projects in Research Lehmann, R., & Lüdtke, (7) Several methodological approaches con- Area III have been designed to investigate, O. (Eds.). (2007a). Schulleistungen von Abi- tribute to the identification of powerful from theoretical and empirical perspectives, turienten. Regionale, learning environments. In our Center, experi- the functional mechanisms underlying inter- schulformbezogene und ments and intervention studies complement vention programs aiming to foster soziale Disparitäten. Münster: Waxmann. large-scale longitudinal studies. metamemory and learning strategies in the domain of reading comprehension. The stud- Baumert, J., Stanat, P., & Watermann, R. (Eds.). Summary Outline ies are longitudinal or use an experimental or (2006a). Herkunfts- Research Area I focuses on the relationship quasi-experimental design. They include a bedingte Disparitäten im between the opportunity structure of schools longitudinal study on the development of Bildungswesen: Differen- zielle Bildungsprozesse and the optimization of individual develop- reading competence, a study of a training und Probleme der Ver- ment, in terms of cognitive competencies, program targeting metamemory and reading teilungsgerechtigkeit: motivational and social resources, value strategies within family contexts, and an ex- Vertiefende Analysen im Rahmen von PISA 2000. commitment, and successful transitions to amination of the role of phonological aware- Wiesbaden: VS Verlag für university education, vocational training, and ness in the language development of bilin- Sozialwissenschaften. the labor market. Two longitudinal studies gual children. Finally, the Jacobs Summer form the basis for this research program. The Camp Project investigates the role of implicit ongoing, multiwave, multiple-cohort study and explicit language learning, with a partic- Learning Processes, Educational Careers, and ular focus on proficiency in school-related Psychosocial Development in Adolescence academic language. and Young Adulthood (BIJU) was initiated in Research Area IV investigates the processes 1991 with a sample of more than 5,000 of learning and instruction, and examines 7th graders. Data from six measurement how successful learning environments can be points are now available for the main cohort. created. The focus is on learning that occurs The longitudinal Transformation of the Upper in authentic real-life classrooms, particularly Secondary School System and Academic Ca- in mathematics classrooms. The question of Kunter, M., & Baumert, reers (TOSCA) study began in 2002; addi- what actually determines good instructional J. (2006). Linking TIMSS tional measurements took place in 2002 and practice is central to the success of educa- to research on learning 2004. The Homework as Academic Learning tion and the functioning of the educational and instruction: A reanalysis of the German Opportunities (HALO) project uses data from system. In Research Area IV, we address this TIMSS and TIMSS video multiple sources (BIJU, PISA) and establishes question by combining pedagogical concepts data. In S. J. Howie & a link between Research Areas I and IV. of instructional quality with the analysis of T. Plomp (Eds.), Contexts of learning mathematics Research Area II deals with institutional, in- individual and collective processes of knowl- and science: Lessons dividual, and familial factors in their relation edge acquisition in mathematics. The learned from TIMSS to transitions in the education system. This is COACTIV data set has become the most im- (pp. 335–351). London: Routledge. a new research area which emerged from the portant database in this area over the last focusing and elaboration of research ques- two years. Cortina, K. S., Baumert, J., Leschinsky, A., Mayer, tions that were initially addressed primarily With Prof. Dr. Elsbeth Stern’s move to the K. U., & Trommer, L. in the context of the longitudinal studies of Swiss Federal Institute of Technology in (Eds.). (2003). Das Bil- dungswesen in der Research Area I and in the PISA study. A main Zurich at the end of 2006, our Center’s work Bundesrepublik Deutsch- focus of the activities in this Research Area is on the ENTERPRISE project that was part of land. Strukturen und Ent- the Center’s participation in the Trends in In- Research Area IV has been concluded. There- wicklungen im Überblick. Reinbek: Rowohlt. ternational Mathematics and Science Study fore, we give an interim report on this pro- (TIMSS), which has involved the development ject’s outcomes over the past years. of an additional module to examine the transition from elementary to secondary school. The research questions being addressed within Research Area III draw on a key finding of PISA 2000. In Germany, at least 25 % of the upcoming generation represent a

70 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 71

Research Area I The BIJU/TOSCA Opportunity Structures of School and Individual Development in Research Team Adolescence and Young Adulthood Jürgen Baumert The successful development of human beings across the entire life span is dependent both on Ulrich Trautwein their individual characteristics and on external socializers, such as significant others and so- Oliver Lüdtke cial institutions. The social institution of school plays a major role during childhood and ado- Gabriel Nagy lescence, particularly in the domain of academic learning and, more generally, cognitive devel- Kai Maaz opment. Furthermore, schools have an impact on the formation and development of motivation, emotions, attitudes, and other personal characteristics. Major research topics of Research Nicole Husemann Area I include the opportunity structures open to students from different backgrounds, the Michael Becker development of school achievement across the secondary school level, the educational stan- Kathrin Jonkmann dards attained in German upper secondary schools, the comparability of the school-leaving Swantje Pieper qualifications awarded across Germany, and determinants and consequences of different aca- Marko Neumann demic biographies. In addition, we analyze the interrelationship between school context and (predoctoral fellows) expectancy and value beliefs, and focus on self-selection versus socialization processes during the transition from school to work or university. Associated scientists: Olaf Köller, Institute for Educational Progress (IQB), Berlin The BIJU Study—Aims and Data Collection Kai S. Cortina, Uni- BIJU has four guiding components: versity of Michigan, (1) providing institutional and individual baseline data on the integration of the East Ann Arbor and West German educational systems; since 1991; (2) analyzing domain-specific learning as a function of personal resources and institu- In collaboration with tional opportunity structures; Jacquelynne S. (3) analyzing long-term trajectories of psychosocial development in adolescence and Eccles, Michigan young adulthood; Study of Adolescent (4) analyzing ways of coping with the transition from school to vocational training or Life Transitions university. (MSALT), University Data collection began with a survey of the main cohort (longitudinal cohort 1) in the of Michigan, Ann 1991/92 school year (see Figure 1). Data was gathered from 7th graders at three meas- Arbor urement points. The first point of measurement coincided with the transformation of the unitary school system of the former East Germany to the tracked system adopted from West Germany. The fourth wave of data collection was conducted in Spring 1995, when the main cohort students were in the final grade of lower secondary school. The www.biju.mpg.de fifth wave took place in Spring 1997, when participants were either in vocational education or in the academic track of upper secondary level. The sixth wave of data collection was conducted in 2001, and focused on how students had mastered the transition from school to university or from vocational education to the labor market. The sample of school classes comprises some 8,000 students from 212 secondary schools of all types in the states of Berlin, Mecklenburg-West Pomerania, North Rhine- Westphalia, and Saxony-Anhalt. In Spring 1993, the sample was supplemented by a second longitudinal cohort of 1,330 students in the final grade of lower secondary schooling. In order to provide a baseline for an East-West comparison at the end of lower secondary education, a separate cross-sectional study of approximately 1,600 10th graders was also carried out. This study concentrated on issues of political socialization, and the transition to vocational training and working life.

Center for Educational Research | 71 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 72

Key References The Empirical Database Educational Institutions as Developmental Stanat, P., & Lüdtke, O. Given its emphasis on institutional influences Environments: An Overview of Recent (in press). Multilevel on human development, the research con- Research issues in international ducted within Research Area I entails longi- Research in Area I has yielded a multitude of large-scale assessment studies on student per- tudinal multilevel studies that collect data at theoretically and practically significant find- formance. In F. J. R. van the country, state, school, class, and individ- ings in recent years. We first give a brief de Vijver, D. A. van ual levels, cover more than one knowledge overview of several central results before de- Hemert, & Y. H. Poortinga (Eds.), Individ- domain, and allow both intraindividual scribing two domains (the interplay of voca- uals and cultures in mul- change across domains and interindividual tional interests and school performance at tilevel analysis. Hillsdale, differences in patterns of intraindividual the transition from school to university and NJ: Erlbaum. change to be investigated. The two flagship the effects of assigning and completing Lüdtke, O., Robitzsch, A., Trautwein, U., & Köller, studies in Research Area I, Learning homework) in more depth. O. (2007). Umgang mit Processes, Educational Careers, and Psycho- fehlenden Werten in der social Development in Adolescence and Young Student Performance and Learning Gains psychologischen For- schung: Probleme und Adulthood (BIJU; see Figure 1), and Transfor- One focus of research has been on student Lösungen. Psycholo- mation of the Secondary School System and performance and learning gains. The assump- gische Rundschau, 58, Academic Careers (TOSCA; see Figure 2) were tion that schools and school types represent 103–117. designed to investigate the effects of learn- differential developmental environments Becker, M., Trautwein, U., Lüdtke, O., Cortina, ing contexts in high school and college envi- leading to different student learning gains at K. S., & Baumert, J. ronments on human development, bearing in different school types (fan-spread) is a main (2006). Bildungsexpan- mind the requirements of multilevel longitu- hypothesis of our Research Center. sion und kognitive Mo- bilisierung. In A. Hadjar dinal designs. In addition to BIJU and TOSCA, There are three main explanatory approaches & R. Becker (Eds.), Die research conducted within Research Area I to these fan-spread effects (cf. Baumert, Bildungsexpansion: Er- makes use of the PISA and TIMSS data sets as Stanat, & Watermann, 2006a, 2000b). The wartete und unerwartete Folgen (pp. 63–89). well as additional data sets collected at our first explanation assumes that differential de- Wiesbaden: VS Verlag für Center. Because the longitudinal modeling of velopmental trajectories in German secondary Sozialwissenschaften. hierarchically structured data is methodolog- schools—vocational track Hauptschule, inter- Lüdtke, O., Trautwein, ically difficult, data analysis requires specific mediate track Realschule, and academic track U., Kunter, M., & and complex methods (cf. Lüdtke & Köller, Gymnasium—are attributable to differences in Baumert, J. (2006b). Re- liability and agreement 2006; Lüdtke, Robitzsch, Trautwein, & Köller, students’ performance and learning speed of student ratings of the 2007). that existed before they entered secondary classroom environment: A reanalysis of TIMSS data. Learning Environ- ments Research, 9, 215–230. 199919981997199619951994199319921991 2000 2001 Baumert, J., Trautwein, U., & Artelt, C. (2003). Grade 7 1 2 3 Schulumwelten. Institu- L 1: Longitudinal cohort 1 tionelle Bedingungen des Lehrens und Lernens. In Grade 8 L 2: Longitudinal cohort 2 J. Baumert, C. Artelt, E. C 1: Cross-sectional cohort 1 L 1 Klieme, M. Neubrand, M. Grade 9 Prenzel, U. Schiefele, W. Schneider, K.-J. Tillmann, 1 4 & M. Weiß (Eds.), PISA Grade 10 1 2000. Ein differenzierter C 1 Blick auf die Länder der Grade 11/vocational L 2 Bundesrepublik Deutsch- education land (pp. 261–331). Opladen: Leske + Grade 12/vocational 2 5 Budrich. education Grade 13/vocational 5a education

University/vocational 3 Figure 1. Study design of education/job 6 the BIJU project.

72 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 73

school (cf. Schneider & Stefanek, 2004). Dif- explanation for the effects of school types on Key References fering developmental trajectories would thus student learning gains refers to composition Neumann, M., Schnyder, be an expression of differential learning rates effects arising from differing performance-re- I., Trautwein, U., Niggli, A., Lüdtke, O., & and, therefore, a function of entrance selec- lated, social, cultural, and learning career Cathomas, R. (in press). tivity to the three school types. The second groupings of students. According to this ap- Schulformen als diffe- approach focuses on the differential effects of proach, differences in performance develop- renzielle Lernmilieus: Institutionelle und kom- school types and school systems relying on ment are not, or are only partially, dependent positionelle Effekte auf differing timetables, curricula, and teaching on attending a certain school type. Rather, die Leistungsentwicklung cultures; these effects are “institutional“ in they are a consequence of the characteristics im Fach Französisch. Zeitschrift für Erzie- nature (cf. Baumert et al., 2003). The third of the specific learning group. hungswissenschaft. Becker, M., Lüdtke, O., Trautwein, U., & Data Collection in TOSCA Baumert, J. (2006). Leis- At Time 1 of the TOSCA 2002 cohort, a representative sample of 4,730 students in their tungszuwachs in Mathe- matik: Evidenz für einen last year of upper secondary education (aged about 17 to 19 years) were sampled be- Schereneffekt im mehr- tween March and May 2002. All students were attending either traditional Gymnasium gliedrigen Schulsystem? schools or one of the five (now six) forms of vocational Gymnasium schools that have Zeitschrift für Pädago- gische Psychologie, 20, been established in the state of Baden-Wuerttemberg. More than 60 % of these stu- 233–242. dents consented to be recontacted for follow-up studies. The second wave of data col- Baumert, J., Stanat, P., lection took place from February to May 2004. A total of 2,315 students participated in & Watermann, R. this follow-up. The third wave took place from February to May 2006 and a total of (2006b). Schulstruktur und die Enstehung dif- 1,912 students participated. At Time 4, in early 2007, a subsample of TOSCA partici- ferenzieller Lern- und pants were administered a set of mathematics and cognitive ability tests. Entwicklungsmilieus. In A second TOSCA cohort (“TOSCA-Repeat“) began in 2006 with almost 5,000 students J. Baumert, P. Stanat, & R. Watermann (Eds.), who were in their last year of upper secondary education at more than 150 schools in Herkunftsbedingte Dis- Baden-Wuerttemberg. TOSCA-Repeat assesses the effects of the structural reform of paritäten im Bildungs- the upper secondary level (the last 3 years of schooling before the Abitur). The school wesen: Differenzielle Bildungsprozesse und system in Baden-Wuerttemberg has experienced major changes since 2002. The most Probleme der Vertei- important change to the upper secondary level has been the introduction of core com- lungsgerechtigkeit: Ver- petence subjects and the abolition of the traditional advanced courses (Leistungskurse). tiefende Analysen im Rahmen von PISA 2000 In 2005, instruments from the TOSCA study were administered to students in their final (pp. 95–188). Wies- Abitur year participating in the LAU study on achievement at school entry and learning baden: VS Verlag für development. This is a longitudinal study that has already been running for several Sozialwissenschaften. years (directed by Prof. Dr. Rainer Lehmann from Humboldt University Berlin), in which all of the students of a Hamburg cohort were assessed numerous times throughout their entire school careers. By using the TOSCA instruments in 2005, a detailed comparison of Abitur students in Baden-Wuerttemberg and Hamburg was possible for the first time.

2002 2003 2004 2005 2006 2007 2008

Grade 13 1 1 1

Vocational education/university/job

Vocational education/university/job 2 2

Vocational education/university/job TOSCA-Repeat Vocational education/university/job TOSCA-2002 3

Vocational education/university/job LAU 4 Figure 2. Study design of the TOSCA project.

Center for Educational Research | 73 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 74

Key References Several recent articles from our Center indi- Wuerttemberg performed significantly better Trautwein, U., Köller, O., cate the importance of institutional and than their peers in Hamburg. These findings Lehmann, R., & Lüdtke, composition effects. Becker, Lüdtke, have led to a renewed discussion of the qual- O. (Eds.). (2007a). Schul- leistungen von Abiturien- Trautwein, and Baumert (2006) were able to ity of the Abitur in Germany. ten. Regionale, schul- demonstrate the fan-spread effect on the formbezogene und basis of TIMSS: Within a school year, Gymna- Expectancy Beliefs, Value Beliefs, and soziale Disparitäten. Münster: Waxmann. sium students learned more mathematics Academic Choices than Realschule or Hauptschule students. Beyond learning gains and performance stan- Baumert, Stanat, and Watermann (2006a, dards, the examination of the relationship 2000b) drew on data from the PISA study to between performance and motivational vari- investigate students’ mean performance and ables (specifically expectancy and value be- socioeconomic background characteristics liefs) and the explanation of academic within school types. In addition to school- choices form a second focus of attention type effects, the authors found evidence of within Research Area I. Again, we are partic- composition effects: When controlling for ularly interested in effects of different learn- individual background characteristics, a con- ing environments as differential developmen- centration of students with low average cog- tal environments. We examine how school nitive abilities or from disadvantaged family systems, school types, and the composition of Trautwein, U., Lüdtke, backgrounds had a negative effect on school classes influence the development of O., Köller, O., & Baumert, achievement. noncognitive measures and student biogra- J. (2006). Self-esteem, A longitudinal study with students from two phies. academic self-concept, and achievement: How Swiss cantons learning French as a foreign Different school types can lead to different the learning environment language revealed similar effects (Neumann choice mechanisms. Nagy et al. (in press) moderates the dynamics et al., in press): On the one hand, students in compared the determinants of the selection of self-concept. Journal of Personality and Social higher school types were shown to have par- of advanced courses in Germany and the Psychology, 90, 334–349. ticularly positive learning development; on USA. In order to increase the chance of ob- the other hand, positive effects of favorable taining a place at a renowned university in class composition were observed. However, the USA, participation in a large number of composition effects decreased or became advanced courses is recommended. In insignificant as soon as the school type was Germany, the number of advanced courses a taken into account. student can take is restricted, and the choice Our studies on standards of performance and of a certain course does not influence univer- the comparability of school-leaving certifi- sity entrance. As expected by the authors, the cates have been continued over the past two selection of advanced courses was especially years (Köller, Watermann, Trautwein, & dependent on students’ school performance Lüdtke, 2004). In Germany, only those stu- in the USA, but on students’ domain-specific dents who achieve the Abitur are eligible to self-concept and interest in Germany. study at universities. Therefore, the Abitur is Trautwein, Lüdtke, Köller, and Baumert a highly desirable school-leaving certificate (2006) also identified effects of different that requires high standards of comparability. school systems when comparing students Together with Rainer Lehmann and Olaf from West and East Germany using the first Köller from the Humboldt University Berlin, waves of the BIJU study. The development of we compared the mathematics, English, and global self-esteem of students in East science performance of Abitur students from Germany was more strongly influenced by different school types in the federal states of academic self-concepts than that of stu- Hamburg (LAU) and Baden-Wuerttemberg dents in West Germany. The reasons for (TOSCA) (Trautwein, Köller, Lehmann, & these differences are probably related to the Lüdtke, 2007a). In mathematics, in particular, different learning cultures existing just after large differences were observed between reunification that were also reflected in the school types within the two states. In addi- transparency of performance assessment in tion, on average, students from Baden- East Germany.

74 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 75

In a series of studies we continued our re- central part of the student’s life, strongly af- Key References search program (see Köller, 2004; Trautwein, fecting his or her cognitive activities, beliefs, Marsh, H. W., Trautwein, Köller, Lüdtke, & Baumert, 2005) examining and behavior outside of the educational in- U., Lüdtke, O., Köller, O., frame-of-reference effects on self-concept. stitution. Educational institutions take up a & Baumert, J. (in press). Big Fish Little Pond Ef- Herbert Marsh has coined the term Big-Fish- large part of children’s, adolescents’, and fect: Persistent negative Little-Pond Effect (BFLPE) to describe the young adults’ time, and thus constitute social effects of selective high finding that students in high-achieving environments in their own right. The social schools on self-concept after graduation. Ameri- groups develop lower self-concepts than rules and regulations of educational institu- can Educational Re- equally proficient students in low-achieving tions not only create the conditions for sys- search Journal. environments. Trautwein, Lüdtke, Marsh, tematic instruction and learning but also Trautwein, U., Lüdtke, Köller, and Baumert (2006) documented the provide the setting for immediate everyday O., Marsh, H. W., Köller, O., & Baumert, J. (2006). powerful role of the BFLPE in the large stu- experiences. Tracking, grading, and dent sample from the German extension to Two current dissertation projects within Re- student motivation: Us- the PISA study. They found strong frame-of- search Area I investigate the effects of school ing group composition and status to predict reference effects on both mathematics self- or university contexts on social behavior, life self-concept and interest concept and mathematics interest. Moreover, goals, and well-being. The dissertation proj- in ninth-grade mathe- the study also showed that teachers’ grading ect by Nicole Husemann sheds light on the matics. Journal of Educa- tional Psychology, 98, practices reflect frame-of-reference effects, development of life goal priorities in different 788–806. and that grading standards that vary from environments and effects of person-environ- Lüdtke, O., Köller, O., school to school may partly explain the ment fit in terms of goal congruence on well- Marsh, H. W., & frame-of-reference effects on self-concept being. The dissertation project by LIFE fellow Trautwein, U. (2005). Teacher frame of refer- and interest. However, although grading Kathrin Jonkmann focuses on the relationship ence and the big-fish- practices may contribute to the BFLPE, they between peer status, achievement, and self- little-pond effect. Con- do not fully explain this phenomenon. concept. temporary Educational Psychology, 30, 263–285. Lüdtke, Köller, Marsh, and Trautwein (2005) examined whether the BFLPE decreases if Interests and Achievement, and the teachers use an individualized frame of refer- Transition From School to University ence instead of a social frame of reference. In In comparison to other countries, the cou- accordance with their hypothesis, they found pling between education and occupational that an individualized frame of reference (as status is particularly strong in Germany, identified in a video study) was indeed asso- across the entire breadth of the professional ciated with a more positive self-concept, but hierarchy. Thus, the selection of a field of ter- that the frame-of-reference effects were tiary education is highly important because largely unaffected. the ensuing occupational specialization Frame-of-reference effects are pervasive and shapes long stretches of each individual life stable, even after educational transitions. path. What defines a good selection? What is Marsh, Trautwein, Lüdtke, Köller, and regarded as a successful transition to tertiary Baumert (in press) found persistent frame-of- education? reference effects from secondary school on Psychological theories relate the success of academic self-concept in young adults 2 and the transition from school to tertiary educa- 3 years after secondary school graduation. tion to the concept of person-environment fit (Holland, 1997). They hypothesize two as- Schooling as an Authentic Part of Life pects of person-environment fit to be central It is a structural paradox of formal education indicators of a transition’s success: (1) the fit that the experiences made available within achieved between individual desires and the institutions of formalized education are opportunities to realize them and (2) the fit always vicarious—selected and prepared with achieved between individual competencies the aim of facilitating learning processes and the environmental requirements. A good which the learner must nevertheless perceive fit secures high levels of individual perform- as personal and authentic. At the same time, ance and satisfaction in the context of terti- however, the school or university context is a ary education.

Center for Educational Research | 75 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 76

Figure 3. The relationship domain of interest by between interest profiles, investing their basic basic cognitive abilities, and performance in cognitive abilities in mathematics and Artistic that domain. PPIK English. theory assumes this Investigative mechanism to ex- Cognitive ability plain the high corre- Social θ = 0°; γ = .38** English lations between do- θ ° γ = 151 ; = .24** Realistic main-specific com- Mathematics petencies and basic θ ° γ = 343 ; = .28** cognitive abilities. The predictions of PPIK theory were Enterprising tested using the Conventional TOSCA data set. Ex- amining whether specific competen- Key Reference Former LIFE fellow Gabriel Nagy’s completed cies are associated with vocational interest Nagy, G. (2005). Beruf- dissertation project examined person- profiles requires a series of complex analysis liche Interessen, kogni- environment fit during the transition from steps. The basic principle was to disentangle tive und fachgebundene the Gymnasium to university using data from persons’ general basic cognitive abilities from Kompetenzen: Ihre Be- deutung für die Studien- the TOSCA study (Nagy, 2005). The role of in- their domain-specific strengths and weak- fachwahl und die Be- terests and competencies in the selection of nesses. währung im Studium. a university degree program, along with sub- The analyses showed clear associations be- Unpublished doctoral thesis, Free University sequent success in the university context, tween a person’s basic cognitive ability pro- Berlin. were at the center of attention. This project file and his or her interests: Those with par- drew on the concept of vocational interests ticularly high basic cognitive abilities or developed by John Holland (1997), which mathematics ability usually had an interest identifies six broad content domains of voca- focus within domains I and R. Specific com- tional interests: (1) realistic (R), (2) investiga- petencies in the verbal domain were related tive (I), (3) artistic (A), (4) social (S), (5) enter- to a profile orientation toward domains A prising (E), and (6) conventional (C) interests. and S. Thus, specific mathematical and verbal It is not the absolute level of interests, but competencies were related to contrasting in- the individual configuration of occupational terest orientations. The results of these com- interests that is important: Strong interests plex analyses are shown graphically in in a certain domain are usually accompanied Figure 3. by weak interests in other domains. Prediction of University Field of Study Development of Interests and Competencies Which characteristics are more important for Various theories suggest that domain-specific the subject taken at university: interests or competencies and vocational interests de- competencies? This question was also exam- velop in a reciprocal process. The process of ined with the TOSCA data set. In this case, ability and interest differentiation is most the analysis strategy focused on using stu- clearly explicated in Ackerman’s (1996) so- dents’ knowledge and interests to predict called PPIK theory, which examines the influ- their field of study. The relationship between ences of intelligence-as-Process, Personality, correct and incorrect predictions was ex- Interests, and intelligence-as-Knowledge. Ac- pressed as the “hit rate.“ Figure 4 shows the cording to this theory, persons’ interest leads hit rates found by Nagy (2005) on the basics them to engage often and intensely with the of ability measures or interest profiles alone topic and content in question. In this way, as well as for the two aspects combined plus they develop increased competencies in the a number of additional predictors. Compari-

76 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 77

Figure 4. Hit rates from hierarchical multinomial Model 1 Model 2 Model 3 logistic regressions pre- • Ability measures • Interest profiles • Ability measures dicting the selection of (GPA, cognitive ability, (level, orientation, (GPA, cognitive ability, university courses based mathematics and differentiation) mathematics and on average ability and English achievement) English achievement) interest profiles. • Interest profiles (level, orientation, differentiation) • School type • Gender

61.7% 71.7% 73.3%

son of Models 2 and 3 shows that the hit demands and characteristics. Moreover, rate of interest alone was barely improved by choices are often motivated by factors such the inclusion of other predictors. This finding as job market opportunities or chance experi- demonstrates the central importance of indi- ence rather than interest alone. These factors vidual interest configurations for the selec- can lead to a decrease in the fit between the tion of a university field of study. individual and the subject chosen. What influences do individual interest con- University Success figurations, on the one hand, and their fit to The selection of a university field of study is the university field of study, on the other, often based on mistaken assessments of its have for various aspects of success at univer-

Figure 5. Variance explained by different .20 aspects of success in .18 hierarchical regression Achievement analyses. )

2 .16 .14 .12 .10 Drop-out .08 intentions .06 Study

Percent explained variance ( R Percent .04 satisfaction .02 .00 Baseline Competencies Interests Interest congruence

Center for Educational Research | 77 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 78

sity? How important are cognitive competen- programs remain the exception to the rule. In cies and those acquired at school? A theoret- fact, the recommendations that scientists ical framework model dealing with these and practitioners make about homework as- questions was established by Lubinski and signment and completion are often based on Benbow (2000). According to this model, the their own experience or on speculation rather performance aspect of success at university than on sound scientific evidence (Trautwein depends mainly on the availability of the & Köller, 2003). necessary abilities, whereas satisfaction with Originating from analyses conducted within the university course depends on interest the BIJU study, the learning opportunities congruence. provided by homework assignments and The relationships between interests, compe- completion have become an additional focus tencies, and success at university were exam- of Research Area I that forges a promising ined using data from the second wave of link to Research Area IV. The Homework as TOSCA. Two years after leaving school N = Academic Learning Opportunities (HALO) 1,756 young adults were studying at univer- project draws on different data sources, such sity. The assessment included a broad range as PISA 2000, PISA 2003, and a study con- of aspects of success: (1) self-reported uni- ducted in collaboration with the University of versity course achievement, (2) satisfaction Teacher Education in Freiburg (Switzerland). with university course, and (3) the intention Two questions are at its core: What is the re- to drop out. lation between homework assignments and Figure 5 provides an overview of the central completion and student achievement? What results of a series of sequential regression are determinants of high homework morale? models. The prediction analyses showed that individual competencies (cognitive ability, The Homework-Achievement Relationship mathematics and English achievement, and Based on a review of prior research, Cooper GPA), school grades, vocational interests, and and colleagues (2006) concluded that more interest congruence are associated with suc- time on homework is associated with higher cess at university. Furthermore, the different achievement. Similarly, the OECD report on characteristics examined were related to dif- the PISA 2000 data suggests that the time fering aspects of success at university. Uni- spent on homework is positively associated versity course achievement was mainly deter- with achievement in practically all partici- mined by measures of competency, whereas pating countries. Our own research shows study satisfaction and intention to drop out that definitive insights into the homework- were best predicted by interest congruence. achievement relationship are as yet pre- cluded by a lack of suitable data sets and by Research Team Learning Opportunities Provided by methodological shortcomings in the analyses. Homework There are at least three potential threats to Ulrich Trautwein There are few aspects of student develop- the validity of typical correlational studies on Oliver Lüdtke ment and learning whose practical signifi- the homework-achievement relationship (see Marko Neumann cance for students, teachers, and parents Trautwein, 2007; Trautwein & Köller, 2003). stands in such stark contrast to the scarcity First, homework can be related to achieve- In cooperation with and arbitrariness of theoretical and empirical ment at two levels. One, a homework effect Alois Niggli and research as homework. Students, teachers, at the class level (or homework assignment Inge Schnyder, Uni- and parents invest millions of hours in home- effect) is found when students in classes with versity of Teacher work every week, but the cost-benefit ratio a higher quantity or quality of homework Education, Freiburg, remains a subject of intense debate. The have more pronounced achievement gains Switzerland available body of research findings on home- than students in other classes. The other, a work is complex, fragmented, and contradic- homework effect at the student level (or tory because of the wide variety of interact- homework completion effect), is found when ing factors that may influence the homework students in the same class who differ in their process, and because systematic research homework behavior (e. g., time spent on

78 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 79

homework) show differential outcomes. In value theory, as described in the work of Key References this sense, homework is a classic example of Jacquelynne Eccles, homework behavior is Trautwein, U. (in press). the multilevel problem, and it is essential to believed to be heavily influenced by motiva- Hausaufgaben. In W. Schneider & M. differentiate between teacher- and student- tional predictors (e. g., belief in being able to Hasselhorn (Eds.), Hand- level effects in all studies that relate home- solve homework problems, perceived utility of buch der Pädagogischen work to achievement. A second major issue in homework tasks). The effects of cognitive Psychologie. Göttingen: several homework studies is that they do not abilities and personality as well as the impact Hogrefe. control for the role of confounding variables. of the family context and parental behavior Lüdtke, O., Trautwein, U., Schnyder, I., & Niggli, For instance, more homework might be set in are seen as (partially) mediated by motiva- A. (2007). Simultane high-quality schools attended by students tional predictors. Likewise, effects of the in- Analysen auf Schüler- from privileged backgrounds. Given these re- structional environment (e. g., homework und Klassenebene: Eine Demonstration der kon- lationships, the finding of a positive relation- quality and control) are expected to be par- firmatorischen Mehr- ship between homework and achievement tially mediated by homework motivation. ebenen-Faktorenanalyse might be attributable to a common cause, Several studies provide support for the cen- zur Analyse der Schüler- wahrnehmungen am and not to time on homework per se. In a re- tral assumptions of the model (e. g., Beispiel der Hausauf- lated vein, the majority of studies are single- Trautwein, Lüdtke, Kastens, & Köller, 2006; gabenvergabe. Zeitschrift measurement studies. Questions pertaining Trautwein, Lüdtke, Schnyder, & Niggli, 2006). für Entwicklungspsycho- logie und Pädagogische to the directionality of homework effects For instance, expectancy and value beliefs Psychologie, 39, 1–11. cannot be readily answered on the basis of predict homework effort in different school Trautwein, U. (2007). such designs. Third, research has concen- subjects. In addition, students’ perceptions of The homework-achieve- trated almost exclusively on time spent on homework quality are closely associated with ment relation reconsidered: Differentiating homework. Yet all sorts of distractions can perceived homework value and homework ef- homework time, home- have detrimental effects on students’ home- fort. Moreover, classmates show comparably work frequency, and work behavior. If a student reports spending high agreement in their perceptions of home- homework effort. Learn- ing and Instruction, 17, a lot of time on his or her homework, this is work quality, indicating that these percep- 372–388. not necessarily a sign of great conscientious- tions are not completely idiosyncratic. At the Trautwein, U., Lüdtke, ness, but may reflect problems of motivation same time, we also found substantial associ- O., Kastens, C., & Köller, or concentration. ations between students’ homework behavior O. (2006). Effort on homework in grades 5–9: Our studies (Trautwein, 2007; Trautwein, across domains, and a meaningful predictive Development, motiva- Köller, Schmitz, & Baumert, 2002) indicate effect of students’ conscientiousness. This tional antecedents, and that frequent homework assignments in indicates that students’ homework behavior the association with effort on classwork. Child mathematics are positively associated with is not only dependent on the quality of Development, 77, achievement gains at the class level, but that homework assignments and domain-specific 1094–1111. time-consuming assignments do not show expectancy and value beliefs, but also on Trautwein, U., Lüdtke, positive effects. Focusing on individual stu- stable personality characteristics. O., Schnyder, I., & Niggli, dents, those who put a lot of effort into their In a recent study (Trautwein & Lüdtke, in A. (2006). Predicting homework effort: Sup- homework (but do not necessarily report long press-b), we tested predictions derived from port for a domain-spe- study times) fare better than those who in- the homework model by taking an intraindi- cific, multilevel home- vest less effort. vidual approach. Whereas the interindividual work model. Journal of Educational Psychology, perspective helps to explain why some stu- 98, 438–456. Determinants of Homework Effort dents put more effort into their homework Trautwein, U., & Köller, Trautwein and colleagues (Lüdtke, Trautwein, than others, the intraindividual approach fo- O. (2003). The relation- Schnyder, & Niggli, 2007; Trautwein & Köller, cuses on within-student variability in home- ship between homework and achievement: Still 2003; Trautwein, Lüdtke, Kastens, & Köller, work effort across different subjects. We much of a mystery. Edu- 2006; Trautwein, Lüdtke, Schnyder, & Niggli, tested whether the homework model devel- cational Psychology Re- 2006) proposed a multilevel homework model oped by Trautwein and colleagues can also view, 15, 115–145. as a general framework for studying home- account for intraindividual variation in work effort (see Figure 6). homework effort. Our findings show that According to this model, students’ homework homework effort is primarily a function of effort impacts their achievement. Moreover, between-student differences in conscien- in line with the predictions of expectancy- tiousness, within-student differences in the

Center for Educational Research | 79 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 80

Key References perceived learning environment (subject-spe- Moving beyond studies that rely on self- Trautwein, U., & cific homework quality and control), and report from one source only (typically stu- Lüdtke, O. (2007). Stu- within-student differences in motivation dents), our homework research is currently dents' self-reported ef- (subject-specific expectancy and value be- focusing on what teachers do and think fort and time on homework in six school sub- liefs). Furthermore, a significant cross-level about homework, and the consequences jects: Between-student interaction was found, with perceived home- for students’ homework motivation and differences and within- work control by teachers having a stronger behavior. student variation. Jour- nal of Educational Psy- effect on less conscientious students than on chology, 99, 432–444. their more conscientious peers. Trautwein, U., Köller, O., Schmitz, B., & Baumert, J. (2002). Do homework assignments enhance achievement? A multilevel analysis in 7th grade mathematics. Con- temporary Educational Psychology, 27, 26–50. Figure 6. Condensed version of the homework model (see Trautwein, Lüdtke, Schnyder, & Niggli, 2006).

Homework assignments – homework Homework quality behavior – homework control Homework expectancy Homework effort Student characteristics – gender Achievement – cognitive – school grades abilities – achievement – conscien- tests tiousness Homework value Role of parents Homework – parental time expectations and attitutes – quantity of homework help – quality of homework help

80 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 81

Research Area II The Research Team Transitions in the Educational System Jürgen Baumert The biographies of young people are characterized by a host of transitions. Beyond the biologi- Kai Maaz cal changes and psychological transitions from childhood to adolescence and adulthood that Nele McElvany each individual needs to negotiate, there are several transitions in the context of the educa- Ulrich Trautwein tional system that are governed by specific legal and societal regulations. These necessitate Oliver Lüdtke complex decisions that are not independent of societal and institutional conditions and, in a Gabriel Nagy differentiated educational system, have a far-reaching influence on students’ educational and vocational biographies. The analysis of transitions has a long tradition within the Center for Cornelia Gresch Educational Research; transitions were not only examined in the context of Research Area I (predoctoral fellow) but also in former Research Areas II and III. The new Research Area II integrates all of the Cen- ter’s projects and subprojects that deal explicitly with the analysis of transitions at various Michaela Kropf stages of educational careers, with a focus on familial background. (coordinator)

The importance of educational transitions is Our scientific interest in the analysis of tran- In cooperation with a result of the German educational system’s sitions in educational biographies centers on Rainer Watermann, structural features. two aspects: (1) On the one hand, we focus Department of Edu- (1) Both children and parents are confronted on the transitions themselves. The starting cation, University of with various decisions from the very begin- points are the social disparities in educa- Göttingen ning of the educational career: from the tional participation identified in contexts Wilfried Bos, Insti- choice of kindergarten and elementary such as the PISA studies (Baumert & tute for School school, to the move to a secondary school, Schümer, 2001; Baumert, Stanat, & Water- Development Re- the transfer to upper secondary level or voca- mann, 2006a). These social disparities are as- search (IFS), Univer- tional training, and, finally, to the choice of a sumed to emerge primarily at transitional sity of Dortmund university course. points in the educational career at which de- Olaf Köller, Institute (2) The educational system is characterized cisions must be taken. In this sense, transi- for Educational by a high degree of standardization and tions are analyzed as transition decisions. Progress (IQB), Berlin stratification in terms of the points within an (2) On the other hand, we concentrate on the educational biography at which transitions period before the transition, examining the take place, with specific regulations varying ways in which those involved (usually stu- across federal states. The decisions taken at dents) approach the transition from various various stages of educational trajectories are theoretical perspectives. thus taken in the context of institutional structures and administrative rules and regu- Transitions From Elementary School to lations. Lower Secondary Level (3) Despite various reform endeavors, there is In order to analyze the transition from pri- still a strong link between the educational mary to lower secondary level, a new project track attended and the school-leaving certifi- named TRANSITION (Übergang) was initiated. Key Reference cate acquired. Although this link has become This study examines performance-based dis- Baumert, J., & Schümer, G. (2001). Familiäre weaker over the past 10 years, 80 % of Abitur tributive justice and regional, social, and cul- Lebensverhältnisse, Bil- qualifications, the certificate qualifying for tural or ethnic disparities. Data are currently dungsbeteiligung und entrance to university, are attained at general being collected (see Figure 1 below). Kompetenzerwerb. In J. Baumert, E. Klieme, M. Gymnasium schools. Relative to other coun- Neubrand, M. Prenzel, U. tries, the association between the educa- Description of the Problem and Its Theoretical Schiefele, W. Schneider, tional system and the occupational system is Context P. Stanat, K.-J. Tillmann, & M. Weiß (Eds.), PISA particularly strong in Germany, across the Over the past decades, the German school 2000: Basiskompetenzen entire occupational spectrum. The first job is system has become more permeable and von Schülerinnen und thus largely determined by the educational flexible. This is mainly due to the new routes Schülern im internatio- nalen Vergleich (pp. 323– system (in terms of the qualifications to the Abitur that have been introduced in 407). Opladen: Leske + acquired). the vocational and general sectors of upper Budrich.

Center for Educational Research | 81 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 82

Data collection phases of TRANSITION study

TRANSITION TRANSITION TIMSS TRANSITION TRANSITION Nov/Dec 2006 (after the recommen- May/June 2007 Feb/March 2008 June/July 2008 Survey of dation) Survey of – student assessment First postal survey Second postal parents and parents and teachers – performance test of students survey of students teachers from Feb 2007 – survey of parents

2006 2007 2008

Transition Registration Transition Mid-term Final recommendation with the to report card report card secondary secondary 5th grade 5th grade school school

Institutional framework for transition to the secondary school system

Figure 1. Schedule of the TRANSITION study following children, parents, secondary schooling. Despite these opportu- independent of ability and attainment, and and teachers through the transition from elemen- nities to correct previous decisions, the tran- are called secondary disparities. They violate tary to secondary school. sition from elementary to secondary school the principles of distributive justice. remains one of the most important changes In taking into account primary and secondary of status in young people’s lives, with long- effects of social class, we follow Boudon’s term consequences for their future education microsociological approach, according to and life chances. The allocation of students which educational decisions result from an to different secondary school types highlights internal connection between students’ school differences in performance that developed in performance, selection mechanisms of the elementary school or were already present at educational system, and familial decision Key References school entry. These performance differences processes. The selection mechanisms are de- Maaz, K., Gresch, C., are not independent of the students’ social, termined by transition rules (e. g., whether or Köller, O., & Trautwein, ethnic, and cultural background, however. In not recommendations made by elementary U. (2007). Schullauf- bahnen, soziokulturelle fact, the assignment of students to different schools are binding) as well as by the institu- Merkmale und kognitive secondary tracks on the basis of their tional structure of the educational system. Grundfähigkeiten. In U. achievement is always associated with social, Familial or parental selection behavior is de- Trautwein, O. Köller, R. Lehmann, & O. Lüdtke ethnic, and cultural disparities. These social termined by the individual decisions of stu- (Eds.), Schulleistungen and ethnic differences in educational partici- dents or their parents, who have several op- von Abiturienten: Regio- pation, which are attributable to differences tions. Social-class-specific educational deci- nale, schulformbezogene und soziale Disparitäten in student ability and performance, are called sions are mainly due to differences in (pp. 43–70). Münster: primary disparities. They are congruent with educational aspirations and students’ school Waxmann. the principles of performance-based distribu- achievement. Thus, educational disparities re- Baumert, J., Stanat, P., tive justice—although they may well be con- sult from background-dependent educational & Watermann, R. (Eds.). (2006a). Herkunftsbe- sidered too large from a normative perspec- decisions and social-class-related achieve- dingte Disparitäten im tive. Additionally, however, parents from dif- ment differences. Selection of an educational Bildungswesen: Differen- ferent social and ethnic backgrounds may option can be regarded as the result of a zielle Bildungsprozesse und Probleme der Ver- differ systematically in the secondary track combination of primary and secondary ef- teilungsgerechtigkeit: they choose for their children. These differen- fects of social class. Whereas the primary Vertiefende Analysen im tial choices produce new disparities in edu- effects mainly determine the likelihood of Rahmen von PISA 2000. Wiesbaden: VS Verlag für cational participation at the transition to success of an educational investment via stu- Sozialwissenschaften. secondary schooling. These disparities are dents’ performance, the secondary effects

82 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 83

additionally lead to differing cost-benefit ing up the secondary system in this way also Key References calculations among the different social leads to a reduction of social disparities. Trautwein, U., & classes. We therefore take an expanded However, it remains unclear whether it can Baeriswyl, F. (in press). value-expectation approach that allows soci- compensate for secondary disparities arising Wenn leistungsstarke Klassenkameraden ein ological rational choice theory to be tested at the transition from elementary to second- Nachteil sind: Referenz- against more psychological value-expecta- ary school because little is known about the gruppeneffekte bei Über- tion models. Moreover, the model makes it extent of these early inequalities. Corre- gangsentscheidungen. Zeitschrift für Pädago- possible to include institutional and adminis- sponding knowledge of the other educational gische Psychologie. trative conditions as well as the norms of the paths is even less satisfactory. Baeriswyl, F., Wandeler, respective social and ethnic reference group. C., Trautwein, U., & These have tended to be disregarded theoret- Main Aims of the TRANSITION Study and Its Oswald, K. (2006). Leis- tungstest, Offenheit von ically so far. Design Bildungsgängen und Comparison of the proportions of students Germany’s participation in the TIMSS 2007 obligatorische Beratung from different elementary schools or admin- study of 4th graders has provided an ideal der Eltern: Reduziert das Deutschfreiburger Über- istrative districts enrolling in the different opportunity to considerably extend scien- gangsmodell die Effekte secondary tracks indicates the possible mag- tific knowledge of how parental intentions, des sozialen Hinter- nitude of these secondary disparities. For ex- cultural, social, and economic backgrounds, grunds bei Übergangs- entscheidungen? ample, the proportion of students attending teachers’ recommendations, and institu- Zeitschrift für Erzie- an academic-track Gymnasium can range tional regulations interact at the transition hungswissenschaft, 9, from 10 % to 80 % depending on the district. from elementary to secondary school. The 373–392. Likewise, the probability that a student from main objective of the TRANSITION study is Maaz, K., Hausen, C., McElvany, N., & a certain background will be enrolled in a to analyze parental decisions on the transi- Baumert, J. (2006). mid-track Realschule or an academic-track tion from elementary to secondary school Stichwort: Übergänge im Gymnasium varies considerably from school against the background of the following Bildungssystem. Zeit- schrift für Erziehungs- to school and from region to region, irrespec- factors: wissenschaft, 9, tive of the permeability of the secondary 299–327. sector. – students’ achievement and attitudes at How do these kinds of differences develop? Is elementary school; it possible to reduce them, and, if so, how? – the parental decision-making process as Any attempts to answer these questions a function of the social, ethnic, and cul- highlight just how little we know about the tural background; transition from elementary to secondary – the secondary track recommended by school. We can only speculate that perform- the elementary school teacher; ance differences and secondary social and – the process of parent-teacher consulta- ethnic disparities in educational participation tion; emerge from the interaction of the cultural – institutional regulations; background and social status of the parental – the regional structure of the secondary home, the students’ actual performance, system and the regional provision of teachers’ recommendations, institutional reg- secondary schools; ulations, and not least the cultural, social, – the cultural environment of the school’s and economic environment. Very little is catchment area; and known about the decision-making process – the regional economic and labor market itself, however. structure. Insights into this process are needed if we are to understand the logic and social dynamics of the German school system as The objectives of the study are to find the a whole. For example, we know that the low critical variables regarding both the parental Gymnasium attendance of certain student decision process and the recommendations groups at middle school level can be com- made by elementary school teachers using pensated by establishing alternative college- specifically developed scales, and then to bound paths at upper secondary level. Open- model the interactions of these indicators.

Center for Educational Research | 83 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 84

Key Reference This would replace the frequently applied, Transition From Lower to Upper Secondary Nagy, G., Köller, O., & but unsatisfactory, approach of using a sin- Level Schooling or to Vocational Training Heckhausen, J. (2005). gle question to indicate central dimensions At the end of lower secondary level, students Der Übergang von der Schule in die berufliche of decision and recommendation behavior. and parents are again confronted with a dif- Erstausbildung: Wer die The analysis follows four main lines of ficult decision. Students can either finish Sorgen scheut, wird von research: school and acquire an educational qualifica- ihnen ereilt. Zeitschrift für Entwicklungspsy- tion to move into vocational training or they chologie und Pädago- (1) modeling the decision process in the can continue school by entering upper sec- gische Psychologie, 37, children’s parental home; ondary level education. 156–167. (2) examining the situation of children from immigrant families separately; Transition From Lower Secondary Level to the (3) reconstructing elementary school Vocational Training System teachers’ recommendation behavior The transition from school to vocational and determining their diagnostic com- training is a developmental task whose ac- petence; and complishment is decisive for the ensuing pro- (4) analyzing the importance of differing fessional career. The role of emotions during institutional regulations. the search for an apprenticeship has hardly been examined so far. Based on the Action Phase Model of Developmental Regulation A further objective of the study is to analyze and Control Process Theory, predictions were how the students and their parents cope with drawn up in a study by Nagy, Köller, and the process of transition. Therefore, there are Heckhausen (2005) regarding affective reac- two further phases of data collection after tions during the search for an apprenticeship. the transition to secondary school: The first These were tested in the longitudinal study covers the first 6 months after the transition Developmental Regulation and Mastering the up to the first mid-term report card in 5th Transition from School to Work (DRAMA) with grade. The second phase continues until the N = 363 students in 10th grade of intermedi- final 5th grade report card in the new school. ate-track secondary school. The study is taking place in cooperation be- Structural equation models showed that the tween the Berlin Max Planck Institute for students’ experience in the course of the ap- Human Development and the Department of plication process is related to their negative Education of the Georg August University vocation-specific affect. In contrast to un- Göttingen, with Professor Rainer Watermann specific mood measures, goal-specific affect from Göttingen in charge. mediates between the timing of the first pos- The TRANSITION study is coordinated with itive response and subsequent commitment the design of TIMSS 2007 (cf. Figure 1). The to the application process. After controlling sample is made up of the majority of classes for gender, mathematics achievement, and taking part in TIMSS. The federal states of the social class of the area served by the Berlin, Brandenburg, and Mecklenburg- school, negative affect remains predictive of Western Pomerania are excluded because the the increase of application activity in the transition from elementary to secondary second half of the 10th grade. school takes place after 6th rather than 4th grade in these states. The sample is extended Transition From Lower to Upper Secondary by an additional 26 classes with a higher rate Level Schooling of students from immigrant families. The en- Whereas social disparities have often been tire sample is made up of 253 classes with shown for the transition from elementary 5,819 students. Parents, students, and teach- school, there have been relatively few studies ers will be interviewed. Moreover, institu- discriminating primary and secondary effects tional and regional context factors will be for the transition at the end of lower second- taken into account. ary school. The TIMSS data were used to examine whether the transition to the upper

84 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 85

Figure 2. Logistical regression predicting Gender 0.39 entry to upper secondary level. Age 0.64 Parental education 1.92 Performance in mathematics 1.53 Performance in physics/biology 3.32 Mathematics self-concept 2.33 Physics self-concept 0.72 Key References Interest in chemistry 0.72 Maaz, K. (2006). Soziale Herkunft und Hochschul- Interest in technology 0.5 zugang: Effekte institu- 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 tioneller Öffnung im Bil- dungssystem. Wies- Odds Ratio baden: VS Verlag für Sozialwissenschaften. Chang, P.-H. (2005). secondary level is socially selective. First of Index of Occupational Status (ISEI). Parents of Transformation of voca- all, significant effects of performance in Gymnasium students had the highest ISEI tional secondary schools: mathematics and physics/biology were found. scores and parents of Hauptschule students A study of the vocational Gymnasium in Germany. An increase of one standard deviation in had the lowest. Opladen: Barbara physics/biology performance was associated A comparison of SES means for the 9th and Budrich. with a more than threefold increase in the 13th grades revealed two differences. Parents Maaz, K., Chang, P.-H., chances of entering upper secondary level whose children attended the 13th grade of a & Köller, O. (2004). Führt institutionelle Vielfalt (cf. Figure 2). Moreover, controlling for math- general-education Gymnasium had a higher zur Öffnung im Bil- ematics and physics/biology performance mean SES than those whose children went to dungssystem? Sozialer revealed that cultural capital and parents’ the 9th grade of a Gymnasium. At the same Hintergrund und kognitive Grundfähigkeit der educational qualifications had a significant time the parental SES of students attending Schülerschaft an allge- influence on the transition. For students with vocational Gymnasium schools who went to mein bildenden und similar achievement, the likelihood of enter- a Haupt- or Realschule at lower secondary beruflichen Gymnasien. In O. Köller, R. Water- ing upper secondary track increased if their level was higher than the parental SES of mann, U. Trautwein, & O. parents had high cultural capital. These ef- students who attended a Realschule in the Lüdtke (Eds.), Wege zur fects can be interpreted as secondary dispari- 9th grade. These findings indicate that social Hochschulreife in Baden- Württemberg: TOSCA — ties because the same performances led to selection processes also occur at the transi- Eine Untersuchung an differing transition decisions (Schnabel & tion to upper secondary level. allgemein bildenden und Schwippert, 2000). One must assume that most students with beruflichen Gymnasien (pp. 153–203). Opladen: The TOSCA study showed that students in low achievement do not manage the transi- Leske + Budrich. different secondary tracks differ in terms of tion to upper secondary level. Thus, the Schnabel, K. U., & social background. Students at vocational higher parental ISEI scores of students at- Schwippert, K. (2000). Gymnasium schools have a weaker social tending general-education Gymnasium Einflüsse sozialer und ethnischer Herkunft background than those at higher track Gym- schools could be due to an interaction beim Übergang in die nasium schools (Chang, 2005; Maaz, 2006). between parents’ SES and school perform- Sekundarstufe II und den In order to see whether additional systematic ance. Comparison of the mean ISEI scores in Beruf. In J. Baumert, W. Bos, & R. Lehmann (Eds.), social selection occurs at the transition from relation to the students’ average grades in TIMSS/III: Dritte Interna- lower to upper secondary level, that is, the three main subjects, namely, German, tionale Mathematik- und whether secondary social disparities play a mathematics, and English, confirms this as- Naturwissenschafts- studie: Mathematische role at this transition, TOSCA data were com- sumption. In Figure 3 it is plain that the und naturwissenschaft- pared with PISA data. Figure 3 shows that at- overlap of ISEI distributions among 9th-grade liche Bildung am Ende tendance of certain school types is related to Realschule students with good or very good der Schullaufbahn (Vol. 1, pp. 261–281). the parents’ Socio-Economic Status (SES) as school grades and students of vocational Opladen: Leske + assessed by the International Socio-Economic Gymnasium schools who have transferred Budrich.

Center for Educational Research | 85 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 86

Figure 3. Highest socioeconomic status (ISEI 80 score) in the family by school type in the 9th and 13th grades. 70

30 Mean ISEI (highest score in family)

20 General Vocational Vocational Hauptschule Realschule Gymnasium Realschule Realschule education Gymnasium Gymnasium school grade school grade Gymnasium Transition Transition 1.0–3.0 3.1–6.0 from from Gymnasium Real- or Hauptschule PISA Baden-Wuerttemberg PISA Baden-Wuerttemberg TOSCA 2002, 13th grade 2000, 9th grade 2000, 9th grade

from a Real- or Hauptschule (13th grade) is fluence the decision of successful Realschule greatest. Therefore, vocational Gymnasium students to continue their school education schools attracted students from socially by transferring to a vocational Gymnasium weaker families who did not attend general- and what determines their subsequent suc- education Gymnasium despite good school cess at that school? Particular emphasis will grades. At the same time, the results also be placed on aspects to be measured in an show that Realschule students from families extensive standardized assessment: (a) school with high ISEI scores were more likely to risk competencies in the domains of mathemat- the step to vocational Gymnasium than stu- ics, language, technology, and economics, dents with the same school grades, but from (b) students’ interest profiles, and (c) familial families with lower ISEI scores. Thus, social background. In addition, we will examine the selection also occurs at the transition to vo- extent to which the decision about further cational Gymnasium schools. education is influenced by the regional structure of schools and vocational training on Educational Careers of Students With offer. Intermediate-Level Certificates (Realschule) So far, there are no reliable data in Germany Transition to University to examine the transition from Realschule to The effect of secondary social disparities at the various types of upper secondary level university entrance is mainly studied by school. Therefore, the project entitled Educa- examining the university aspirations of tional Careers of Students with Intermediate- students with Abitur. Based on BIJU data, the Level Certificates: The Role of School Achieve- longitudinal study carried out at the Max ment, Interest Profiles, and Familial Back- Planck Institute for Human Development, ground for the Transition to Vocational Schnabel, Alfeld, Eccles, Köller, and Baumert Gymnasium or Vocational Training was initi- (2002) showed that the intention to enter ated to systematically analyze the transition university increases 1.5-fold if one parent behavior of students graduating from Real- holds the Abitur rather than a lower school schule. Two questions are asked: Which char- certificate. The surprising aspect of this find- acteristics predict successful integration into ing is that the authors controlled for aca- vocational training tracks? Which factors in- demic performance in various domains,

86 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 87

Figure 4. Results of logis- University expectation (mother) tical regression of university intentions by University expectation (father) gender, school perform- Material orientation ance, as well as economic, social, and per- Performance anxiety sonal characteristics. Number of books Classical music Vocational qualification of parents Material goods Performance Gender –0.3 –0.2 –0.1 0 0.1 0.2 0.3 0.4 0.5 Standardized regression coefficient

Key References Maaz, K., & Water- school grades, and domain-specific self- with sociocultural characteristics of the fam- mann, R. (in press). Re- concepts. ily of origin. produktion oder Mobili- tät? Zur Wirkung fami- Similar observations were made when ana- In many studies on the transition to univer- liärer Prozessmerkmale lyzing TIMSS data. The parents’ highest edu- sity, it is not possible to consider different auf die Studienintention cational qualification had a positive effect on facets of social background or to distinguish am Ende der gymnasialen Oberstufe. Zeitschrift university intentions after control of stu- between primary and secondary background für Soziologie der Erzie- dents’ school performance (see Figure 4). effects. Maaz and Watermann (in press) have hung und Sozialisation. However, the effect of the educational milieu explicitly taken up the concept of primary Trautwein, U., Maaz, K., at home clearly decreased after controlling and secondary effects of social background, Lüdtke, O., Nagy, G., for the perception of parents’ university ex- and examine the effect of familial structure Husemann, N., Water- mann, R., et al. (2006). pectations (Schnabel & Gruehn, 2000). This (e. g., SES, educational level) and process Studieren an der Berufs- shows that the university aspirations of characteristics (e. g., familial book ownership, akademie oder an der young adults develop in the context of their cultural participation) on university aspira- Universität, Fachhoch- schule oder Pädago- parents’ hopes which differ by social status tions at the end of the upper secondary level gischen Hochschule: Ein of the family. of school. With TOSCA data, they were able Vergleich des Leistungs- Using the TOSCA data set, it was possible to to show that primary and secondary dispari- stands, familären Hinter- grunds, beruflicher Inte- analyze the transition to university longitudi- ties take effect, even at this late juncture, ressen und der Studien- nally. Maaz (2006) showed that, in the case and also that the exclusive consideration of wahlmotive von of students with Abitur from Baden- familial structure in the context of disparity (künftigen) Studierenden aus Baden-Württemberg. Wuerttemberg, social class only had a small analysis provides an incomplete image of the Zeitschrift für Erzie- effect (see also Watermann & Maaz, 2006b). effects of familial background characteristics. hungswissenschaft, 9, However, there were differences in the type A communality analysis was carried out 393–412. of university selected by students from dif- allowing a separation of the explained vari- Watermann, R., & Maaz, K. (2004). ferent familial backgrounds. Students with ance into predictor-specific and confounded Studierneigung bei Ab- Abitur from a privileged familial background variance components in order to study the solventen allgemein were more likely than their peers with less respective influence of familial process and bildender und beruflicher Gymnasien. In O. Köller, favorable backgrounds to select university structure characteristics. Whereas the predic- R. Watermann, U. (rather than a lower track polytechnic tor-specific variance component symbolizes Trautwein, & O. Lüdtke [Fachhochschule], college of education the proportion of explained variance that can (Eds.), Wege zur Hoch- schulreife in Baden- [Pädagogische Hochschule], or vocational be regarded as representing structure, re- Württemberg: TOSCA — academy [Berufsakademie]; Maaz, 2006; spectively process characteristics, the con- eine Untersuchung an Trautwein, Maaz, Lüdtke et al., 2006; Water- founded variance component represents the allgemein bildenden und beruflichen Gymnasien mann & Maaz, 2004). In addition, the selec- proportion of explained variance resulting (pp. 403–450). Opladen: tion of the degree to be taken is associated from combinations of both characteristics. Leske + Budrich.

Center for Educational Research | 87 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 88

Figure 5 shows the results of the communal- the total variance independently of structure ity analysis. Taken together, the proportion of characteristics. This shows that process variance explained by structure and process characteristics are highly confounded with characteristics is rather low, but there is a structure characteristics, but that they have proportion of specific process-related vari- their own independent effect on university ance that contributes to the explanation of aspirations.

Figure 5. Communality analysis: Separation of the variance explained Structure Process by structure and process characteristics characteristics characteristics in univer- 0.9% 1.2% sity aspirations after controlling for the grade attained in the Abitur.

Confounded Specific variance Specific variance 0.6% variance 0.4% 0.6%

Total variance 1.6%

Figure 6. Percentual agreement of parents’ and students’ responses 100 when asked about par- 90 Mother Father ents’ educational and vocational level. 80 70 60 50 40

Agreement (in %) 30 20 10 0 School-leaving certificate Vocational qualification

88 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 89

Key References Validity of Students’ Responses Kreuter, F., Maaz, K., & Students usually provide the information on their parents’ socioeconomic status (SES) Watermann, R. (2006). Der Zusammenhang and educational level in educational research. Validity of this information thus depends zwischen der Qualität strongly on students’ knowledge of their parents’ educational and vocational achieve- von Schülerangaben zur ments. Even if such information is never free of measurement error, it is expected that sozialen Herkunft und den Schulleistungen. In parents themselves should give more reliable information about their educational and K.-S. Rehberg (Ed.), vocational characteristics than their children, particularly when the children asked are Soziale Ungleichheit, very young. Analyses of PISA data were able to show that, in general, there is good kulturelle Unterschiede (pp. 3465–3478). Frank- agreement between parents’ and students’ responses about central features of their furt a. M.: Campus. social background (Maaz, Kreuter, & Watermann, 2006). The analyses revealed that Maaz, K., Kreuter, F., & responses on general educational qualifications are less error-prone than responses on Watermann, R. (2006). vocational training (see Figure 6). Students were also better able to give responses on Schüler als Informanten? their parents’ profession than on their vocational qualification level. Die Qualität von Schülerangaben zum In further studies, Kreuter, Maaz, and Watermann (2005, 2006) examined the conse- sozialen Hintergrund. In quences of various types of measurement error for analyses of interactions. Collecting J. Baumert, P. Stanat, & data on parents’ educational and vocational characteristics from students is associated R. Watermann (Eds.), Herkunftsbedingte Dis- with cognitive performance that additionally demands a certain degree of abstraction paritäten im Bildungs- from them. The arising measurement error and its correlation with the dependent vari- wesen: Differenzielle able can lead to a systematic distortion of results. For example, the effect of a measure- Bildungsprozesse und Probleme der Vertei- ment error can be shown by bivariate regression of performance in a mathematics test lungsgerechtigkeit: Ver- on various measures of a father’s vocational training. The coefficients were lower when tiefende Analysen im the students’ responses about their parents’ educational background were used. This Rahmen von PISA 2000 (pp. 31–59). Wiesbaden: simplified bivariate regression clarifies the effect of measurement errors and of potential VS Verlag für Sozial- bias. The resulting underestimation is low, but statistically significant. It can be assumed wissenschaften. that the underestimation of the effect of parents’ vocational training can be offset in Kreuter, F., Maaz, K., & multivariate models when several indicators are linked in order to measure social back- Watermann, R. (2005). Messung sozialer Her- ground. kunft in Schulleistungs- untersuchungen: Der Zusammenhang zwischen der Qualität von Schülerangaben und den Leistungen. In U. Engel (Ed.), Bildung und soziale Ungleichheit: Methodo- logische und strukturelle Analysen (pp. 9–35). Bonn: Informations- zentrum Sozialwissen- schaften.

Center for Educational Research | 89 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 90

The Research Team Research Area III Language Skills and Reading Literacy Cordula Artelt (until 11/2005) Introduction and Project Overview Nele McElvany The examination of the development and effective promotion of students’ language skills and Petra Stanat reading literacy using longitudinal and several (quasi-)experimental intervention studies is at (until 10/2005) the heart of Research Area III. This research program was initiated partly in response to the national and international PISA 2000 and 2003 results, which revealed that students’ reading Michael Becker literacy and motivation in Germany is well below the average in international comparison Axinja Hachfeld (Baumert et al., 2001; Schiefele, Artelt, Schneider, & Stanat, 2004). The situation in the lower Andrea G. Müller performance range is particularly problematic: almost 25 % of the 15-year-old students did (predoctoral fellows) not even reach the second of five proficiency levels defined by the OECD. With reading abilities that allow the decoding of letters and words or the retrieval of simple information from a Christina Limbird written text at most, considerable difficulties are to be expected for these students, not only (associated predoc- in school education but also in vocational training and working life. In-depth analyses show toral fellow, LIFE, that familial background plays a crucial role in competence acquisition and educational par- Free University ticipation. In Germany, the coupling of families’ social background and students’ educational Berlin, until success is particularly strong in international comparison. Students with a weak socioeco- 08/2005) nomic background are more often found on the low-proficiency levels. Students from immigrant families, that is, about one fifth of students in Germany, proved to be disadvantaged Gayle Christensen both in proficiency acquisition and in educational participation. The determinants of perform- (visiting research ance development in these children and opportunities for their in- and out-of-school promo- scientist, until tion are a topic of particular interest to this Research Area. 09/2005) The foundation for proficiency acquisition at intervention study on the implementation higher level schools is laid at elementary and efficacy of a newly developed reading school. At the same time, the transition from support program, the Berliner Eltern-Kind elementary to secondary schools is one of the Leseprogramm (Berlin Parent-Child Reading key junctures in the German educational sys- Program), was analyzed in an in-depth evalu- tem that influences an individual’s further ation study. educational and vocational career consider- All three projects focused on reading profi- ably (cf. Research Area II). Therefore, the ciency and included the special situation of projects on the acquisition and promotion of children with a migration background. Three language skills and reading literacy focus on further central projects of the Research Area elementary school. concentrated explicitly on the development Reading proficiency was at the center of at- and promotion of language skills and reading tention for three central projects within this literacy of children with a migration back- Research Area in 2005 and 2006: ground: (1) A wide-ranging and interdisciplinary ex- (1) Phonological Processing, Verbal Abilities, pertise on Promotion of Reading Literacy was and Second Language Literacy Development commissioned by the German Federal Educa- Among Bilingual Turkish Children in Germany: Key Reference tion and Research Ministry (Artelt et al., In this study, central dimensions in second Artelt, C., McElvany, 2005). language acquisition were examined as pre- N., Christmann, N., Richter, T., Groeben, N., (2) The development of reading proficiency requisites for the acquisition of reading liter- Köster, J., Schneider, W., and central individual and social influence acy. Stanat, P., Ostermeier, factors from 3rd to 6th grade were examined (2) Barriers to Reaching Proficiency in School- C., Schiefele, U., Valtin, R., & Ring, K. (2005). in a new longitudinal study, the Berliner Related Language Among Second Language Förderung von Lesekom- Leselängsschnittstudie (Berlin Longitudinal Learners: In this experimental study, system- petenz – Eine Expertise. Reading Study). atically selected language aspects (vocabu- Bonn: Bundesminis- terium für Bildung und (3) In response to research questions on pro- lary, grammar, contextual embeddedness, Forschung (BMBF). motion possibilities, a quasi-experimental content related to everyday life) were empiri-

90 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 91

cally analyzed to see whether those can be ture on the promotion of reading proficiency Key Reference identified that create particular difficulties and on a detailed online survey of all 16 Stanat, P., & for children with migration backgrounds and states in Germany about existing in- and Christensen, G. (2006). that should be taken into account when de- out-of-school support offers. The expertise Where immigrant students succeed: A com- signing promotion concepts. intentionally takes an interdisciplinary per- parative review of per- (3) The Jacobs Summer Camp: In this experi- spective, and was put together under the formance and engage- mental intervention study, a newly developed leadership of the Max Planck Institute for ment in PISA 2003. Paris: OECD. promotion concept was tested to see Human Development (Cordula Artelt, Nele whether it was possible to improve the lan- McElvany, Petra Stanat) with reading re- guage skills of children from immigrant fami- search experts from various universities and lies in the summer vacation. institutions (Prof. U. Christmann, University The research focus on children with a migra- of Heidelberg; Dr. T. Richter and Prof. N. tion background was also intensified in an Groeben, University of Cologne; Prof. J. internationally comparative perspective on Köster, University of Jena; Prof. W. Schneider, the PISA 2003 data. The results of the analy- University of Würzburg; Dr. C. Ostermeier, ses were made available to the scientific Leibniz Institute for Science Education, Kiel; community and the general public in the Prof. U. Schiefele, University of Bielefeld; OECD report entitled: Where immigrant stu- Prof. R. Valtin, Humboldt University Berlin; dents succeed—A comparative review of Prof. K. Ring, Stiftung Lesen, a foundation performance and engagement (Stanat & sponsoring reading). After an extensive Christensen, 2006). overview of the current state of research and of theoretical models available in the domain Reading Proficiency of reading literacy, a scientific perspective Expertise “Promotion of Reading Proficiency“ examines which insights can be gained from Which approaches are there to promote the viewpoints of different disciplines and reading literacy in Germany? How can sup- where measures promoting reading literacy port measures in the German states be seem productive. The review of existing state supplemented in a useful way? These are the activities showed that most projects’ main questions to which Promotion of Reading aim is to promote reading motivation (see Literacy, the expertise commissioned by the Figure 1). Bundesministerium für Bildung und The expertise arrived at three central recom- Forschung, responds. Since being published, it mendations regarding content, processes, and has been distributed widely and has elicited cooperation for further and sustained promo- much interest. The expertise is based both on tion of reading literacy from a statewide per- an extensive viewing of the scientific litera- spective:

Figure 1. Main aims of 35 the projects in German 33 states (single nomina- tions). 30 Promotion of reading motivation 25 Promotion of societal recognition of reading 20 20 Training measures Promotion of reading frequency 15 Teaching of reading techniques

Frequency (in %) and strategies 10 8 Development of materials 7 6 5 Promotion of reading of different 3 text types 11 0 Other

Center for Educational Research | 91 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 92

(1) Emphasizing subcompetencies and pre- In all, 33 elementary school classes in Berlin requisites of reading literacy as goal dimen- took part. The assessment began at the end sions of promotion; of 3rd grade because basic reading tech- (2) placing communication about demands niques have usually been acquired by this and competency expectations regarding stu- time and more complex reading und compre- dents’ reading literacy at the center of pro- hension processes play an increasing role. fessional exchange between teachers and These provide the basis for continued in- and reading promoters; and out-of-school learning and everyday life. The (3) forms of structured cooperation: making development and complex interaction of interfaces and transitions in the educational reading motivation, reading self-concept, system points of departure for reading pro- reading practice, reading-related metacogni- motion and guaranteeing content transfer tive knowledge, vocabulary, and decoding and coordination via structured in- and out- ability were examined. Vocabulary was par- of-school cooperation. ticularly important for reading level and liter- Clear promotion potentials are seen in the acy development. stronger content alignment of promotion The longitudinal design of the reading study measures toward the construction of mean- provides information on development before ing in reading and the reading process, in the and after the tracking in secondary schools. focusing on the needs of individual target This made it possible for Michael Becker to groups, and in the expansion of promotion- estimate, as part of his doctoral thesis, how related collaboration between in- and out- developmental trajectories are influenced by of-school support agencies. the tracked school system. Moreover, in view of the tight coupling between familial back- Berliner Leselängsschnittstudie—Berlin ground and school performance, we examine Longitudinal Reading Study how school developments in reading literacy The ability to read and understand texts is can be related to specific familial factors, not only important in the school context, it is such as cultural capital (e. g., reading to- foremost a crucial precondition for participa- gether, parental attempts to motivate read- tion in social, political, cultural, and eco- ing, or parents as reading models). The analy- nomic life in our society. The results of inter- sis of the use of cognitive and metacognitive national comparison, such as PISA, are all the reading strategies in this age group, the as- more dramatic because they reveal consider- sessment of school support environments by able deficiencies in students’ reading abilities surveying teaching staff on teaching practice in Germany (e. g., Baumert et al., 2001; Bos and reading promotion activities, and the ex- et al., 2003). In the context of a longitudinal amination of the diagnostic competence of study, Nele McElvany examined individual German teachers regarding reading literacy and social factors influencing the develop- of their students are all areas of interest. ment of reading literacy, reading motivation, For the group of children with a familial mi- and reading behavior from 3rd to 6th grade. gration background, the longitudinal data were used to determine whether the develop-

Figure 2. Design of the Berlin Longitudinal End of 3rd grade Beginning of 4th grade Middle of 4th grade End of 6th grade Reading Study. Students: Students: Students: Students: Questionnaire Questionnaire Questionnaire Questionnaire & test booklet & test booklet & test booklet & test booklet Parents: Parents: Parents: Parents: Questionnaire Questionnaire Questionnaire Questionnaire Teachers: Teachers: Teachers: Questionnaire Questionnaire Questionnaire

March 2003

92 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 93

ment of reading competency and important Parent-Child Reading Program) to promote Key Reference reading prerequisites, such as vocabulary, de- reading literacy of students at elementary McElvany, N. (in press). coding abilities, and motivation, is compara- school age systematically within the context Die Bedeutung fami- of their families (McElvany, 2007; McElvany, liärer Merkmale für ble in the time period from 3rd to 6th grade Lesekompetenz, Lese- with that of the group of children without a Artelt, & Holler, 2005). Based on findings on motivation und Lesever- migration background and whether the ex- reading literacy promotion, reading socializa- halten. Psychologie in tion, and familial influence opportunities the Erziehung und Unter- amined factors play the same role for both richt. groups. It became clear that competence dif- program intends to promote the implicit un- ferences favoring children without a migra- derstanding and habituation of text compre- tion background were mainly due to the bet- hension strategies and processes, with the ter social status of these families. This had a long-term goal of reading literacy enhance- significant effect on performance level at the ment. To achieve this, the program makes use end of 3rd grade, but not on further literacy of parent-child interaction structured by development up to the end of 6th grade. For questions and answers about individual texts. both groups, vocabulary had a greater influ- The program consists of 43 parent-child ses- ence on baseline literacy level and subse- sions and lasts 14 to 15 weeks, with 3 ses- quent change than decoding ability. sions per week. Each session takes about 30 minutes. At the beginning of the session, Berlin Parent-Child Reading Program the parent and child decide whether they Although the family is regarded as the most want to work on a longer or shorter text. important out-of-school context for reading These are very varied in topic and kind and socialization of children (e. g., Hurrelmann & aim at the interests of 4th graders. After the Hammer, 1994) and the PISA study in parent or child has read the text out loud, it Germany and other large-scale studies have is first established whether the child has had repeatedly demonstrated the tight link be- trouble understanding the content or individ- tween school performance and familial back- ual words (cf. Figure 3). Simple or more com- ground, pedagogical-psychological interven- plex comprehension questions follow and, on tion approaches in the field of systematic average, four follow-up questions or tasks promotion of reading literacy have mostly re- stimulate an in-depth conversation that is stricted themselves to small groups or the supposed to lead to an elaborated under- class context and have disregarded the pro- standing of text content. The final task usu- motion potential of the family. Within this ally requires an oral summary of the text, but Research Area, Nele McElvany developed the can also be of a creative nature (e. g., draw a Berliner Eltern-Kind Leseprogramm (Berlin picture about the story).

Figure 3. Conception of Text Decoding ability, the reading program Questions vocabulary sessions. K0: Is there anything that you didn’t understand in the text? Or are there words you don’t know? Try to clarify things Metacognitive processes, together if anything is still unclear! Vocabulary

K1: How do you identify a princess according to the story? E1: Whose idea was the pea? Questions ensuring .... comprehension

W1: Did the king and queen choose a good way to find a Elaboration strategies, wife for the prince? Metacognitive processes, ... Vocabulary

Final task for the child: Summary Cognitive strategies, Imagine you meet a friend and want to tell her or him the Motivation story you just read. You haven’t got much time, so you only give a summary. That means you ....

Center for Educational Research | 93 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 94

Key References In contrast to typical homework supervision, in the family. The analyses on implementa- McElvany, N., & Artelt, the parent and child have equal rights tion show that the structures and processes C. (in press-a). Das throughout the session, carrying out all of intended in the program worked well in the Berliner Eltern-Kind Leseprogramm: Konzep- the tasks in alternation or together and participating families but also that many tion und Effekte. Psy- giving each other feedback. In this way, the families were not prepared to take part in chologie in Erziehung program combines approaches of implicit this kind of voluntary supplementary reading und Unterricht. strategy promotion and of “guided oral read- promotion measure in their own home (cf. McElvany, N., & Artelt, ing,“ complementing the role of the parental Figure 4). C. (in press-b). System- atic reading training in model and including scaffolding elements. There were clear differences between partici- the family: Development, The implementation and effectivity of the pants and nonparticipants. Participating chil- implementation, and program were tested in an evaluation study dren were more likely to come from more evaluation of the Berlin Parent-Child Reading (McElvany, 2007; McElvany & Artelt, 2006). educated families, and were relatively good Program. Learning and Between September 2003 and January 2004, readers at baseline. This highlights the need Instruction. 509 4th graders from 15 Berlin elementary for the development of additional strategies McElvany, N. (2007). schools took part, to counteract participation selectivity in Förderung von Lesekom- petenz im Kontext der with the nonpartici- terms of performance-related child charac- Familie. Münster: Wax- pating peers (N = teristics and the social background of fami- mann. 393) representing lies. McElvany, N., & Artelt, the control group in With the Berlin Parent-Child Reading Pro- C. (2006). Das Berliner Eltern-Kind Lesepro- the quasi-experi- gram, the Research Area has provided a gramm. Diskurs Kind- mental pre-post- promising systematic intervention program heits- und Jugend- control group design. supporting concrete reading competencies forschung, 1, 157–159. In the framework of and familial reading culture. McElvany, N., Artelt, C., this evaluation study, & Holler, S. (2005). Lese- förderung in der Familie. the effects of participation on prerequisites Promoting Children With a Migration In A. Sasse & R. Valtin and subcompetencies of reading literacy Background (Eds.), Lesen lehren could be observed. The vocabulary of the par- Phonological Processing, Verbal Abilities, and (pp. 151–167). Berlin: Deutsche Gesellschaft ticipants developed more strongly than that Second Language Literacy Development für Lesen und Schreiben. of the nonparticipants over the time of the Among Bilingual Turkish Children in Germany study. A particular effectivity of the program Focusing on early reading skills, a doctoral for weak students was revealed regarding the thesis by LIFE fellow Christina Limbird used a development of text-related metacognition. longitudinal study to examine whether the However, there was no direct promotion ef- same component processes are involved for fect on standard reading comprehension or both bilingual and monolingual children from decoding abilities. Beyond enhancement of the 1st to the 3rd grade. A sample of 169 concrete competencies, the findings show children (69 monolingual, 100 Turkish- positive effects for familial reading culture. German bilingual) was investigated with re- Further analyses of longer term effects are in gard to nonverbal cognitive abilities, phono- planning. Longer term effects could be due to logical awareness, verbal abilities, short-term habituation of individual reading processes or verbal memory, word decoding, and reading to the general promotion of reading culture comprehension at multiple points of measurement over a 2-year period. International Figure 4. Distribution of research in the realm of literacy acquisition families invited to regis- repeatedly describes these skills as the build- ter for, and participate 13% in, the program. No registration ing blocks of reading. Because both theoreti- Registration, but cal and empirical models of reading point to 21 % no participation phonological awareness as an essential as- 66% Participation pect of emerging reading skills and because a large body of literature indicates that phonological awareness may be influenced by diverse linguistic experiences, this study placed

94 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 95

Figure 6. Model of read- 60 ing tested for Turkish TB bilingual (TB) and 50 GM Vocabulary German monolingual 40 (GM) groups. 30

20 Reading Decoding 10 comprehension

German vocabulary score 0 T1 T2

Figure 5. Number of German words identified by Phonological Turkish bilingual (TB) and German monolingual (GM) awareness groups for German vocabulary in the middle and at the end of 2nd grade.

particular emphasis on bettering our under- monolingual group but also indicated that vocabulary may be more relevant for the standing of the role of phonological aware- Key References ness among the growing number of bilingual bilingual group. Limbird, C. K. (2006). children in Germany. A series of analyses based on a typical model Phonological processing, Theories of bilingualism and empirical data of reading for children (from Näslund & verbal abilities, and sec- led to the prediction that bilingual children Schneider, 1991) showed poorer fit for the ond language literacy development among bilin- would perform better on tasks involving bilingual group. These findings indicate that gual Turkish children in phonological awareness than their mono- the bulk of research on reading is not ac- Germany. Unpublished lingual peers. Monolingual children, on the counting for the unique characteristics of doctoral thesis, Free Uni- versity Berlin. other hand, were expected to outperform multilingual children in primary education, Limbird, C. K., & their bilingual counterparts on measures of and that this deficit clearly needs to be ad- Stanat, P. (2006a). German vocabulary. It was predicted that the dressed in future research. Practical implica- Prädiktoren von Lesever- factors involved in reading performance and tions of this study could include the imple- ständnis bei Kindern deutscher und türkischer reading development would differ between mentation of vocabulary and phonological Herkunftssprache: Ergeb- the groups, with vocabulary playing a more awareness training programs for multilingual nisse einer Längsschnitt- central role for bilingual children and phono- preschoolers and young elementary school studie. In A. Ittel & H. Mertens (Eds.), Verän- logical awareness demonstrating more im- children in Germany. derungsmessung und portance for monolingual children. The com- Längsschnittstudien in mon existing models of reading development Barriers to Reaching Proficiency in School- der empirischen Erzie- hungswissenschaft were expected to function more poorly with Related Language Among Second Language (pp. 93–123). Wies- a bilingual sample than with a monolingual Learners baden: VS Verlag für sample. Despite the assumption that less knowledge Sozialwissenschaften. Results indicated that while the bilingual of German contributes greatly to the differ- Limbird, C. K., & Stanat, P. (2006b). children scored consistently higher on tasks ences in school performance between chil- Sprachförderung bei of phonological awareness, the difference dren with, and without, a migration back- Schülerinnen und was not statistically significant. The data ground and to the increase of programs to Schülern mit Migrations- hintergrund: Ansätze clearly showed though that the bilingual promote language skills, information about und ihre Wirksamkeit. In group performed much more poorly on meas- the barriers preventing acquisition of German J. Baumert, P. Stanat, & ures of German vocabulary (see Figure 5). as a second language is limited. Specifically, R. Watermann (Eds.), Herkunftsbedingte Dis- On measures of word decoding and reading little is known about the aspects of language paritäten im Bildungs- comprehension, no differences between the that create particular difficulties for children wesen: Differenzielle Bil- bilingual and monolingual children emerged. with a non-German language background dungsprozesse und Probleme der Vertei- The two groups were then tested for good- when learning school language. Everyday- lungsgerechtigkeit: Ver- ness-of-fit on the selected model of reading. and school-related types of communication tiefende Analysen im Analyses not only showed that phonological appear to place differential demands on lan- Rahmen von PISA 2000 (pp. 257–307). Wies- awareness played a significantly stronger role guage competencies. Theoretical assumptions baden: VS Verlag für in developing reading comprehension for the for such a differentiation were formulated by Sozialwissenschaften.

Center for Educational Research | 95 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 96

Figure 7. Interaction be- guage background (cf. Figure 7). However, tween written language 100 this interaction effect disappeared when performance and language background. 90 controlling for socioeconomic background. 80 This shows that performance differences 67.52 among students are related to their families’ 70 63.91 social background and language competen- 60 61.14 cies. Thus, the significant interaction effect 50 54.92 revealed in the initial analysis is not only due 40 to lesser language competencies of children 30 with a non-German language background. German background 20 Rather, social background and the language Non-German background 10 spoken at home seem to combine and lead to Proportion of correct responses (in %) 0 worse performance, with texts related both C test with C test with to school and to everyday experience. everyday content school-related In a second step, the influence of selected content aspects of language on auditory comprehension was examined. The interaction hypotheses that performance disadvantages are Key Reference Cummins (1979, 1981) and Gogolin (2003), greater for children with a non-German lan- Müller, A. G. (2007). As- for example. However, there is a lack of em- guage background than for those with a pekte schulbezogener pirical studies validating these assumptions. German language background when working Sprache für schulischen LIFE fellow Andrea G. Müller’s dissertation on texts with a difficult vocabulary or com- Erfolg von Schülerinnen und Schülern deutscher project aimed to establish that everyday- plex grammar versus texts with easy vocabu- und nichtdeutscher related language is indeed easier than lary or simple grammar could not be con- Herkunftssprache. Un- school-related language. Furthermore, the in- firmed. Furthermore, a greater performance published doctoral thesis, Free University fluence of selected aspects of language in disadvantage in conditions of language with- Berlin. the field of vocabulary and grammar as well out contextualization than in conditions with as a contextualization of language on the contextualization were not observed when auditory comprehension of children with a comparing children with, and without, a German and non-German language back- German language background. ground was examined. Two (quasi-)experi- Taken together, the results establish that mental studies were carried out to answer children with a non-German language back- these research questions. On the one hand, ground do not perform as well as those with the analyses were based on data from a Ja- a German language background, but an addi- cobs Summer Camp Project subsample (N = tional performance disadvantage in the ex- 190; see below). On the other hand, data amined aspects of the German language was from Berlin elementary school children were not confirmed. The results indicate the need analyzed (N = 529). In this quasi-experimen- for integrated promotion of German lan- tal study that took place in February 2006, guage competencies. auditory comprehension performance of children with, and without, a migration back- Promotion of Language Competencies in ground was tested. Children With Immigrant and Socially Based on oral proficiency data from the ele- Disadvantaged Families: The Jacobs Summer mentary school children, the hypothesis of an Camp Project additional performance disadvantage for Findings from international school achieve- children with a non-German versus German ment studies show that Germany is less suc- language background could not be confirmed cessful in supporting students with a migra- regarding texts of school-related content. In tion background in the acquisition of the the case of written work dealing with school- school language than other countries. At the related texts, weaker performances were ob- same time, little is known about the efficacy served for children with a non-German lan- of measures that aim at promoting second

96 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 97

Figure 8. Participants in the Jacobs Summer Camp Project.

language acquisition. This was the Jacobs proficiency level in their first language. Key References Summer Camp Project’s starting point (Jürgen Therefore, it is unclear whether the findings Stanat, P., Baumert, J., Baumert, Petra Stanat, Andrea G. Müller). A of the analyses can be transferred to children & Müller, A. G. (2005). central aim was to obtain information on the of primary school age who are learning a Förderung von deutschen Sprachkompetenzen bei effectiveness of different approaches to help- second language and whose knowledge of Kindern aus zugewan- ing students acquire proficiency in German as their first language is also often restricted. derten und sozial be- a second language. The literature on foreign The Jacobs Summer Camp Project explored nachteiligten Familien: Evaluationskonzeption language teaching and learning distinguishes this question by operationalizing the implicit für das Jacobs-Sommer- between implicit and explicit approaches strategy of second language support with a camp Projekt. Zeitschrift (DeKeyser, 1995; Ellis, 1994, 2003; Gasparini, theater-pedagogical program and the explicit für Pädagogik, 51, 856–875. 2004; Hulstijn, 2005; Nunan, 1999). Implicit strategy with systematic classes in German Stanat, P., & Müller, A. approaches create language-intensive situa- as a second language. A field-experimental G. (2005). Förderung von tions in which learners are stimulated to use design with random assignment to the exper- Schülerinnen und the target language. No measures are taken imental conditions was applied in order to Schülern mit Migrations- hintergrund. In H. to steer their attention to linguistic rules. In estimate the effects of these measures Bartnitzky & A. Speck- contrast, explicit teaching strategies focus (Stanat, Baumert, & Müller, 2005; Stanat & Hamdan (Eds.), Deutsch specifically on linguistic structures, for Müller, 2005). Following up on the literature als Zweitsprache lernen (pp. 20–32). Frankfurt example, by explaining grammatical rules on evidence-based policy and reform (e. g., a. M.: Grundschulver- and following up with their use (explicit de- Feuer, Towne, & Shavelson, 2002; Mosteller & band. duction) or by working on material revealing Boruch, 2002; National Research Council, Stanat, P., Müller, A. rules based on systematically structured lan- 2002; Slavin, 2004), the study also intended G., & Baumert, J. (2005). Die Kofferbande guage acquisition contexts (explicit induc- to demonstrate how well-controlled field ex- auf Reisen in das Land tion). periments can be implemented in educational der Sprache und des The distinction between implicit and explicit science. Theaters: Das Jacobs- Sommercamp Projekt. teaching and learning of foreign languages Children from the 3rd grade of schools in Pädagogik, 57, 57–58. has been examined both in class and under Bremen participated in the study. Of the chil- laboratory conditions. A meta-analysis of 49 dren who had applied for participation, 150 studies indicates that explicit approaches were randomly selected for the summer camp lead to better learning achievements than (experimental group). The remaining 82 chil- implicit approaches (Norris & Ortega, 2000). dren formed the baseline group. Children in However, almost 80 % of the studies included the experimental group were distributed be- in the meta-analysis examined adult learners, tween three camps. In one camp, only the and 65 % were carried out in the context of theater and leisure components of the pro- university teaching of foreign languages. In gram (morning and afternoon) were imple- only one of the studies did the target popula- mented (implicit language support). In the tion consist of primary school children. More- other two camps, the children also took part over, it can be assumed that most partici- in the theater and leisure program in the af- pants in the programs already had a high ternoons, but attended lessons for German as

Center for Educational Research | 97 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 98

Key Reference a second language in the mornings (implicit OECD Report: “Where Immigrant Students Stanat, P., & and explicit language support). To estimate Succeed—A Comparative Review of Christensen, G. (2006). the effects of the interventions, children Performance and Engagement” (2006) Where immigrant students succeed: A compar- completed tests of grammar, reading, and vo- Successful integration of immigrant students ative review of perform- cabulary shortly before and after the summer into the education system presents a central ance and engagement in vacation (pretest, posttest) and again concern to many countries worldwide. There- PISA 2003. Paris: OECD. 3 months after the completion of the pro- fore, as another core activity within this Re- gram (follow-up). search Area, Petra Stanat and Gayle The results for the written tests show that Christensen analyzed data from PISA 2003 on the summer camp program improved chil- outcomes of schooling for an OECD report. dren’s language competencies despite the in- Measures were immigrant students’ perform- tervention’s relatively brief duration. This was ance in key school subjects at the age of 15, particularly the case for the combination of their self-assessment as learners, and their implicit and explicit language support. A feelings about school. Two groups of immi- multivariate analysis of variance revealed grant students were analyzed: first-genera- that students in this treatment group tion students who were born outside the reached significantly higher performance lev- country of assessment and whose parents els in the posttest than students in the base- were also born in a different country; and line group. The advantage was most pro- second-generation students who were born nounced for grammar (d = .86) and reading in the country of assessment themselves but (d = .51). A similar tendency was observed for whose parents were born in a different coun- vocabulary (d = .37), yet the effect failed to try, that is, students who completed all of reach significance (p = .095). Students in the their schooling in the country of assessment. implicit support condition also tended to The report compares immigrant students to perform better than the control group in native students who were born in the country grammar (d = .38) and reading (d = .49). of assessment and who have at least one However, none of these differences were parent born in that country. The analyses insignificant. clude 17 countries with significant immi- A similar pattern emerged for the follow-up grant student populations: the OECD coun- tests 3 months after the intervention, al- tries Australia, Austria, Belgium, Canada, though the group differences tended to be Denmark, France, Germany, Luxembourg, the smaller and only a few of them reached sig- Netherlands, New Zealand, Norway, Sweden, nificance. The effect sizes, however, clearly Switzerland and the United States as well as indicate that students who had received both the partner countries Hong Kong China, implicit and explicit language support Macau China, and the Russian Federation. For reached higher scores in the tests of gram- the majority of these countries, as well as for mar (d = .38), reading (d = .49), and vocabu- the United Kingdom, Finland, and Spain, the lary (d = .41) than students in the baseline report also presents information on policies group. The somewhat weaker tendency for and programs to help immigrant students at- students in the implicit condition to perform tain proficiency in the language of instruc- better than students in the baseline condition. Among others, the following seven im- tion was also apparent at the follow-up. portant conclusions were drawn from the Overall, the findings of the Jacobs Summer analyses: Camp Project indicate that a combination of (1) PISA results suggest that high levels of implicit and explicit language support is ef- immigration do not necessarily impair inte- fective in helping elementary school children gration. with an immigration background to improve (2) Immigrant students are motivated learn- their second-language skills. Implicit lan- ers and have positive attitudes toward guage support alone seems to yield similar, school. Such strong learning dispositions can but weaker, effects. be developed by schools to help these students succeed in the education system.

98 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 99

(3) Despite these strong learning dispositions, tics only partially explain differences in immigrant students often perform at levels mathematics performance. significantly lower than their native peers. At (6) Similarly, performance differences in the same time, however, their performance mathematics are not fully explained by the levels vary across countries. fact that some immigrant students do not (4) In the majority of countries, at least 25 % speak the language of instruction at home. of immigrant students could face consider- However, in several countries this relation- able challenges in their future professional ship is quite strong, and may warrant and personal lives as they do not have basic stronger language support in schools. mathematics skills according to the PISA (7) Policies to help immigrant students attain 2003 assessment. proficiency in the language of instruction (5) Background characteristics of immigrant have common characteristics, but vary in student populations and school characteris- terms of explicit curricula and focus.

The Center for Educational Research in March 2007

Left to right: Gabriel Nagy, Thamar Dubberke, Stefan Krauss, Michael Becker, Nele McElvany, Uta Klusmann, Yi-Miau Tsai, Kai Maaz, Mareike Kunter, Oliver Lüdtke, Nicole Husemann, Kathrin Jonkmann, Marko Neumann, Cornelia Gresch, Ulrich Trautwein, Erin Furtak, Jürgen Baumert.

Center for Educational Research | 99 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 100

The Research Team Research Area IV Cognitive Activation in the Classroom Jürgen Baumert Mareike Kunter Introduction to Research Area IV Martin Brunner Research Area IV investigates the processes of learning and instruction, and examines how (until 11/2006) powerful learning environments can be created. The focus is on learning that occurs in Stefan Krauss authentic real-life classrooms, particularly in mathematics classrooms. Classroom instruction is not the only factor that determines the learning outcomes of students. It is, however, the Erin Furtak factor most likely to be affected by the institutions of the educational system and the profes- (postdoctoral fellow, sional activity of teachers. Recent findings emphasize that classroom instruction rather than since 09/2006) the school environment or management structures has the main impact on school effectiveness in terms of learning outcomes. As such, the question of what actually determines good Thamar Dubberke instructional practice is central to the success of education and the functioning of the educa- Uta Klusmann tional system. In Research Area IV, we address this question by combining concepts of Yi-Miau Tsai instructional quality with the analysis of individual and collective processes of knowledge (predoctoral fellows) acquisition in a specific domain, that is, in mathematics.

In collaboration The research program builds on earlier work learning. In addition, in our secondary analy- with Werner Blum that was carried out in the context of our ses of the 1996 TIMSS Videotape Classroom (University of Kassel) large-scale and longitudinal studies, such as Study, we have linked classroom observation Michael Neubrand BIJU, TIMSS, and PISA, which had provided data with student data and have been able to (University of first evidence about which features of class- describe, for example, how the types of tasks Oldenburg) room instruction are essential for students’ set in lessons influence learning processes,

In cooperation with The aim of the COACTIV study is to investigate mathematics teachers’ competence and the National PISA the way this competence relates to instructional processes. By assessing the knowledge, 2003 Consortium: beliefs, motivation, and self-regulatory skills of mathematics teachers, and then linking Jürgen Baumert these aspects to features of their classroom instruction and to the development of their (MPI for Human students’ mathematical literacy, the study is intended to provide unique insights into the Development), prerequisites for students’ mathematical learning. Werner Blum (Uni- The COACTIV study is embedded in the longitudinal component of the PISA 2003 study. versity of Kassel), Both the students sampled for PISA and their mathematics teachers were assessed Detlev Leutner twice—once at the end of 9th grade, and once at the end of 10th grade. We are thus (University of able to combine student achievement and questionnaire data with their teachers’ data, Duisburg-Essen), and to observe changes over the course of a school year. Rainer Lehmann Measures on students’ development are obtained from the students’ PISA test scores and (Humboldt Univer- additional questionnaire data. Information on mathematics instruction comes from three sity Berlin), Michael sources: teacher reports, student reports (both standardized questionnaires), and analy- Neubrand (Univer- ses of the teaching material that was used in the given period. For this, teachers were sity of Oldenburg), asked to submit the tasks they actually employed in their PISA classes (homework as- Reinhard Pekrun signments, exams, and tasks used in introductory lessons). These tasks were coded by (University of trained raters using a newly developed classification scheme to tap the cognitive poten- Munich), Manfred tial inherent in the tasks. To assess teachers’ competence, we developed an array of new Prenzel, Jürgen Rost instruments, some of which are computer-based. In particular, we focused on developing (Leibniz Institute for a standardized test to assess teachers’ content and pedagogical content knowledge. Science Education, Our teacher sample consists of 351 teachers and their mathematics classes in the first Kiel), Ulrich wave, and 240 teachers and their classes in the second wave (the reduction in sample Schiefele (University size is due to students from vocational schools no longer being included in the assess- of Bielefeld) ment at the second wave). A total of 178 teachers participated in both waves of the assessment, and taught the PISA classes over the whole school year.

100 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 101

and ultimately impact on students’ learning process-mediation-product model, and the Key References outcomes, including motivational develop- (social-)constructivist approach. The general Kunter, M., Klusmann, ment. In our current research program, we approach that structures our framework is U., Dubberke, T., Blum, take the idea of learning environments one outlined in Figure 1. W., Brunner, M., Jordan, A., Krauss, S., Löwen, K., step further and put a new focus on the Our theoretical framework is based on the Neubrand, M., & Tsai, teacher as the prime agent in the teaching notion that conceptual understanding is a Y.-M. (in press). Linking and learning process. One of our core objec- central aim of mathematics instruction. It is aspects of teacher competence to their instruc- tives therefore is to investigate the prerequi- now widely accepted that new concepts and tion: Results from the sites that teachers need to have in order to insights are not acquired through passive COACTIV project. In M. be able to create successful learning environ- knowledge transmission of the teachers’ Prenzel (Ed.), Studies on the educational quality ments. knowledge to the learner’s mind, but rather of schools: The final re- Methodologically, in Research Area IV, we that they are the result of the learner’s active port on the DFG priority build on correlational and quasi-experimental process of constructing increasingly complex program. Münster: Wax- mann. studies in which we often combine teacher and elaborated cognitive structures. This Brunner, M., Kunter, and student data to gain valid insights into process of insightful learning is characterized M., Krauss, S., the instructional processes. The flagship of by the learner’s active mental engagement Klusmann, U., Baumert, the research program of Area IV is the that occurs in interaction with the learning J., Blum, W., et al. (2006). Die professionelle COACTIV study which is embedded in the environment. The product of this process is a Kompetenz von Mathe- PISA 2003 assessment and investigates thorough understanding of domain-specific matiklehrkräften: Kon- mathematical teachers’ competence and the concepts comprising declarative knowledge, zeptualisierung, Erfas- sung und Bedeutung für way several aspects of competence relate to skills, and procedures. The process of insight- den Unterricht: Eine teachers’ instruction in their mathematics ful learning is supported by motivational Zwischenbilanz des classes. In our ongoing research program, we variables, such as interest, and self-related COACTIV-Projekts. In M. Prenzel & L. Allolio- complement the findings from this study cognitions, such as control beliefs. At the Näcke (Eds.), Unter- with new research projects that follow some same time, insightful learning processes may suchungen zur Bildungs- of the questions that arose in more detail. also result in the development of functional qualität von Schule: Abschlussbericht des motivational and self-related cognitions. DFG-Schwerpunktpro- Insightful Learning in Powerful Learning We see the classroom instruction as a struc- gramms (pp. 54–82). Environments tured learning opportunity that may, or may Münster: Waxmann. The research in Area IV is guided by a theo- not, provide the occasion for insightful learn- retical framework that conceptualizes teaching. Powerful classroom environments are ers, lessons, and students as the three cor- those that stimulate students to apply them- nerstones of teaching and learning processes selves cognitively and that are structured in in classroom instruction. The framework such a way that active and independent draws on aspects of teacher expertise, the knowledge construction is possible. The most

Figure 1. Teachers’ competence, mathematics Student instruction lessons, and Teacher Instruction students’ learning: The underlying structure of Knowledge the theoretical frame- Cognitively activating Knowledge work in the COACTIV elements study. Motivation Classroom Beliefs management Beliefs Psychological Personal learning functioning support Self-regulation Insightful learning process

Center for Educational Research | 101 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 102

Key References important key to trigger and uphold active ture of the learning environment, for in- Kunter, M., & Baumert, knowledge construction is the attunement of stance, the general lesson scripts or the use J. (2006a). Linking TIMSS the tasks to be mastered to the students’ of particular learning settings, such as whole to research on learning and instruction: A re- prior knowledge and understanding. In our class discussion, group work, or partner work. analysis of the German work—which will be described in more detail However, instructional research has shown TIMSS and TIMSS video in the next paragraph—we identified three that these surface features alone do not de- data. In S. J. Howie & T. Plomp (Eds.), Contexts of general features of instruction that are cru- cide whether students will engage in insight- learning mathematics cial to enable insightful learning processes in ful learning processes, and thus do not nec- and science: Lessons secondary school mathematics, namely, cog- essarily predict learning outcomes. In order learned from TIMSS (pp. 335–351). London: nitively activating elements, classroom man- to describe strengths and weaknesses of in- Routledge. agement, and individual learning support. struction, it is fruitful to describe the interac- Kunter, M., & Baumert, It is important to note that the uptake of the tion between teacher and students in the in- J. (2006b). Who is the learning opportunity depends both on the structional process in more detail. expert? Construct and students themselves (according to their indi- Prior research, for instance from the 1996 criteria validity of student and teacher ratings vidual prerequisites) and on situational con- TIMSS Videotape Classroom Study (e. g., of instruction. Learning straints (Baumert, Blum, & Neubrand, 2004; Baumert et al., 1997; Klieme et al., 2001; Environments Research, Helmke, 2003). Successful instruction thus Kunter, 2005; Kunter & Baumert, 2006a), has 9, 231–251. hinges on the degree to which instructional shown that both formal organization and the Kunter, M., Dubberke, T., Baumert, J., Blum, strategies are suited to the needs of the situ- instructional interactions of mathematics W., Brunner, M., Jordan, ation and the students. Instructors therefore teaching in German secondary schools are A., et al. (2006). Mathe- need to provide challenging tasks, monitor not optimally suited to trigger insightful matikunterricht in den PISA-Klassen 2004: Rah- student learning, and need to adapt their learning processes. Formally, the “monocul- menbedingungen, For- teaching accordingly if they want to support ture of mathematics teaching” has often men und Lehr-Lern- active and independent knowledge construc- been noted, with lessons typically following a prozesse. In M. Prenzel, J. Baumert, W. Blum, R. tion for many learners (de Corte et al., 1996). rather rigid script of teacher presentation Lehmann, D. Leutner, M. For teachers in classroom situations, this is a with class discussion and ensuing seat work. Neubrand et al. (Eds.), demanding task—it is no easy matter to In terms of the instructional process, the PISA 2003: Untersuchun- gen zur Kompetenzent- create challenging and suitable learning con- TIMSS Video study has shown that German wicklung im Verlauf eines ditions for groups of students who may differ mathematics lessons have typically focused Schuljahres (pp. 161– greatly in terms of motivation or prior knowl- on drilling routines rather than on developing 194). Münster: Wax- mann. edge. Such deliberate, but, at the same time, conceptual knowledge. Teachers guided stu- flexible and adaptive action, can only be dents through new topics step-by-step, pre- Kunter, M. (2005). Mul- tiple Ziele im Mathe- achieved based on a solid knowledge base senting the new ideas themselves, and rarely matikunterricht. that is supported by adaptive beliefs and ad- making reference to students’ conceptions. Münster: Waxmann (Pädagogische Psycholo- equate psychological functioning. From this The tasks students worked on were very rou- gie und Entwicklungs- perspective, particular relevance is put on the tine-oriented, and their solutions required psychologie 51). person of the teacher as an important agent the application of procedures rather than in the instructional process. Consequently, as conceptual understanding. one major part in our recent work, we inves- The findings from the TIMSS Video study and tigate which aspects of teacher compe- the disappointing results of the PISA 2000 tence—such as knowledge, beliefs, and their study have received much attention in psychological functioning—are relevant for Germany, leading to several programs in re- educational outcomes. Before we turn to this search and teacher education aiming to im- issue, we will first describe our notion of high prove the quality of instruction, particularly quality in more detail. in the domain of mathematics. One of the questions addressed by the COACTIV study What Are the Features of High Quality was whether improvements in the current Instruction? German mathematics instruction were ob- Classroom instruction can be described in servable. With our assessment being embed- various ways. As a first approach, it is often ded in the PISA 2003 study, we can draw on useful to describe the organizational struc- a representative sample of mathematics

102 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 103

Figure 2. Usage of partic- 100 ular learning settings in mathematics lessons: 90 Results from the teacher survey in COACTIV. 80

50 (in %) 40

0 Teacher Group Partner Individual Individual presentation work work seat work learning schedule

Rarely or never Sometimes More often Regularly

classes and their teachers. We used a multi- use this type of learning arrangement. These methodological approach in which we com- results indicate that formal learning arrange- bined teacher self-reports, student ratings of ments in the mathematics lessons are still instruction, and analyses of the tasks that —10 years after the TIMSS Video study—very teachers employed in their lessons. We exam- teacher-centered. Moreover, learning ined the formal arrangements of the lessons arrangements that allow for collaborative as well as the interaction between teacher learning or explicitly cater to individual stu- and students in the instructional process. dents’ needs are rarely used.

Formal Arrangements of Mathematics Features of the Instructional Process Lessons Within the COACTIV framework, we distin- Our results indicate that the picture of the guish three aspects of the instructional inter- uniform organization has not changed much action between teacher and students that in secondary school mathematics classes theoretically seem crucial for initiating and since the results from TIMSS in the 1990s. sustaining insightful learning processes in Figure 2 presents results from the teacher mathematics lessons (see Figure 1). First, we survey in which teachers were asked how of- consider the frequency of cognitively activat- ten they implement several learning settings, ing elements of instruction. Cognitively acti- such as teacher presentation, partner or vating elements refers to all learning situa- group work, or individual seat work. tions that may trigger students’ conceptual As can be seen in Figure 2, the most frequent involvement. In the mathematics classroom, learning arrangements are teacher presenta- it is determined primarily by the types of tion and individual seat work, but collabora- tasks and the way they are implemented. tive learning settings are rarely used. Of par- Tasks that draw on students’ prior knowledge ticular interest is the last bar of the figure, by challenging their beliefs is one example. which reports the implementation of individ- While implementing tasks in class discussion, ualized learning schedules. It shows that over the teacher does not simply declare students’ 80 % of the teachers do not, or hardly ever, answers to be “right” or “wrong,” but encour-

Center for Educational Research | 103 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 104

Table 1 Even Tasks on a Low Difficulty Level May Imply Conceptual Understanding: Examples of Tasks With Diverse Levels of Curricular Knowledge and Cognitive Demands

Low level High level of curricular knowledge of curricular knowledge

Technical task Consider the function Let f be the quadratic function f(x) = 3x – 1. f(x) = 2x2 + 5x – 3. Find the value of x Write f(x) in vertex form. when f(x) = 11.

Conceptual How does the surface area Thomas’ father wants to give him some money towards a modeling task of a square change when motorbike. He makes Thomas two offers: the side length is tripled? (1) 15€ today, a further 20€ tomorrow, a further 25€ the Show your reasoning. next day, and so on for 2 weeks (i.e., increasing the sum by 5€ every day). (2) 5 cents today, a further 10 cents tomorrow, a further 20 cents the next day, and so on (i.e., doubling the sum every day), again for 14 days. Which offer should Thomas choose? Show your reasoning.

ages students to evaluate the validity of their quired to solve them. Technical tasks require solutions for themselves. The second impor- only factual knowledge or computational tant aspect is classroom management. Class- skills. Modeling tasks, on the other hand, re- rooms with less disruption and more effi- quire students first to construct a represen- ciency in time use are a prerequisite for tation of the problem situation by interpret- learning to occur. Third, we consider the per- ing the information given in the task state- sonal learning support provided by the ment. This situational model then has to be teacher. Motivational theories have shown translated into a mathematical model, from that students engage in insightful learning which the mathematical solutions needed to processes and develop intrinsic motivation solve the problem can be derived. These solu- when they feel personally supported in their tion strategies then have to be implemented, learning environments. Therefore, teachers interpreted, and validated. The match be- need to be acutely aware of possible difficul- tween the situational model and the mathe- ties and offer assistance to stimulate individ- matical model determines the quality of the ual learning processes. task solution. Modeling tasks are further cat- Particularly with regard to cognitive potential egorized as computational modeling tasks or inherent in lessons, our results also tend to conceptual modeling tasks. If the situational replicate the TIMSS findings of routine-ori- model is close to the mathematical model re- ented rather than problem-solving-oriented quired to solve it, and the solution strategies instruction. In the COACTIV study, we asked primarily entail calculations and mathemati- the teachers to submit (among others) all the cal algorithms (even demanding ones), it is exams they set in the PISA classes during the called a computational modeling task. If the whole school year. On average, each teacher mathematical model requires the students to submitted 53 exam tasks which were then link several concepts or strategies, or to draw coded by trained raters, using a newly devel- inferences going beyond the information oped classification scheme to tap the cogni- given in the task statement, the task is called tive potential inherent in the tasks. Several a conceptual modeling task. It is important to categories of cognitive potential were evalu- note that this classification does not equate ated. For instance, all tasks were classified as with problem difficulty in terms of the aver- either technical tasks or modeling tasks, age student success rate on a problem. All based on the cognitive processes that are re- three types of problems—technical tasks,

104 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 105

computational modeling tasks, and concep- Figure 3. Cognitive tual modeling tasks—can occur at all diffi- Task types potential of the exam tasks set by the teachers culty levels. Table 1 illustrates this point. 8% in the COACTIV mathe- Figure 3 shows that of all exam tasks, almost Technical tasks matics. half were technical tasks, meaning that their Conceptual modeling solution required only factual knowledge or 43% tasks computational skills. The other half were 49% Computational modeling tasks modeling tasks. Of these, most were computational modeling tasks that required procedural thinking during the processing phase. Only 8 % of all tasks were conceptual modeling tasks in which conceptual thinking was we also investigate the effects of instruction needed during the processing phase, such as on students’ motivational development. For linking several concepts or strategies or instance, a reanalysis of the 1996 TIMSS data drawing inferences beyond the information in showed that students’ interest development the task statement. Considering the fact that may be fostered if students perceive their exam tasks are meant to represent the ac- mathematics instruction as transparent and complishment of learning goals at the end of feel monitored by their teacher. In another a unit, this finding illustrates the computa- current research project, LIFE fellow Yi-Miau tional orientation of German mathematics Tsai investigates instructional effects on stu- instruction in a very striking way. dent motivation in an intraindividual design. Nine classes including 261 students and 25 Effects of Instructional Features on teachers participated in the study with a re- Educational Outcomes peated-measurement design. For 3 consecu- Our results thus indicate that truly powerful tive weeks, students and teachers reported mathematics instruction that supports inde- their lesson experience immediately after pendent and insightful student learning is a each lesson, particularly their interest and rather rare occurrence in German secondary self-view. First analyses of the students’ data school mathematics classes. However, despite showed that students’ self-reported instruc- this general tendency, the instructional qual- tional quality predicts their interest experi- ity varies between classes and teachers. ence and perceived competence during the Using the data from the longitudinal compo- lesson. Students enjoyed lessons more when nent of the PISA 2003/04 assessment, we teachers provided them with personal and therefore investigated whether higher in- cognitive autonomy support. In addition, structional quality can systematically be such beneficial effects are not limited to linked to students’ learning over the course mathematics lessons, they are also applicable of 1 school year. We found that classes in in language lessons. which tasks with high cognitive potential were set showed higher mathematics How Can Teachers Provide High Quality achievement than other classes. In addition Instruction? to the cognitive potential, efficient classroom Educational research has recently seen a re- management was another factor that influ- markable upsurge in interest in the person of enced the learning of students positively. the teacher as one of the main agents in the In our future work, we will investigate the in- instructional process. Providing cognitively terplay between the three aspects of the in- activating instruction, managing the class- structional process in more detail. Moreover, room efficiently, and supporting students in our work, we have, as a first step, concen- personally require high levels of attention trated on the learning gains of students as and engagement of the teacher. Moreover, one of many desirable outcomes of instruc- teachers need a broad knowledge of instruction. However, taking the multifacetted national methods and strategies to establish ture of educational objectives into account, well-structured and cognitively activating

Center for Educational Research | 105 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 106

Key Reference learning situations, and need to implement Knowledge Kunter, M., Baumert, J., them flexibly according to students’ needs. In order to create a challenging and adaptive & Köller, O. (in press). For the pursuit of cognitively activating ele- learning environment in the mathematics Effective classroom ments in particular, domain-specific back- classroom, teachers need to select mathe- management and the development of subject- ground knowledge seems to be of high im- matical tasks that meet the differing needs related interest. Learning portance. of individual students, that is, tasks that vary and Instruction. To date, empirical evidence on which teacher in terms of their difficulty and cognitive de- characteristics are particularly important to mands as well as in structural and contextual create high-quality learning situations is still terms, thus helping learners gain transferable scarce, one reason for this being the lack of knowledge. These tasks must be presented in adequate measurement instruments to tap an order that allows students to gradually teachers’ knowledge and other characteris- expand their knowledge and practice existing tics. The COACTIV study is one of the first skills. This requires a profound base of knowl- studies that succeeded in developing a com- edge on the teacher’s part. In COACTIV, we prehensive assessment instrument measuring therefore focus on teachers’ domain-specific several aspects of mathematics teachers’ knowledge. Drawing on Shulman (1986, competence. Drawing on our framework of 1987), we differentiate between mathemat- powerful mathematics instruction outlined ics-related content knowledge (CK) and above, we derived several prerequisites that mathematics-related pedagogical content seem crucial for teachers to successfully knowledge (PCK). CK concerns the content to meet the demands of (mathematics) teach- be taught in school, whereas PCK is the ing. These include aspects of knowledge, be- awareness of how best to present this con- liefs, and variables of psychological function- tent to students. Important facets of PCK in- ing, and are seen as being at the core of clude knowledge of instructional strategies mathematical teachers’ competence. Figure 4 and knowledge of students’ understanding as gives an overview of the competence aspects well as knowledge of the potential of mathe- that we examined. matical tasks.

Figure 4. Teacher competence as conceptualized in the COACTIV study. Aspects of teacher competence Knowledge Beliefs Psychological functioning

Areas of knowledge

Content Pedagogical Pedagogical Organizational Counseling knowledge content knowledge knowledge knowledge knowledge

Facets of knowledge Deep understanding of school mathematics Knowledge of students mathematical thinking Knowledge about representing and explaining mathematics … Knowledge about classroom management Knowledge about student’s learning in general …

106 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 107

Beliefs regulatory style, particularly where general Key References What teachers think about their subject, occupational engagement and coping are Baumert, J., & Kunter, and—even more importantly—whether they concerned. M. (2006). Stichwort: Professionelle Kompe- feel responsible for their students and seek to tenz von Lehrkräften. support their personal growth, may deter- Assessing Teacher Competence Zeitschrift für Erzie- mine whether or not their teaching serves to Although notions of teacher competence are hungswissenschaft, 9, create powerful learning environments. In convincing from the theoretical perspective, 469–520. COACTIV, we consider teacher’s beliefs about empirical evidence to support them is as yet Klusmann, U., Kunter, M., Trautwein, U., & the nature of knowledge and knowing (epis- sporadic. An empirical approach to the ques- Baumert, J. (2006). temological beliefs), theories about the nation is complicated for both theoretical and Lehrerbelastung und Un- ture of learning and teaching, and teachers’ methodological reasons. Theoretically, a con- terrichtsqualität aus der Perspektive von Lehren- instructional goals, all referring to the conceptualization of teacher competence as a den und Lernenden. text of mathematics. We differentiate be- multidimensional construct needs to unite Zeitschrift für Pädago- tween a transmission view (drawing on tradi- several theoretical approaches: domain-spe- gische Psychologie, 20 (3), 161–173. tional learning theories) and a constructivist cific theories of student learning, general view (e. g., Staub & Stern, 2002). learning theories, and psychological theories of human functioning. Methodologically, an Psychological Functioning: Engaging and empirical assessment of competence requires Coping reliable measures, which are scarce at pres- We assume that a combination of high en- ent. There are few standardized instruments gagement and a high capacity to deal with —particularly in the area of teacher knowl- the pressures of school life is crucial for edge but also in all other areas of interest— teachers’ psychological functioning. On the and most previous findings are based on one hand, teaching—perhaps more than many qualitative and anecdotal data. Moreover, it other professions—calls for continuous en- is vital to ensure variability of the target gagement, and requires teachers to apply variables in the teacher samples, for example, themselves personally: Students demand by drawing sufficiently large samples or us- teachers’ attention, standardized curricula ing quasi-experimental approaches. Finally, create time pressures, and the lack of exter- cross-sectional data need to be supple- nal guidance and feedback necessitates high mented by either longitudinal or experimen- levels of self-motivation. At the same time, tal approaches if conclusions are to be drawn teachers work in a complex social environ- on the causal influences of teacher compe- ment that requires immediate responses to tence. multiple social stimuli, always under the stu- The COACTIV study was designed to address dents’ scrutiny (Doyle, 1986). To do well in these questions by conceptualizing, assess- this situation, teachers need to be able to ing, and validating a model of teacher com- distance themselves from these pressures, petence for mathematics instruction. The de- making themselves less vulnerable to criti- velopment of valid and reliable instruments cism and failure. In COACTIV, we examine that are suited to assess these aspects of different motivational variables (i. e., their teachers’ competence was one of the major general professional engagement, such as ca- challenges of our study. This was particularly reer ambitions or the subjective significance the case for the assessment of teachers’ of their work, and their mathematics-specific knowledge, as no standard instruments ex- engagement, e. g., teachers’ enthusiasm for isted thus far. The theory-based construction mathematics) as one motor for their engage- of the tests of PCK and CK was a multidisci- ment. In terms of teachers’ ability to cope plinary and a multistep project. Researchers with pressures and failures, we consider in mathematics education, psychology, and aspects of their resilience (e. g., emotional secondary mathematics teachers collaborated distancing and application of active coping in writing, piloting, and analyzing items. The strategies). The interplay between the two test on teachers’ PCK included items that dimensions may constitute a specific self- asked teachers to provide explanations for

Center for Educational Research | 107 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 108

Figure 5. Sample items from the test on mathematics teachers’ CK and “Square” “Parallelogram” PCK. (“Tasks” subfacet of PCK) (“Students” subfacet of PCK)

How does the surface area of a square change The area of a parallelogram can be calculated when the side length is tripled? by multiplying the length of its base by its Show your reasoning. height.

Please note down as many different ways of height solving this problem (and different reasonings) as possible. base Please sketch an example of a parallelogram to which students might fail to apply this formula.

“–1 times –1” “Infinite decimal” (“Instruction” subfacet of PCK) (Test of CK)

A student says: I don’t understand why Is it true that 0.999999....=1? (–1).(–1)=1 Please give detailed reasons for your answer.

Please outline as many different ways as possible of explaining this mathematical fact to your student.

mathematical problems, items that required The First Results From COACTIV: Linking teachers to detect typical student errors, and Teacher Competence and Instruction items in which teachers had to find multiple Using our newly developed measures, we ways for solving tasks. The items in the CK have started the systematic investigation of task tapped conceptual and/or procedural interindividual differences in teacher compe- mathematical skills. Although the items re- tence and in instruction, and have linked as- lated to mathematical content that is typi- pects of teacher competence to their teach- cally taught at high school, they were essen- ing. tially too difficult for students, but required a As a first step, it was important to learn more deep understanding of mathematics. Sample about the configuration of the various as- items of the test are displayed in Figure 5. pects of teacher competence. Is there some- The successful construction of a set of valid one resembling a “perfect teacher” who and reliable measures tapping aspects of scores high on all our competence measures teacher competence was one of the most im- (and vice versa)—or is it, for instance, typical portant milestones of the COACTIV study. that some teachers have high knowledge These new measures now allow us to learn scores, but show rather low engagement? A more about the relations between the differ- first answer to the question can be found in ent aspects of teacher competence, to com- Figure 6 which shows the intercorrelations pare different groups of teachers, and, most between selected aspects of teacher compe- importantly, to investigate which of the com- tence, namely, teachers’ CK, PCK, their belief petence aspects are particularly relevant for that mathematics is best taught in direct educational outcomes. transmission mode, their enthusiasm for mathematics, and a coping style that is marked by high engagement and high resilience at the same time. The figure shows that both knowledge types and teachers’

108 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 109

Figure 6. The relation between multiple aspects of teacher competence (latent correlation from a –.36 Structural Equation Model (SEM). .60 –.44 Pedagogical Content Transmission Enthusiasm for Adaptive content knowledge view mathematics coping style knowledge

beliefs are related to each other, suggesting the student reports: The more teachers know Key Reference (a) that teachers with higher mathematics- about how to make mathematical content Brunner, M., Kunter, specific CK usually also know better how accessible to students, the more cognitively M., Krauss, S., Baumert, mathematics content can be made accessible demanding tasks they set in class (e. g., in J., Blum, W., Dubberke, T., et al. (2006). Welche to students, and (b) that typically teachers their exams), and the more their students feel Zusammenhänge beste- with higher scores in the knowledge tests challenged in the classroom. Teachers with hen zwischen dem tend to reject the traditional transmission higher PCK also show higher levels of individ- fachspezifischen Profes- sionswissen von Mathe- view of teaching. However, teachers’ knowl- ual learning support, indicating that a certain matiklehrkräften und edge and beliefs are not directly linked with level of content-specific knowledge is needed ihrer Ausbildung sowie teachers’ psychological functioning, as to provide students with adequate guidance. beruflichen Fortbildung? Zeitschrift für Erzie- teacher’s enthusiasm and coping style do not Teachers’ pure CK, however, does not directly hungswissenschaft, 9, correlate with the other measures. contribute to teachers’ instructional quality: 521–544. In terms of interindividual differences, we Whether teachers have a deep understanding found that teachers vary substantially in all of their subject does not seem to be reflected aspects of competence, and that these differ- directly in their teaching—at least not in a ences are especially pronounced in knowl- way that the students notice. Instead, CK edge. These knowledge differences are at seems to exert an indirect influence, as it least partially dependent on the type of uni- predicts teachers’ PCK. Concerning other versity education teachers received, as we variables, teachers’ coping style seems to be found substantially higher levels of PCK and the most significant predictor: Teachers who CK or those teachers who were trained to show high levels of commitment, but are able teach the academic track (see Figure 7). In- to distance themselves from problems at the terestingly, person variables, such as age or same time, seem best equipped to provide gender, were not significantly related to students with patient and constructive sup- teachers’ knowledge. With regard to psycho- port. Interestingly, there were no systematic logical functioning, however, older teachers associations between teachers’ enthusiasm were less likely to be enthusiastic or to report for the subject of mathematics and their in- effective coping skills. struction. Evidently, teachers’ interest in, and The crucial question of course is whether enjoyment of, a subject is not necessarily re- teachers’ competence is systematically linked flected in specific classroom behaviors. to their mathematics instruction. The Figure 8 illustrates our findings by showing COACTIV design provides an excellent data- an SEM in which we specified the teachers’ base to explore this question in detail, as we characteristics as predictors of their teaching are able to tap information about the quality behaviors, as rated by their students. of instruction from various sources, such as With the establishment of the link between task analysis and students’ reports, thus go- teachers’ characteristics and their instruc- ing beyond teacher self-reports. tion, we have a first indication that teacher We found that teachers’ PCK uniquely pre- competence is an array of different features, dicts the degree of cognitive activation in including cognitive, motivational and do- class, both measured by the task material and main-specific as well as general person char-

Center for Educational Research | 109 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 110

Figure 7. Differences in teachers’ pedagogical 1.5 context knowledge and content knowledge depending on their univer- 0.79 1.0 sity education. 0.53 0.35 0.5 0.02 0.03

0.0

–0.5 –0.16

Mean and 95%-confidence interval –0.31 –0.41 –0.50 –1.0 Content knowledge Pedagogical content Pedagogical content knowledge knowledge with equal level of content knowledge

Gymnasium teachers Non-Gymnasium teachers GDR teachers

acteristics. In our future analyses, we will compare exhausted and nonexhausted now further examine the importance of these teachers. Another direction of our research aspects for the creation of meaningful learn- program is the extension of our framework ing environments, and eventually for stu- over and above the domain of mathematics, dents’ cognitive and motivational develop- investigating aspects of general pedagogical ment. In order to learn more about the psy- knowledge as well as knowledge in other chological functioning of teachers, we are domains, such as reading (see Research currently running a quasi-experimental inter- Area III). view and questionnaire study in which we

Figure 8. Aspects of teacher competence as R 2 =.36 R 2 =.11 predictors of student- Pedagogical reported teaching behav- .32* Cognitive content iors (standardized re- activation .60* knowledge gression coefficients and 2 R of the structural 2 model). R =.13 –.36* Transmission 2 view R =.04 Content Classroom knowledge R 2 =.01 management

Enthusiasm for mathematics

R 2 =.00 R 2 =.10

Adaptive .20* Individual coping style learning support

110 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 111

Completed Project ENTERPRISE: Early tools that support an understanding of pro- The Research Team Learning of Science and Mathematics portional concepts in a scientific context, that is, speed, density, or concentration. It Elsbeth Stern The ENTERPRISE project (Enhancing was shown that with the help of tools, like Ilonca Hardy KNowledge Transfer and Efficient Rea- balance beams or graphs, 3rd graders were soning by Practicing Representation In already able to develop a deeper and trans- Henrik Saalbach Science Education) aims to explore the ferable understanding of scientific concepts. Michael Schneider conditions under which graphs and dia- In several PhD theses and in cooperation (predoctoral grams can serve as tools for structuring projects with the Department of Science Ed- fellows) learning environments and, thus, foster ucation in Elementary School of the Univer- conceptual understanding in science as sity of Münster, evidence for the short-term well as other content areas. as well as long-term effects of tool use on deeper conceptual understanding could be In 1997, Elsbeth Stern was hired as a senior obtained (Hardy, Schneider, Jonen, Stern, & Understanding research scientist to establish a research Möller, 2005; Hardy, Jonen, Stern, & Möller, comes from having project on science and mathematics learning 2006; see also the past three annual reports). a good represen- for elementary school children. The effects of cognitive tool use on the emer- tation The project was based on a domain-specific gence of conceptual knowledge was ad- Novick & Hmleo perspective of cognitive development, argu- dressed in two dissertations finished in 2006. ing that the difficulties experienced by Michael Schneider investigated how practice younger children in coping with specific cog- with number line games affects the relation- nitive demands can be traced back to their ship between procedural and conceptual lack of domain-specific conceptual and pro- knowledge. Henrik Saalbach ran cross-cul- cedural knowledge. In this perspective, learn- tural studies in order to explore the impact of ing environments for young children should grammar as a cognitive tool on the concep- focus on laying the foundations for the kind tual development of young children of knowledge acquisition that is most rele- (Saalbach & Stern, 2004). vant and challenging in the upper grades of school. Knowledge dealt with in all kinds of Cognitive Tool Use and Its Effects on academic contexts is based on symbolic sys- the Interaction Between Conceptual and Key References tems, such as script, formal mathematical Procedural Knowledge Hardy, I., Jonen, A., language, pictures, and diagrammatic repre- Conceptual knowledge, as seen by cognitive Möller, K., & Stern, E. (2006). Effects of in- sentations. Symbols can be understood as theories of learning, provides an in-depth un- structional support in mental tools that support the construction of derstanding of the principles of, and inter- constructivist learning meaning in concepts, ideas, or plans. relations among, pieces of knowledge in a environments for elementary school students' The longitudinal studies Elsbeth Stern was domain. Procedural knowledge allows for the understanding of "float- involved in before she started her project at efficient and goal-directed solution of rou- ing and sinking." Journal the MPIB clearly suggest a strong impact of tine problems, and is represented, on the of Educational Psychol- ogy, 98, 307–325. early number understanding on later mathe- cognitive level, by production rules with dif- Hardy, I., Schneider, M., matical understanding (Stern, 2005; Stern & ferent degrees of automaticity. Jonen, A., Stern, E., & Felbrich, 2006). During the past decade, ele- In the literature addressing the interrelations Möller, K. (2005). Foster- mentary school mathematics in German between conceptual and procedural knowl- ing diagrammatic reasoning in science educa- schools has considerably improved: Curricula edge, three basic approaches can be distin- tion. Swiss Journal of focusing on the development of the number guished: concepts-first approaches, which Psychology, 64, 207–217. sense rather than on practicing numerical are based on the assumption that the acqui- Saalbach, H., & Stern, E. facts are on the advance. Early science edu- sition of conceptual knowledge takes place, (2004). Differences between Chinese mor- cation, in contrast, still had a lot of catching as a rule, prior to that of procedural knowl- pho-syllabic and German up to do in the 1990s. As a consequence, the edge and, subsequently, causes an increase in alphabetic readers in the major goal of the ENTERPRISE project was to procedural knowledge; procedures-first ap- Stroop interference effect. Psychonomic foster science learning in elementary school proaches, which postulate the exact opposite Bulletin & Review, 11, children by providing students with reasoning with respect to the sequence of acquisition 709–715.

Center for Educational Research | 111 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 112

Key References and the direction of causation; and the Itera- Stern, E., Felbrich, A., & tive Model, which holds that the sequence of Schneider, M. (2006). acquisition depends on experience and that Mathematik lernen. In causal relations are bidirectional. In his the- D. Rost (Ed.), Hand- wörterbuch Pädago- sis, Michael Schneider took up the challenge gische Psychologie (3rd, to find measures allowing for the separation rev. ed., pp. 461–469). of conceptual and procedural knowledge Weinheim: Beltz. which, in actual behavior, are inextricably in- FANG Stern, E. (2005d). DAS MONSTER !!! Knowledge restructuring terwoven. as a powerful mecha- Predictions based on the Iterative Model “Catch the Monster!!!” nism of cognitive devel- were tested in the well-researched field of opment: How to lay an Figure 2. The catch-the-monster game. early foundation for decimal fractions against predictions based conceptual understand- on the concepts-first and the procedures- ing in formal domains. first approaches. Four measures of concep- In P. D. Tomlinson, J. Bewege den blauen Schieber mit der Maus auf die Stelle des Zahlenstrahls, Dockrell, & P. Winne tual knowledge (evaluation of procedures, die den Dezimalbruch angibt. Klicke danach auf “Weiter”. (Eds.), Pedagogy—teach- translation into diagrams, size comparison, ing for learning and written explanations) and four measures (pp. 153–169). Leicester: British Psychological of procedural knowledge (problem-solving

Society (British Journal accuracy, problem-solving duration, asymme- 0.3 of Educational Psychol- try of access, dual-task costs) were used, all ogy Monograph 0 1 Series II, 3). of them adapted from published studies. In a first study, knowledge types were mod- weiter eled in structural equation models (SEM) as two latent factors underlying the eight mani- Figure 3. Practicing procedural knowledge: The task fest measures, thus permitting predictive re- was: “Use the mouse to move the blue slide to the position of the number line which matches the deci- lations between factor pairs to be examined mal fraction.” across three points of measurement in a cross-lagged panel design. The sample con-

sisted of 204 5th graders and 6th graders. Erkläre auf Seite 1, warum die Zahl dort liegt, wo das Monster steht. Klicke dann auf “Weiter”. The eight measures were shown to possess

.57 .75 .66 .33 .51 .74 .77 .32 0.3 CE CT CS CW CE CT CS CW 0 1

weiter .45 C1 C2 .36 Figure 4. Practicing conceptual knowledge: The task was: “Explain briefly why the number is placed at the .86 .93 position of the monster.” .34 P1 P2 .19 high convergent validities, but low divergent validities. Intercorrelations between the latent factors were very high. This proved to be an obstacle to the analysis of predictive rela- PC PD AS DU PC PD AS DU CFI = 0.956 .85 .10 .32 .30 .87 .30 .32 .42 RMSEA = 0.049 tions, which provided some clues, but no un- ambiguous evidence, for the validity of the Figure 1. SEM on the relationship between the measures of conceptual knowledge Iterative Model, as can be seen in Figure 1. (C) (evaluation of procedures [CE], translation into diagrams [CT], size comparison In a consecutive study with 84 5th graders, it [CS], and written explanations [CW]) and four measures of procedural knowledge (problem-solving accuracy [PC], problem-solving duration [PD], asymmetry of ac- was investigated whether, nonetheless, process [AS], dual-task costs[DU]). cedural and conceptual knowledge can be

112 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 113

fostered independently of each other by (2) In the word learning task, participants practicing with cognitive tools either in a were asked to extend a novel label that was more procedural or in a more conceptual given to the standard (“This is called FEP. manner. Two learning environments based on What else is called FEP?”). the “catch-the-monster game” were realized, (3) In the property induction task, partici- as shown in Figures 2 to 4. In a 2 x 3 experi- pants were taught a novel nonperceptual mental design with two independent treat- property of the standard object (e. g., “Look, ment groups and three measurement points, this one has IDOFORM inside”) and asked to no between-group differences were revealed. select the item that would be most likely to All in all, results show that conceptual and have the same property. procedural knowledge are much more According to the widely accepted linguistic strongly interrelated than one would expect relativity approach, the conceptual structure from all previous studies which, using only of children and of adults can be affected by manifest measures, were more prone to the grammar of their mother tongue. Cross- measurement errors. Therefore, the Iterative language studies have revealed that certain Model could be neither validated nor invali- grammar elements can affect cognitive de- dated. Although the distinction between both velopment mainly by influencing the foci of kinds of knowledge remains useful from a children’s habitual attention, thus leading heuristic point of view, cognitive competence them to engage in certain processes of cate- in a subject area like mathematics seems to gorization and conceptualization. A particu- rely on the integration of well-developed larly strong impact of grammatical struc- procedural as well as conceptual knowledge. tures, such as count/mass syntax or gender grammar, has been shown by several studies. The Impact of Grammatical Language The grammatical structure that was the focus Tools on Conceptual Structures of Henrik Saalbach’s research was the Chi- From the very beginning, humans structure nese system of numeral classifiers by which their complex environment by grouping stim- nouns are grouped into linguistic categories, uli according to their similarity by various just as they are by the system of grammatical criteria. In the case of concrete objects, gender in German. Unlike gender systems, shape is a very salient feature strongly calling however, there are more than 20 classifier for categorization. Another common way of categories in Chinese that are actually used, classifying objects uses thematic relations organized around semantic features, such as which can be based on co-occurrence or in- shape, function, or animacy. For example, teraction in space or time, or functional or tiao, a common Chinese classifier for long causal relationships (e. g., carrot/rabbit, scis- and flexible things, includes many things sors/paper, cow/milk). Still another basis of from different taxonomic categories—fish, conceptual structuring already found in dogs, rivers, roads, pants, and more—and younger children concerns nonobvious taxo- even crosses the animate/inanimate bound- nomic relations. Taxonomic categories are ary (see Figure 5). differentiated into levels of varying speci- The core question was whether the classifier ficity (e. g., animal, dog, collie) related by system affects children’s concepts in any class inclusion (e. g., a collie is a dog, a dog is significant ways. According to the semantic an animal, a collie is an animal). nature of classifier categories (i. e., the fact Three kinds of tasks were used to find out that shape is an important semantic feature), whether objects are grouped on the basis of it was hypothesized that both Chinese chil- their form, or of their thematic or taxonomic dren and adults may have a stronger tenden- relations, respectively: cy than speakers of nonclassifier languages (1) In the nonlexical classification task, par- to organize objects on the basis of their ticipants were asked to select the item that shape rather than on taxonomic or thematic best matched the standard object. relations.

Center for Educational Research | 113 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 114

Figure 5. Objects for In sum, this research demonstrates that chil- which the classifier dren are extremely flexible and adaptive “tiao” is used in the Chinese language. learners who make use of various environmental cues which may help them to structure the complexity of the world. Grammati- cal language tools are among such cues and are particularly used in situations not constrained by any more salient information. The “Words serve as results suggest that deliberate language in- invitation to form put is important for young children’s environ- categories.” ment. Long before the youngsters can ac- Waxman tively use the correct grammar rules in language production, they make use of them German and Chinese speakers of three age when complex incoming information has to groups (3-year-olds, 5-year-olds, adults) be handled to form useful conceptual struc- were tested on a match-to-standard general- tures. ization paradigm in the three mentioned formats: nonlexical classification, word learning, Activities of the ENTERPRISE Group and property inference. In all task contexts, a Continued at the ETH Zürich child was shown a picture of the standard Since 2004, Elsbeth Stern has become in- object (e. g., banana), and was asked to creasingly involved in discussions on whether choose one item out of three choice items: a neuroscience can inform educational reform taxonomic item from the same super-ordi- (Stern, 2005a, 2005b). On the one hand, she nate category (grape), a shape/classifier item felt responsible for scaling down unrealistic (feather), and a thematic item (monkey). All expectations about the educational implica- objects were represented by colored line tions of neuroscience and emphasizing the drawings (see Figure 6). importance of more traditional empirical ed- Differences between German and Chinese ucational research. On the other hand, she participants were only found in the nonlexi- acknowledged that collaboration between cal classification task (see Figure 7). Chinese neuroscience and educational research pro- speakers, especially children, pay more atten- vides unique opportunities to gain insight tion to shape similarity than German speak- into the interaction between mind and brain ers, given that shape is a prominent semantic in the cultural context. In 2005, Elsbeth Stern feature in classifier categories. In the prop- and Michael Schneider headed and coordi- erty inference task, the majority of partici- nated the research program Neuroscience, In- pants referred to taxonomy relations. In the struction, and Learning (NIL) which is funded word learning task, Chinese as well as by the German Federal Ministry of Education German children judged on the basis of and Science. Meanwhile, NIL provides fund- shape, while all adults considered taxonomy ing for six research projects in which neuro- relations. scientists and educational researchers ad- Figure 6. Type of rela- dress questions of brain functioning during tionships to banana: school-related activities. thematic (monkey), In 2006, the ENTERPRISE group moved to the shape (feather), taxonomic (grape). Swiss Federal Institute of Technology Zurich (ETH Zürich), where Elsbeth Stern was appointed Professor for Research on Learning and Instruction, and where she heads the masters program for teacher students in secondary higher education. Despite this shift toward research in learning and instruction in the upper grades, the theoretical focus of

114 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 115

Figure 7. Percentage of German and Chinese 100 speakers who chose the three kinds of relation- 80 ships in the nonlexical classification task. 60

(in %) 40

0 China Germany China Germany China Germany 3 years 3 years 5 years 5 years adults adults Key References Shape Taxonomy Theme Grabner, R. H., Neubauer, A. C., & Stern, E. (2006). Supe- rior performance and the ENTERPRISE project has been maintained. neural efficiency: The The major concern will be to explore how, in impact of intelligence the field of science, learning with cognitive and expertise. Brain Research Bulletin, 69, tools may optimize the way incoming infor- 422–439. mation is connected with knowledge that Stern, E. (2005a). Brain already exists in the mind. In continuation of goes to school. [Review the research on the interaction between of the book The learning brain: Lessons for edu- intelligence and knowledge (Grabner, cation]. Trends in Cog- Neubauer, & Stern, 2006), special emphasis nitive Science, 9, will be placed on the question of how indi- 563–565. vidual differences in cognitive resources Stern, E. (2005b). Edi- affect learning in the area of science and torial: Pedagogy meets neuroscience. Science, engineering. 310, 745.

Center for Educational Research | 115 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 116

Projects of W. Edelstein, Director Emeritus

Learning and Living Democracy: A Program of Education for Democracy and Citzenship in 13 German States Around the turn of the century, in the face of right-wing extremist acts of violence and xeno- phobic attacks on individuals and refugee asylums, the Freudenberg Foundation convened a group of social scientists, experts in youth development, and educators and community workers to discuss the problems of adolescent right-wing extremism in Germany as well as strategies to deal with it with some hope of success. It soon became clear that it was essential to distinguish between police repression of violent and criminal acts, social intervention in view of a regional response to the local situation—a type of action mostly to be entrusted to social workers—and, finally, general prevention defined as a task of education for democracy in the schools. For all three domains of action, the Freudenberg group was able to marshal a number of knowledgeable experts (Edelstein, 2005a, 2005b).

For the prevention aspect, Wolfgang form, and published 97 publications (mostly Edelstein and Peter Fauser, professor of edu- materials for practical use by teachers: cational science at the University of Jena, “building blocks” for democratic schools; a undertook to develop a concept of educa- set of “Contributions to Democracy Educa- tional action in schools. Their report served tion” that provide the basis for a series to be as a basis for deliberation in the Joint Federal published by Beltz starting in 2007; and three and States’ Commission on Educational Plan- book publications, one of which was the www.blk-demokratie.de ning and Research (BLK), which decided to starting volume of the above mentioned se- implement a program of democracy or citi- ries. This volume is a major contribution to Key References zenship education ultimately involving some the development of a more substantial role de Haan, G., Edelstein, 170 schools in 13 of the 16 German States. of democracy education in German schools, W., & Eikel, A. (2007). The report specified four modules or areas of defining standards, criteria and learning op- Qualitätsrahmen Demo- action in the schools: instructional strategies portunities for citizenship education in sec- kratiepädagogik: Demokratische Hand- and interaction processes (including feedback ondary schools (de Haan, Edelstein, & Eikel, lungskompetenz fördern, and assessment modalities), projects involv- 2007). demokratische Schul- ing democratic concerns and procedures, On the face of it, the program has been ef- qualität entwickeln (Vols. 1–7). Weinheim: democratic participation and governance in fective and quite successful. In fact, we shall Beltz. the schools, and opening the school to par- have to await the evaluation results for a re- Edelstein, W. (2005a). ticipation in community life. The report was alistic appraisal of effects. There has been an The rise of a right-wing published by the Commission in the year initial formative evaluation and a summative culture among German youth. The effects of so- 2001 under the title “Demokratie lernen und evaluation of outcomes as well as of cial transformation, leben” (Learning and Living Democracy), and processes (i. e., a longitudinal sequence of identity construction, became the basic document of reference in quantitative assessments). The data will be and context. In D. B. Pillemer & S. H. White the program. (For the theoretical underpin- analyzed in the coming months. Moreover, (Eds.), Developmental nings of the report and, ultimately, of the some 30 qualitative case analyses have been psychology and social program itself see the previous Research Re- completed. A group of some 120 transfer change: Research, history and policy (pp. 314– port [2002–2004, pp. 142–143].) agents have been trained and certified as 351). Cambridge, UK: The program has now been running for democracy counsellors in a 3-year sequence Cambridge University 5 years, coming to its close on March 31, of courses involving, for each counsellor, a Press. 2007. It has been coordinated by a staff cen- set of four selected modules (out of a total of ter at the Department of Education at the 12 available modules) plus an obligatory Free University Berlin and headed by profes- module on managing democratic school de- www.degede.de sors Gerhard de Haan and Wolfgang velopment. Finally, an Association for Democ- Edelstein. The center coordinated 26 school ratic Education has been founded. Its main networks in the participating States, pro- goal is to maintain and extend the activities duced a remarkably effective internet plat- developed and orchestrated in the program,

116 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 117

help to transfer these into the practice of ferent manner, to the enhancement of the Key Reference schools, and to serve the development of a norms and values of citizenship in European Edelstein, W., & de democratic school culture that is character- schools: Human rights, democracy, social in- Haan, G. (2004). Empfehlung 5: Lern- ized by the spirit of recognition, the practice clusion, and sustainable development were to konzepte für eine of participation, the fostering of self-efficacy, become the maxims and normative criteria of zukunftsfähige Schule – and the sharing of responsibility. education in European schools (Eurydice, von Schlüsselkompeten- zen zum Curriculum. In The basic principle leading to this construc- 2005). The goals of the European Year of Citi- Heinrich-Böll-Stiftung tion is the maxim that an individual’s experi- zenship through Education 2005 converge und Bildungskommis- ence of a democratic form of life is the fun- with the goals of the program “Learning and sion der Heinrich-Böll- Stiftung (Ed.), Selbst- damental condition for the development of Living Democracy” and the policies of its suc- ständig lernen: Bildung democratic habits of the heart. Complex so- cessors in the Association for Democratic Ed- stärkt Zivilgesellschaft: cieties with highly abstract social and politi- ucation in Germany. The program shares with Sechs Empfehlungen der Bildungskommission der cal organization and formal rules of repre- OECD’s Defining and Selecting Competencies Heinrich-Böll-Stiftung sentative government require a democratic platform (DeSeCo; Rychen & Salganic 2001, (pp. 130–188). Wein- culture of the school that provides all chil- 2003) the insistence on individual autonomy heim: Beltz. dren and youth with the essential set of op- and social competence (Edelstein & de Haan, portunities to acquire these habits. Such 2004). But it also shares with the Council of habits, it is believed, are essential in order to Europe the critical assessment of the compli- obviate the disaffection with democracy that ance gap between the policy level and the is characteristic of right-wing extremism level of practice and implementation (see among adolescents but also, as many studies Bîrcea, Kerr, Mikkelsen et al., 2004). Finally, of political attitudes across Europe have however, the program described in these shown, among an increasing number of pages has been able to produce, and to pro- adults. vide to interested actors and agencies, a sys- When starting the program in 2002, the fo- tematic set of criteria, standards, competen- cus was on the situation and political per- cies, learning opportunities and evaluation spectives of youth in Germany. Only in the concepts that will serve the development of course of the program’s implementation did democratic schools beyond the confines of its coordinators realize that the European the German States that participated in the Union, the Council of Europe, and the OECD program. were all committed, each in a somewhat dif-

Center for Educational Research | 117 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 118

Publications 2005–2006 (last update: May 2007)

Artelt, C. (2006). Lernstrategien erwerbsprozesse: Eine Antwort Konzepte und empirische Evalu- in der Schule. In H. Mandl & H. auf Heiner Rindermann. Psy- ation reformpädagogischer F. Friedrich (Eds.), Handbuch chologische Rundschau, 58, Praxis (pp. 9–11). Weinheim: Lernstrategien (pp. 337–351). 118–128. Juventa. Göttingen: Hogrefe. Baumert, J., Carstensen, C. H., Becker, M., Lüdtke, O., Artelt, C. (2005). Cross-cultural & Siegle, T. (2005). Wirtschaft- Trautwein, U., & Baumert, J. approaches to measuring moti- liche, soziale und kulturelle (2006). Leistungszuwachs in vation. Educational Assessment, Lebensverhältnisse und regio- Mathematik: Evidenz für einen 10, 231–255. nale Disparitäten des Kompe- Schereneffekt im mehrgliedri- tenzerwerbs. In M. Prenzel, J. gen Schulsystem? Zeitschrift für Artelt, C., McElvany, N., Baumert, W. Blum, R. Lehmann, Pädagogische Psychologie, 20, Christmann, N., Richter, T., D. Leutner, M. Neubrand, et al. 233–242. Groeben, N., Köster, J., (Eds.), PISA 2003: Der zweite Schneider, W., Stanat, P., Oster- Vergleich der Länder in Deutsch- Becker, M., Trautwein, U., meier, C., Schiefele, U., Valtin, R., land – Was wissen und können Lüdtke, O., Cortina, K. S., & & Ring, K. (2005). Förderung von Jugendliche? (pp. 323–365). Baumert, J. (2006). Bildungs- Lesekompetenz – Eine Expertise. Münster: Waxmann. expansion und kognitive Mobi- Bonn: Bundesministerium für Baumert, J., Stanat, P., & lisierung. In A. Hadjar & R. Bildung und Forschung (BMBF). Baumert, J., & Köller, O. (2005). Watermann, R. (Eds.). (2006a). Becker (Eds.), Die Bildungs- Sozialer Hintergrund, Bildungs- Herkunftsbedingte Disparitäten expansion: Erwartete und uner- Baeriswyl, F., Wandeler, C., beteiligung und Bildungsver- im Bildungswesen: Differenzielle wartete Folgen (pp. 63–89). Trautwein, U., & Oswald, K. läufe im differenzierten Sekun- Bildungsprozesse und Probleme Wiesbaden: VS Verlag für (2006). Leistungstest, Offenheit darschulsystem. In V. Frederking, der Verteilungsgerechtigkeit: Sozialwissenschaften. von Bildungsgängen und obliga- H. Heller, & A. Scheunpflug Vertiefende Analysen im Rah- torische Beratung der Eltern: (Eds.), Nach PISA: Konsequenzen men von PISA 2000. Wiesbaden: Brunner, M., Artelt, C., Krauss, Reduziert das Deutschfreiburger für Schule und Lehrerbildung VS Verlag für Sozialwissen- S., & Baumert, J. (2007). Übergangsmodell die Effekte des nach zwei Studien (pp. 9–21). schaften. Coaching for the PISA test. sozialen Hintergrunds bei Über- Wiesbaden: VS Verlag für Learning and Instruction, 17, gangsentscheidungen? Zeit- Sozialwissenschaften. Baumert, J., Stanat, P., & 111–122. schrift für Erziehungswissen- Watermann, R. (2006b). Schul- schaft, 9, 373–392. Baumert, J., & Kunter, M. struktur und die Enstehung dif- Brunner, M., & Krauss, S. (2006). Stichwort: Professionelle ferenzieller Lern- und Entwick- (2005). Der Mathematikunter- Baltes, P. B. (2007). Foreword. Kompetenz von Lehrkräften. lungsmilieus. In J. Baumert, P. richt an der Laborschule: Em- In M. G. Ash & T. Sturm (Eds.), Zeitschrift für Erziehungswis- Stanat, & R. Watermann (Eds.), pirische Befunde zu den mathe- Psychology’s territories: Histori- senschaft, 9, 469–520. Herkunftsbedingte Disparitäten matischen Kompetenzen. In R. cal and contemporary perspec- im Bildungswesen: Differenzielle Watermann, S. Thurn, K.-J. Till- tives form different disciplines Baumert, J., & Maaz, K. Bildungsprozesse und Probleme mann, & P. Stanat (Eds.), Die (pp. IX–XIV). Mahwah, NJ: (2006). Das theoretische und der Verteilungsgerechtigkeit: Laborschule im Spiegel ihrer Erlbaum. methodische Konzept von PISA Vertiefende Analysen im Rah- PISA-Ergebnisse: Pädagogisch- zur Erfassung sozialer und kul- men von PISA 2000 (pp. 95– didaktische Konzepte und em- Baumert, J. (2006). Was wissen tureller Ressourcen der 188). Wiesbaden: VS Verlag für pirische Evaluation reformpäda- wir über die Entwicklung von Herkunftsfamilie: Interna- Sozialwissenschaften. gogischer Praxis (pp. 121–131). Schulleistungen? Pädagogik, 58 tionale und nationale Rah- Weinheim: Juventa. (4), 40–46. menkonzeption. In J. Baumert, Baumert, J., Stanat, P., & P. Stanat, & R. Watermann Watermann, R. (2006c). Vor- Brunner, M., Kunter, M., Baumert, J. (2005). Schluss- (Eds.), Herkunftsbedingte Dis- wort. In J. Baumert, P. Stanat, & Krauss, S., Baumert, J., Blum, folgerungen mit Blick auf Hoch- paritäten im Bildungswesen: R. Watermann (Eds.), Herkunfts- W., Dubberke, T., Jordan, A., schulen, Bund und Länder: Differenzielle Bildungsprozesse bedingte Disparitäten im Bil- Klusmann, U., Tsai, Y.-M., & Akzentsetzungen für die Em- und Probleme der Verteilungs- dungswesen: Differenzielle Bil- Neubrand, M. (in press). Wie er- pirische Bildungsforschung. In gerechtigkeit: Vertiefende dungsprozesse und Probleme der werben Mathematiklehrkräfte H. Mandl & B. Kopp (Eds.), Im- Analysen im Rahmen von PISA Verteilungsgerechtigkeit: Ver- ihr Fachwissen und ihr fach- pulse für die Bildungsforschung: 2000 (pp. 11–29). Wiesbaden: tiefende Analysen im Rahmen didaktisches Wissen? Zeitschrift Stand und Perspektiven: Doku- VS Verlag für Sozialwissen- von PISA 2000 (pp. 9–10). für Erziehungswissenschaft. mentation eines Experten- schaften. Wiesbaden: VS Verlag für gesprächs (pp. 127–137). Berlin: Sozialwissenschaften. Brunner, M., Kunter, M., Akademie-Verlag. Baumert, J., & Stanat, P. Krauss, S., Baumert, J., Blum, (2006). Internationale Schul- Baumert, J., & Tillmann, K.-J. W., Dubberke, T., et al. (2006). Baumert, J., Brunner, M., leistungsvergleiche. In D. H. (2005). Vorwort. In R. Water- Welche Zusammenhänge beste- Lüdtke, O., & Trautwein, U. Rost (Ed.), Handwörterbuch mann, S. Thurn, K.-J. Tillmann, & hen zwischen dem fachspezi- (2007). Was messen interna- Pädagogische Psychologie (3rd, P. Stanat (Eds.), Die Laborschule fischen Professionswissen von tionale Schulleistungsstudien? rev. ed., pp. 291–302). Wein- im Spiegel ihrer PISA-Ergebnisse: Mathematiklehrkräften und Resultate kumulativer Wissens- heim: Beltz. Pädagogisch-didaktische ihrer Ausbildung sowie beruf-

118 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 119

lichen Fortbildung? Zeitschrift Edelstein, W. (in press). Zur sozialen Entwicklung in Ganz- schaft (pp. 238–249). Münster: für Erziehungswissenschaft, 9, Genese des rechtsextremen tagsschulen. Recht der Jugend Westfälisches Dampfboot. 521–544. Potentials in Deutschland und und des Bildungswesens, 54, mögliche Antworten seitens der 3–10. Edelstein, W., & Nunner- Brunner, M., Kunter, M., Schule. In C. Petry & E. Welz Winkler, G. (Eds.). (2005). Moral- Krauss, S., Klusmann, U., (Eds.), Demokratie lernen und Edelstein, W. (2005a). The rise ity in context. Amsterdam: Baumert, J., Blum, W., et al. leben (Vol. 3). Weinheim: Beltz. of a right-wing culture among Elsevier (Advances in psychology (2006). Die professionelle Kom- German youth: The effects of 137). petenz von Mathematiklehr- Edelstein, W. (2007). Schule als social transformation, identity kräften: Konzeptualisierung, Er- Armutsfalle – wie lange noch? construction, and context. In D. Felbrich, A., Hardy, I., & Stern, fassung und Bedeutung für den Schulmagazin, 5–10 (1), 53–56. B. Pillemer & S. H. White (Eds.), E. (in press). Mathematische Unterricht: Eine Zwischenbilanz Developmental psychology and Kompetenz. In M. Hasselhorn & des COACTIV-Projekts. In M. Edelstein, W. (2006a). Bildung social change: Research, history W. Schneider (Eds.), Handbuch Prenzel & L. Allolio-Näcke (Eds.), und Armut: Der Beitrag des Bil- and policy (pp. 314–351). der Psychologie: Vol. Entwick- Untersuchungen zur Bildungs- dungssystems zur Vererbung Cambridge, UK: Cambridge Uni- lungspsychologie. Göttingen: qualität von Schule: Abschluss- und zur Bekämpfung von Armut. versity Press. Hogrefe. bericht des DFG-Schwerpunkt- Zeitschrift für Soziologie der programms (pp. 54–82). Erziehung und Sozialisation, 26, Edelstein, W. (2005b). The rise Gerlach, E., Trautwein, U., & Münster: Waxmann. 120–134. of right-wing extremist youth Lüdtke, O. (in press). Selbst- culture in postunification Ger- konzept und Effekte der sozialen Brunner, M., & Süß, H.-M. Edelstein, W. (2006b). Demo- many. In L. Nucci (Ed.), Conflict, Umwelt. In A. Conzelmann & F. (2005). Analyzing the reliability kratie lernen: Kompetenzen für contradiction, and contrarian Hänsel (Eds.), Sport und Selbst- of multidimensional measures: die Zukunft und Lehren aus der elements in moral development konzept. Schorndorf: Hofmann. An example from intelligence Vergangenheit. In A. Rogoss, E. and education (pp. 157–172). research. Educational and Psy- Hemmer, & E. Zimmer (Eds.), Mahwah, NJ: Erlbaum. Gerlach, E., Trautwein, U., & chological Measurement, 65, Georg Elser – ein Attentäter als Lüdtke, O. (2007). Referenz- 227–240. Vorbild (pp. 133–140). Bremen: Edelstein, W. (2005c). Wissen- gruppeneffekte im Sportunter- Edition Temmen. schaftsdynamik und Gesell- richt: Kurz- und langfristig Chang, P.-H. (2005). Transfor- schaftsentwicklung: Skizze eines negative Effekte von sportlichen mation of vocational secondary Edelstein, W. (2006c). Entge- Forschungsprogramms. In R. M. Klassenkameraden auf das schools: A study of the voca- genkommmende Verhältnisse – Kiesow, R. Ogorek, & S. Simitis Selbstkonzept. Zeitschrift für tional Gymnasium in Germany. Aufgaben der Ganztagsschule (Eds.), Summa: Dieter Simon zum Sozialpsychologie, 38, 73–83. Opladen: Barbara Budrich. für die Zivilgesellschaft. In Ar- 70. Geburtstag (pp. 133–140). beitsgemeinschaft für Kinder- Frankfurt a. M.: Klostermann Grabner, R. H., Neubauer, A. C., Cortina, K. S., Alfeld, C., Köller, und Jugendhilfe – AGJ (Ed.), (Studien zur europäischen & Stern, E. (2006). Superior O., Baumert, J., & Eccles, J. S. Zukunftsprojekt: Gemeinsame Rechtsgeschichte 193). performance and neural effi- (2006). Zur Bedeutung innerer Gestaltung von Lern- und ciency: The impact of intelli- und äußerer Differenzierung für Lebenswelten. Zusammenspiel Edelstein, W. (2005d). gence and expertise. Brain Re- den Bildungsverlauf – eine ver- von Kinder- und Jugendhilfe & Zukunftsfeste Kompetenzen: search Bulletin, 69, 422–439. gleichende Längsschnittanalyse Schule im Sozialraum (pp. 85– Werte für eine nachhaltige Bil- zum Hochschulübergang. In A. 93). Berlin: Arbeitsgemeinschaft dung. In A. Knauf, K. Liebers, & Grabner, R. H., Stern, E., & Ittel, L. Stecher, H. Merkens, & J. für Kinder- und Jugendhilfe. A. Prengel (Eds.), Länderüber- Neubauer, A. C. (2007). Individ- Zinnecker (Eds.), Jahrbuch Ju- greifende Curricula für die ual differences in chess expert- gendforschung (Vol. 6, pp. 15– Edelstein, W. (2006d). Extrem- Grundschule (pp. 7–11). Bad ise: A psychometric investiga- 32). Wiesbaden: VS Verlag für ist youth in Germany: The role Heilbrunn: Klinkhardt. tion. Acta Psychologica, 124, Sozialwissenschaften. of history, development, and co- 398–420. hort experience. In C. Daiute, Z. Edelstein, W., & Grundmann, Cortina, K. S., Baumert, J., Beykont, C. Higson-Smith, & L. M. (2006). The persistence of Grundmann, M., Bittlingmayer, Leschinsky, A., Mayer, K. U., & Nucci (Eds.), International per- social class: A macro-micro U. H., Dravenau, D., & Edelstein, Trommer, L. (Eds.). (in press). Das spectives on youth conflict and analytical approach to educa- W. (2006). Bildungsstrukturen Bildungswesen in der Bundes- development (pp. 43–56). tional performance and attain- und sozialstrukturelle Sozialisa- republik Deutschland. Reinbek: Oxford: Oxford University Press. ment in three generations. In tion. In M. Grundmann, D. Rowohlt. M. Grundmann, D. Dravenau, Dravenau, U. H. Bittlingmayer, & Edelstein, W. (2006e). Kompe- U. H. Bittlingmayer, & W. Edel- W. Edelstein (Eds.), Handlungs- Dieckmann, A., & Krauss, S. tenzen für die Zivilgesellschaft. stein (Eds.), Handlungsbefähi- befähigung und Milieu: Zur (2005). Wenn weniger Wissen In R. Boenicke, A. Hund, T. Rihm, gung und Milieu: Zur Analyse Analyse milieuspezifischer All- mehr sein kann: Einfache & V. Strittmatter-Haubold (Eds.), milieuspezifischer Alltagsprak- tagspraktiken und ihrer Un- Heuristiken zur psychologischen Innovativ Schule entwickeln: tiken und ihrer Ungleichheits- gleichheitsrelevanz (pp. 13–35). Entscheidungsfindung. Zeit- Kompetenzen, Praxis und Visio- relevanz (pp. 96–153). Berlin: Berlin: LIT-Verlag (Individuum schrift für Erziehungswissen- nen. 7. Heidelberger Dienstags- LIT-Verlag (Individuum und und Gesellschaft 2). schaft, 8, 187–201. seminar (pp. 13–26). Heidelberg: Gesellschaft 2). Mattes (Schriftenreihe der Grundmann, M., Dravenau, D., Edelstein, W. (in press). Demo- Pädagogischen Hochschule Edelstein, W., & de Haan, G. Bittlingmayer, U. H., & kratie als Praxis und Demokratie Heidelberg 47). (2006). Lernkonzepte für eine Edelstein, W. (Eds.). (2006). als Wert: Überlegungen zu einer zukunftsfähige Schule. In K. Handlungsbefähigung und Mi- demokratiepädagogisch aktiven Edelstein, W. (2006f). Zur För- Gerlof & A. Ulrich (Eds.), Die lieu: Zur Analyse milieuspezi- Schule. Pädagogische Führung. derung der individuellen und Verfasstheit der Wissensgesell- fischer Alltagspraktiken und

Center for Educational Research | 119 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 120

ihrer Ungleichheitsrelevanz. Klusmann, U., Trautwein, U., & Kreuter, F., Maaz, K., & Water- richt der PISA-Schülerinnen und Münster: LIT-Verlag (Individuum Lüdtke, O. (2005). Intrinsische mann, R. (2005). Messung -Schüler: Schulformunter- und Gesellschaft 2). und extrinsische Lebensziele: sozialer Herkunft in Schulleis- schiede in der Unterrichtsquali- Reliabilität und Validität einer tungsuntersuchungen: Der tät. Zeitschrift für Erziehungs- de Haan, G., Edelstein, W., & deutschen Fassung des Aspira- Zusammenhang zwischen der wissenschaft, 8, 502–520. Eikel, A. (Eds.). (2007). Quali- tions Index. Diagnostica, 51, Qualität von Schülerangaben tätsrahmen Demokratiepäda- 40–51. und den Leistungen. In U. Engel Kunter, M., Dubberke, T., gogik: Demokratische Hand- (Ed.), Bildung und soziale Un- Baumert, J., Blum, W., Brunner, lungskompetenz fördern, Köller, O., Trautwein, U., gleichheit: Methodologische und M., Jordan, A., et al. (2006). demokratische Schulqualität Cortina, K. S., & Baumert, J. strukturelle Analysen (pp. 9–35). Mathematikunterricht in den entwickeln (Vols. 1–7). (2006). Rezeptive Kompetenzen Bonn: Informationszentrum PISA-Klassen 2004: Rahmenbe- Weinheim: Beltz. in Englisch am Ende der gym- Sozialwissenschaften. dingungen, Formen und Lehr- nasialen Oberstufe: Verankerung Lernprozesse. In M. Prenzel, J. Hardy, I., Jonen, A., Möller, K., & deutscher Abiturienten am Kunter, M. (2005). Multiple Baumert, W. Blum, R. Lehmann, Stern, E. (2006). Effects of in- Gemeinsamen Europäischen Ziele im Mathematikunterricht. D. Leutner, M. Neubrand et al. structional support in construc- Referenzrahmen für die Münster: Waxmann (Pädago- (Eds.), PISA 2003: Untersuchun- tivist learning environments for Sprachen. Unterrichtswissen- gische Psychologie und Ent- gen zur Kompetenzentwicklung elementary school students' un- schaft, 34, 239–255. wicklungspsychologie 51). im Verlauf eines Schuljahres derstanding of "floating and (pp. 161–194). Münster: sinking". Journal of Educational Köller, O., Trautwein, U., Waxmann. Psychology, 98, 307–325. Lüdtke, O., & Baumert, J. (2006). Zum Zusammenspiel von Kunter, M., Klusmann, U., Hardy, I., Schneider, M., Jonen, schulischer Leistung, Selbst- Dubberke, T., Blum, W., Brun- A., Stern, E., & Möller, K. konzept und Interesse in der ner, M., Jordan, A., Krauss, S., (2005). Fostering diagrammatic gymnasialen Oberstufe. Löwen, K., Neubrand, M., & Tsai, reasoning in science education. Zeitschrift für Pädagogische Y.-M. (in press). Linking aspects Swiss Journal of Psychology, 64, Psychologie, 20, 27–39. of teacher competence to their 207–217. instruction: Results from the Krauss, S., Baumert, J., Blum, COACTIV project. In M. Prenzel Hardy, I., & Stadelhofer, B. W., Neubrand, M., Jordan, A., (Ed.), Studies on the educational (2006). Concept maps wirkungs- Brunner, M., et al. (2006). Die quality of schools: The final voll als Strukturierungshilfen Konstruktion eines Tests zum report on the DFG priority pro- einsetzen: Welche Rolle spielt fachlichen und zum fachdidak- gram. Münster: Waxmann. die Selbstkonstruktion? tischen Wissen von Mathe- Zeitschrift für Pädagogische matiklehrkräften. In Beiträge Kunter, M., Stanat, P., & Psychologie, 20, 175–187. zum Mathematikunterricht Klieme, E. (2005). Die Rolle von 2006: Vorträge auf der individuellen Eingangsvoraus- Jonkmann, K., Köller, O., & 40. Tagung für Didaktik der Kunter, M., & Baumert, J. setzungen und Gruppenmerk- Trautwein, U. (2007). Englisch- Mathematik vom 6.3. bis (2006a). Linking TIMSS to re- malen beim kooperativen Lösen leistungen am Ende der Sekun- 10.3.2006 in Osnabrück search on learning and instruc- eines Problems. In E. Klieme, D. darstufe II. In U. Trautwein, O. (pp. 319–322). Hildesheim: tion: A re-analysis of the Ger- Leutner, & J. Wirth (Eds.), Pro- Köller, R. Lehmann, & O. Lüdtke Franzbecker. man TIMSS and TIMSS video blemlösekompetenz von Schü- (Eds.), Schulleistungen von data. In S. J. Howie & T. Plomp lerinnen und Schülern: Diagnos- Abiturienten: Regionale, schul- Krauss, S., & Brunner, M. (Eds.), Contexts of learning tische Ansätze, theoretische formbezogene und soziale (2005). Der naturwissen- mathematics and science: Grundlagen und empirische Be- Disparitäten (pp. 113–142). schaftliche Unterricht an der Lessons learned from TIMSS funde der deutschen PISA-2000- Münster: Waxmann. Laborschule: Empirische Be- (pp. 335–351). London: Studie (pp. 99–115). Wiesbaden: funde zu den naturwissen- Routledge. VS Verlag für Sozialwissen- Jordan, A., Ross, N., Krauss, S., schaftlichen Kompetenzen. In R. schaften. Baumert, J., Blum, W., Watermann, S. Thurn, K.-J. Till- Kunter, M., & Baumert, J. Neubrand, M., et al. (2006). mann, & P. Stanat (Eds.), Die (2006b). Who is the expert? Lang, F. R., & Lüdtke, O. (2005). Klassifikationsschema für Laborschule im Spiegel ihrer Construct and criteria validity of Der Big Five-Ansatz der Persön- Mathematikaufgaben: Doku- PISA-Ergebnisse: Pädagogisch- student and teacher ratings of lichkeitsforschung: Instrumente mentation der Aufgabenkate- didaktische Konzepte und em- instruction. Learning Environ- und Vorgehen. In S. Schumann gorisierung im COACTIV-Projekt. pirische Evaluation reformpäda- ments Research, 9, 231–251. (Ed.), Persönlichkeit: Eine ver- Berlin: Max-Planck-Institut für gogischer Praxis (pp. 155–166). gessene Größe der empirischen Bildungsforschung (Materialien Weinheim: Juventa. Kunter, M., Baumert, J., & Sozialforschung (pp. 29–39). aus der Bildungsforschung 81). Köller, O. (in press). Effective Wiesbaden: VS Verlag für Kreuter, F., Maaz, K., & Water- classroom management and the Sozialwissenschaften. Klusmann, U., Kunter, M., mann, R. (2006). Der Zusam- development of subject-related Trautwein, U., & Baumert, J. menhang zwischen der Qualität interest. Learning and Instruc- Limbird, C. K., & Stanat, P. (2006). Lehrerbelastung und Un- von Schülerangaben zur sozia- tion. (2006a). Prädiktoren von Lese- terrichtsqualität aus der Per- len Herkunft und den Schulleis- verständnis bei Kindern spektive von Lehrenden und Ler- tungen. In K.-S. Rehberg (Ed.), Kunter, M., Brunner, M., deutscher und türkischer nenden. Zeitschrift für Pädago- Soziale Ungleichheit, kulturelle Baumert, J., Klusmann, U., Herkunftssprache: Ergebnisse gische Psychologie, 20 (3), Unterschiede (pp. 3465–3478). Krauss, S., Blum, W., et al. einer Längsschnittstudie. In A. 161–173. Frankfurt a. M.: Campus. (2005). Der Mathematikunter- Ittel & H. Mertens (Eds.),

120 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 121

Veränderungsmessung und (2007). Umgang mit fehlenden Effekte institutioneller Öffnung von PISA 2000 (pp. 31–59). Längsschnittstudien in der em- Werten in der psychologischen im Bildungssystem. Wiesbaden: Wiesbaden: VS Verlag für pirischen Erziehungswissen- Forschung: Probleme und Lö- VS Verlag für Sozialwissen- Sozialwissenschaften. schaft (pp. 93–123). Wiesbaden: sungen. Psychologische Rund- schaften. VS Verlag für Sozialwissen- schau, 58, 103–117. Maaz, K., & Waleschkowski, U. schaften. Maaz, K. (2005). Wie nutzen (2005). Erlebnispädagogik im Lüdtke, O., & Trautwein, U. Jugendliche einen Computer? Stadtteil: Entwicklungen mo- Limbird, C. K., & Stanat, P. (2007). Aggregating to the Prädiktoren vielfältiger derner Erlebnispädagogik. (2006b). Sprachförderung bei between-person level in idio- Nutzungsmöglichkeiten. In U. Deutsche Jugend, 53, 309–316. Schülerinnen und Schülern mit graphic research designs: Per- Burrmann (Ed.), Sport im Kon- Migrationshintergrund: Ansätze sonal goal research as an exam- text von Freizeitengagements Maaz, K., & Watermann, R. (in und ihre Wirksamkeit. In J. ple of the need to distinguish Jugendlicher: Aus dem Branden- press). Reproduktion oder Mo- Baumert, P. Stanat, & R. Water- between reliability and homo- burgischen Längsschnitt 1998– bilität? Zur Wirkung familiärer mann (Eds.), Herkunftsbedingte geneity. Journal of Research in 2002 (pp. 75–94). Köln: Sport & Prozessmerkmale auf die Disparitäten im Bildungswesen: Personality, 41, 230–238. Buch Strauß. Studienintention am Ende der Differenzielle Bildungsprozesse gymnasialen Oberstufe. Zeit- und Probleme der Verteilungs- Lüdtke, O., Trautwein, U., Maaz, K., & Baumert, J. (in schrift für Soziologie der gerechtigkeit: Vertiefende Ana- Kunter, M., & Baumert, J. press). Bildungserträge: Soziale Erziehung und Sozialisation. lysen im Rahmen von PISA 2000 (2006a). Analyse von Lern- und regionale Disparitäten der (pp. 257–307). Wiesbaden: VS umwelten: Ansätze zur Bestim- Bildungsbeteiligung und des Maaz, K., & Watermann, R. Verlag für Sozialwissenschaften. mung der Reliabilität und Kompetenzerwerbs. In K. S. (2006). Soziale Disparitäten im Übereinstimmung von Cortina, J. Baumert, A. Bildungssystem im Spannungs- Lüdtke, O. (2006). Persönliche Schülerwahrnehmungen. Leschinsky, K. U. Mayer, & L. feld zwischen traditioneller Ziele junger Erwachsener. Zeitschrift für Pädagogische Trommer (Eds.), Das Bildungs- Dreigliedrigkeit und institu- Münster: Waxmann (Pädago- Psychologie, 20, 85–96. wesen in der Bundesrepublik tionellen Erweiterungen. In gische Psychologie und Ent- Deutschland: Strukturen und K.-S. Rehberg (Ed.), Soziale wicklungspsychologie 53). Lüdtke, O., Trautwein, U., Entwicklungen im Überblick. Ungleichheit, kulturelle Unter- Kunter, M., & Baumert, J. Reinbek: Rowohlt. schiede (pp. 2609–2620). Frank- Lüdtke, O., Becker, M., (2006b). Reliability and agree- furt a. M.: Campus. Neumann, M., Nagy, G., ment of student ratings of the Maaz, K., & Burrmann, U. Jonkmann, K., & Trautwein, U. classroom environment: A re- (2005). Freizeitmuster Jugend- Mähler, C., & Stern, E. (2006). (2007). Durchführung und analysis of TIMSS data. Learning licher. In U. Burrmann (Ed.), Transfer. In D. H. Rost (Ed.), methodische Grundlagen. In U. Environments Research, 9, Sport im Kontext von Freizeit- Handwörterbuch Pädagogische Trautwein, O. Köller, R. Lehmann, 215–230. engagements Jugendlicher: Aus Psychologie (3rd, rev. ed., & O. Lüdtke (Eds.), Schulleistun- dem Brandenburgischen Längs- pp. 782–793). Weinheim: Beltz. gen von Abiturienten: Regionale, Lüdtke, O., Trautwein, U., schnitt 1998–2002 (pp. 33–55). schulformbezogene und soziale Schnyder, I., & Niggli, A. (2007). Köln: Sport & Buch Strauß. Marsh, H. W., Gerlach, E., Disparitäten (pp. 31–42). Simultane Analysen auf Trautwein, U., Lüdtke, O., & Münster: Waxmann. Schüler- und Klassenebene: Eine Maaz, K., Gresch, C., Köller, O., Brettschneider, W.-D. (in press). Demonstration der konfirma- & Trautwein, U. (2007). Schul- Physical self-concept and per- Lüdtke, O., & Köller, O. (2006). torischen Mehrebenen- laufbahnen, soziokulturelle formance: Tests of reciprocal ef- Mehrebenenanalyse. In D. H. Faktorenanalyse zur Analyse der Merkmale und kognitive Grund- fects model of causal ordering Rost (Ed.), Handwörterbuch Päd- Schülerwahrnehmungen am fähigkeiten. In U. Trautwein, O. in preadolescence. Child Devel- agogische Psychologie (3rd, rev. Beispiel der Hausaufgabenver- Köller, R. Lehmann, & O. Lüdtke opment. ed., pp. 469–474). Weinheim: gabe. Zeitschrift für Entwick- (Eds.), Schulleistungen von Abi- Beltz. lungspsychologie und Pädago- turienten: Regionale, schulform- Marsh, H. W., Hau, K.-T., Artelt, gische Psychologie, 39, 1–11. bezogene und soziale Dispari- C., Baumert, J., & Peschar, J. L. Lüdtke, O., Köller, O., Marsh, H. täten (pp. 43–70). Münster: (2006). OECD's brief self-report W., & Trautwein, U. (2005). Waxmann. measure of educational psychol- Teacher frame of reference and ogy's most useful affective con- the big-fish-little-pond effect. Maaz, K., Hausen, C., McElvany, structs: Cross-cultural, psycho- Contemporary Educational Psy- N., & Baumert, J. (2006). Stich- metric comparisons across 25 chology, 30, 263–285. wort: Übergänge im Bildungs- countries. International Journal system. Zeitschrift für Erzie- of Testing, 6, 311–360. Lüdtke, O., Robitzsch, A., hungswissenschaft, 9, 299–327. Trautwein, U., Kreuter, F., & Marsh, H. W., Trautwein, U., Ihme, J. M. (in press). Are there Maaz, K., Kreuter, F., & Water- Lüdtke, O., Köller, O., & test administrator effects in mann, R. (2006). Schüler als In- Baumert, J. (in press). Big Fish large-scale educational assess- formanten? Die Qualität von Little Pond Effect: Persistent ments? Using cross-classified Schülerangaben zum sozialen negative effects of selective multilevel analysis to probe for Hintergrund. In J. Baumert, P. high schools on self-concept effects on math achievement Stanat, & R. Watermann (Eds.), after graduation. American Edu- and sample attrition. Metho- Herkunftsbedingte Disparitäten cational Research Journal. dology. im Bildungswesen: Differenzielle Bildungsprozesse und Probleme Marsh, H. W., Trautwein, U., Lüdtke, O., Robitzsch, A., Maaz, K. (2006). Soziale Her- der Verteilungsgerechtigkeit: Lüdtke, O., Köller, O., & Trautwein, U., & Köller, O. kunft und Hochschulzugang: Vertiefende Analysen im Rahmen Baumert, J. (2006). Integration

Center for Educational Research | 121 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 122

of multidimensional self- (pp. 161–193). Münster: Zeitschrift für Entwicklungspsy- die Leistungsentwicklung im concept and core personality Waxmann. chologie und Pädagogische Psy- Fach Französisch. Zeitschrift für constructs: Construct validation chologie, 37, 156–167. Erziehungswissenschaft. and relations to well-being and Möller, J., & Husemann, N. achievement. Journal of Person- (2006). Internal comparisons in Nagy, G., Neumann, M., Niggli, A., Trautwein, U., ality, 74, 403–456. everyday life. Journal of Educa- Becker, M., Watermann, R., Schnyder, I., Lüdtke, O., & tional Psychology, 98, 342–353. Köller, O., Lüdtke, O., et al. Neumann, M. (2007). Elterliche Marsh, H. W., Trautwein, U., (2007). Mathematikleistungen Unterstützung kann hilfreich Lüdtke, O., Köller, O., & Müller, A. G., Gallschütz, C., am Ende der Sekundarstufe II. In sein, aber Einmischung schadet: Baumert, J. (2005). Academic Puhr, K., & Thimm, K. (2005). U. Trautwein, O. Köller, R. Familiärer Hintergrund, elter- self-concept, interest, grades, Qualitätsstandards für Schulver- Lehmann, & O. Lüdtke (Eds.), liches Hausaufgabenengage- and standardized test scores: weigererprojekte in Kooperation Schulleistungen von Abiturien- ment und Leistungsentwicklung. Reciprocal effects models of von Schule und Kinder- und Ju- ten: Regionale, schulformbezo- Psychologie in Erziehung und causal ordering. Child Develop- gendhilfe. In B. Herz, K. Puhr, & gene und soziale Disparitäten Unterricht, 54, 1–14. ment, 76, 397–416. H. Ricking (Eds.), Problem Schul- (pp. 71–112). Münster: Wax- absentismus: Wege zurück in die mann. Nunner-Winkler, G., & Edel- McElvany, N. (in press). Die Be- Schule (pp. 273–288). Bad Heil- stein, W. (2005). Introduction. deutung familiärer Merkmale brunn: Klinkhardt. Nagy, G., Trautwein, U., In W. Edelstein & G. Nunner- für Lesekompetenz, Lesemotiva- Baumert, J., Köller, O., & Winkler (Eds.), Morality in con- tion und Leseverhalten. Psy- Müller, A. G., & Stanat, P. Garrett, J. (2006). Gender and text (pp. 1–24). Amsterdam: chologie in Erziehung und Un- (2006). Schulischer Erfolg von course selection in upper sec- Elsevier (Advances in psychology terricht. Schülerinnen und Schülern mit ondary education: Effects of ac- 137). Migrationshintergrund: Analy- ademic self-concept and intrin- McElvany, N. (2007). Förderung sen zur Situation von Zuwan- sic value. Educational Research Oesterreich, D. (2005a). Demo- von Lesekompetenz im Kontext derern aus der ehemaligen So- and Evaluation, 12, 323–345. kratieerziehung in Mittel- und der Familie. Münster: Waxmann. wjetunion und aus der Türkei. In Osteuropa: Ergebnisse aus dem J. Baumert, P. Stanat, & R. Neubauer, A., & Stern, E. CIVIC-Education Projekt der IEA. McElvany, N., & Artelt, C. (in Watermann (Eds.), Herkunfts- (2007). Lernen macht intelligent: Berlin: Stiftung Erinnerung, Ver- press-a). Das Berliner Eltern- bedingte Disparitäten im Bil- Warum Begabung gefördert antwortung und Zukunft. Kind Leseprogramm: Konzeption dungswesen: Differenzielle Bil- werden muss. München: und Effekte. Psychologie in dungsprozesse und Probleme der Deutsche Verlags-Anstalt. Oesterreich, D. (2005b). Flight Erziehung und Unterricht. Verteilungsgerechtigkeit: Ver- into security: A new approach tiefende Analysen im Rahmen and measure of the authoritar- McElvany, N., & Artelt, C. (in von PISA 2000 (pp. 221–255). ian personality. Political Psy- press-b). Systematic reading Wiesbaden: VS Verlag für chology, 26, 275–297. training in the family: Develop- Sozialwissenschaften. ment, implementation, and Prenzel, M., & Baumert, J. evaluation of the Berlin Parent- Müller, A. G., & Wegener, E. (2005). Editorial (Schwerpunkt: Child Reading Program. Learn- (2005). Kooperative Zusammen- PISA-E). Zeitschrift für Erzie- ing and Instruction. arbeit in Schulverweigererpro- hungswissenschaft, 8, 485–486. jekten – Erfahrungen aus Pro- McElvany, N. & Artelt, C. jekten für schulabsente Kinder Prenzel, M., Baumert, J., Blum, (2006). Das Berliner Eltern-Kind und Jugendliche. In B. Herz, K. W., Lehmann, R., Leutner, D., Leseprogramm. Diskurs Kind- Puhr, & H. Ricking (Eds.), Pro- Neubrand, M., et al. (Eds.) heits- und Jugendforschung, 1, blem Schulabsentismus: Wege (2006). PISA 2003: Unter- 157–159. zurück in die Schule (pp. 289– suchungen zur Kompetenz- 303). Bad Heilbrunn: Klinkhardt. entwicklung im Verlauf eines McElvany, N., Artelt, C., & Schuljahres. Münster: Holler, S. (2005). Leseförderung Nagy, G., Garrett, J., Trautwein, Waxmann. in der Familie. In A. Sasse & R. U., Cortina, K. S., Baumert, J., & Valtin (Eds.), Lesen lehren Eccles, J. (in press). Gendered Neumann, M., & Nagy, G. (pp. 151–167). Berlin: Deutsche high school course selection as (2007). Naturwissenschaftliche Gesellschaft für Lesen und a precursor of gendered occupa- Grundbildung am Ende der Schreiben. tional careers: The mediating Sekundarstufe II. In U. role of self-concept and intrin- Trautwein, O. Köller, R. Lehmann, Möller, K., Hardy, I., Jonen, A., sic value. In H. M. G. Watt & & O. Lüdtke (Eds.), Schulleistun- Kleickmann, T., & Blumberg, E. J. S. Eccles (Eds.), Explaining gen von Abiturienten: Regionale, (2006). Naturwissenschaften in gendered occupational out- schulformbezogene und soziale der Primarstufe: Zur Förderung comes: Examining individual and Disparitäten (pp. 143–159). konzeptuellen Verständnisses social explanations through Münster: Waxmann. durch Unterricht und zur Wirk- school and beyond. APA Books. samkeit von Lehrerfortbildun- Neumann, M., Schnyder, I., gen. In M. Prenzel & L. Allolio- Nagy, G., Köller, O., & Heck- Trautwein, U., Niggli, A., Näcke (Eds.), Untersuchungen hausen, J. (2005). Der Übergang Lüdtke, O., & Cathomas, R. (in zur Bildungsqualität von Schule: von der Schule in die berufliche press). Schulformen als differen- Abschlussbericht des DFG- Erstausbildung: Wer die Sorgen zielle Lernmilieus: Institutionelle Schwerpunktprogramms scheut, wird von ihnen ereilt. und kompositionelle Effekte auf

122 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 123

Prenzel, M., Baumert, J., Blum, Lüdtke, O. (2006). Wer lange Herkunftsbedingte Disparitäten engagement in PISA 2003. Paris: W., Lehmann, R., Leutner, D., lernt, lernt noch lange nicht viel im Bildungswesen: Differenzielle OECD. Neubrand, M., et al. (2005). PISA mehr: Korrelate der Hausauf- Bildungsprozesse und Probleme 2003: Der zweite Vergleich der gabenzeit im Fach Französisch der Verteilungsgerechtigkeit: Stanat, P., & Lüdtke, O. (in Länder in Deutschland – Was und Effekte auf die Leistungs- Vertiefende Analysen im Rahmen press). Multilevel issues in inter- wissen und können Jugendliche? entwicklung. Psychologie in von PISA 2000 (pp. 189–219). national large-scale assessment Münster: Waxmann. Erziehung und Unterricht, 53, Wiesbaden: VS Verlag für studies on student performance. 107–121. Sozialwissenschaften. In F. J. R. van de Vijver, D. A. van Hemert, & Y. H. Poortinga (Eds.), Schümer, G. (2006). Zur bil- Stanat, P. (2005a). Jungen und Individuals and cultures in mul- dungspolitischen Bedeutung Mädchen in der Laborschule: tilevel analysis. Hillsdale, NJ: internationaler Schulleistungs- Empirische Befunde zu Erlbaum. studien. In C. Brinkmann, S. Geschlechterunterschieden. In R. Koch, & H. G. Mendius (Eds.), Watermann, S. Thurn, K.-J. Till- Stanat, P., & Lüdtke, O. (2007). Wirkungsforschung und Politik- mann, & P. Stanat (Eds.), Die Internationale Schulleistungs- beratung – eine Gratwande- Laborschule im Spiegel ihrer vergleiche. In G. Trommsdorf & rung? Fachtagung am 24./ PISA-Ergebnisse: Pädagogisch- H. J. Kornadt (Eds.), Enzyklopä- 25.2.2005 in Bonn (pp. 245– didaktische Konzepte und em- die der Psychologie: Vol. C, VII, 3. 273). Nürnberg: Institut für Ar- pirische Evaluation reformpäda- Anwendungsfelder der Kultur- beitsmarkt- und Berufsfor- gogischer Praxis (pp. 257–277). vergleichenden Psychologie schung der Bundesagentur für Weinheim: Juventa. (pp. 279–347). Göttingen: Arbeit (Beiträge zur Arbeits- Hogrefe. markt- und Berufsforschung Stanat, P. (2005b). Sprachliches 300). Lernen an der Laborschule: Em- Stanat, P., & Müller, A. G. pirische Befunde zur Lesekompe- (2005). Förderung von Schü- Schümer, G. (2005). Schule und tenz. In R. Watermann, S. Thurn, lerinnen und Schülern mit Ramm, G., Prenzel, M., Baumert, soziale Ungleichheit: Zum Um- K.-J. Tillmann, & P. Stanat (Eds.), Migrationshintergrund. In H. J., Blum, W., Lehmann, R., gang mit unterschiedlichen Die Laborschule im Spiegel ihrer Bartnitzky & A. Speck-Hamdan Leutner, D., et al. (Eds.). (2006). Lernvoraussetzungen in PISA-Ergebnisse: Pädagogisch- (Eds.), Deutsch als Zweitsprache PISA 2003: Dokumentation der Deutschland und anderen didaktische Konzepte und em- lernen (pp. 20–32). Frankfurt Erhebungsinstrumente. Münster: OECD-Ländern. Die Deutsche pirische Evaluation reformpäda- a. M.: Grundschulverband. Waxmann. Schule, 97, 266–284. gogischer Praxis (pp. 103–117). Weinheim: Juventa. Stanat, P., Müller, A. G., & Saalbach, H., & Imai, M. (2005). Schwanzer, A. D., Trautwein, U., Baumert, J. (2005). Die Koffer- Do classifier categories structure Lüdtke, O., & Sydow, H. (2005). Stanat, P., Baumert, J., & bande auf Reisen in das Land our concepts? In Proceedings Entwicklung eines Instruments Müller, A. G. (2005). Förderung der Sprache und des Theaters: of the 27th Annual Conference zur Erfassung des Selbst- von deutschen Sprachkompe- Das Jacobs Sommercamp Pro- of the Cognitive Science Society konzepts junger Erwachsener. tenzen bei Kindern aus zuge- jekt. Pädagogik, 57, 57–58. (unpaginated). Mahwah, NJ: Diagnostica, 51, 183–192. wanderten und sozial benach- Erlbaum. teiligten Familien: Evaluations- Stanat, P., & Watermann, R. Shajek, A., Lüdtke, O., & konzeption für das (2005). Schulevaluation im An- Scheunpflug, A., Baumert, J., & Stanat, P. (2006). Akademische Jacobs-Sommercamp Projekt. schluss an Schulleistungsunter- Kunter, M. (2006). Schwer- Selbstkonzepte bei Jugendlichen Zeitschrift für Pädagogik, 51, suchungen: Das Beispiel der punkt: Professionelle Kompetenz mit Migrationshintergrund. 856–875. Laborschule Bielefeld. In H. G. von Lehrkräften. Editorial. Unterrichtswissenschaft, 34, Holtappels & K. Höhmann (Eds.), Zeitschrift für Erziehungswis- 125–145. Schulentwicklung und Schul- senschaft, 9, 465–467. wirksamkeit: Systemsteuerung, Sniehotta, F. F., Nagy, G., Bildungschancen und Entwick- Schnyder, I., Niggli, A., Scholz, U., & Schwarzer, R. lung der Schule (pp. 206–222). Cathomas, R., & Trautwein, U. (2006). The role of action con- Weinheim: Juventa. (2006). Unterscheidet sich das trol in implementing intentions Hausaufgabenverhalten von during the first weeks of behav- Staudinger, U. M., & Baumert, J. Schülern aus unterschiedlichen iour change. British Journal of (2007). Bildung und Lernen jen- Klassen? Befunde einer Mehr- Social Psychology, 45, 87–106. seits der 50: Plastizität und Re- ebenenanaylse im Fach Franzö- alität. In P. Gruss (Ed.), Die Zu- sisch [Are there any differences Stanat, P. (2006a). Disparitäten kunft des Alterns: Die Antwort in homework behavior in stu- im schulischen Erfolg: For- der Wissenschaft (pp. 240–257). dents from different classroom? schungsstand zur Rolle des Mi- München: Beck. Results from a multilevel analy- grationshintergrunds. Unter- sis in French as a second lan- richtswissenschaft, 34, 98–124. Stern, E. (in press). Wodurch guage]. Revue suisse de sciences unterscheiden sich Kinder von de l’education = Schweizerische Stanat, P. (2006b). Schulleis- Erwachsenen? Neuere Ansätze Zeitschrift für Bildungswissen- tungen von Jugendlichen mit zur Erklärung der geistigen Ent- schaften, 28, 295–478. Migrationshintergrund: Die Stanat, P., & Christensen, G. wicklung im Kindesalter. Schüler. Rolle der Zusammensetzung der (2006). Where immigrant stu- Schnyder, I., Niggli, A., Schülerschaft. In J. Baumert, P. dents succeed: A comparative Stern, E. (2006a). Frühe Lern- Cathomas, R., Trautwein, U., & Stanat, & R. Watermann (Eds.), review of performance and prozesse aus kognitionspsycho-

Center for Educational Research | 123 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 124

logischer Sicht. In G. Opp, T. schaft: Ein Almanach (pp. 123– Publikationen, Internetredaktion Disparitäten. Münster: Hellbrügge, & L. Stevens (Eds.), 129). Stuttgart: Lucius & Lucius. (Bildungsreform 13). Waxmann. Kindern gerecht werden: Kontro- verse Perspektiven auf Lernen in Stern, E. (2005d). Knowledge Stern, E., & Hardy, I. (2005). der Kindheit (pp. 93–100). Bad restructuring as a powerful Anspruchsvolle Lernaufgaben. In Heilbrunn: Klinkhardt. mechanism of cognitive devel- W. Einsiedler, M. Götz, H. opment: How to lay an early Hacker, J. Kahlert, R. W. Keck, & Stern, E. (2006b). Intelligentes foundation for coneptual under- U. Sandfuchs (Eds.), Handbuch Wissen als der Schlüssel zum standing in formal domains. In Grundschulpädagogik und Können. In W. Forner & S. P. D. Tomlinson, J. Dockrell, & Grundschuldidaktik (pp. 396– Habscheid (Eds.), Sprachliche P. Winne (Eds.), Pedagogy— 402). Bad Heilbrunn: Klinkhardt. und fachliche Kompetenzen: Teaching for learning (pp. 153- Zwei Seiten eines Blattes? 169). Leicester: British Psycho- Trautwein, U. (in press). 20 Jahre Siegener Institut für logical Society (British Journal Hausaufgaben. In W. Schneider Sprachen im Beruf (pp. 15–23). of Educational Psychology & M. Hasselhorn (Eds.), Hand- Frankfurt a. M.: Lang. Monograph Series II, 3). buch der Pädagogischen Psy- chologie. Göttingen: Hogrefe. Stern, E. (2006c). Lernen: Was Stern, E. (2005e). Kognitive wissen wir über erfolgreiches Entwicklungspsychologie des Trautwein, U. (2007). The Lernen in der Schule? Päda- mathematischen Denkens. In M. homework-achievement relation gogik, 58 (1), 45–49. van Aster & J. H. Lorenz (Eds.), reconsidered: Differentiating Rechenstörungen bei Kindern: homework time, homework fre- Trautwein, U., Köller, O., Stern, E. (2006d). Was Häns- Neurowissenschaft, Psychologie, quency, and homework effort. Lehmann, R., & Lüdtke, O. chen nicht lernt, lernt Hans hin- Pädagogik (pp. 137–149). Learning and Instruction, 17, (2007b). Öffnung von Bildungs- terher: Der Erwerb geistiger Göttingen: Vandenhoeck & 372–388. wegen und Vergleichbarkeit von Kompetenzen bei Kindern und Ruprecht. Leistungsniveaus: Zusammen- Erwachsenen aus kognitions- Trautwein, U. (2005). Selbst- fassung und Ausblick. In U. psychologischer Perspektive. In Stern, E. (2005f). Optimierung ständiges Lernen an der Labor- Trautwein, O. Köller, R. Lehmann, E. Nuissl (Ed.), Vom Lernen zum durch Auslese? Umgang mit un- schule: Empirische Befunde zu & O. Lüdtke (Eds.), Schulleistun- Lehren: Lern- und Lehrforschung terschiedlichen Lernvorausset- Selbstregulation und Selbstkon- gen von Abiturienten: Regionale, für die Weiterbildung (pp. 93– zungen. Erziehung und Wissen- zept. In R. Watermann, S. Thurn, schulformbezogene und soziale 105). Bielefeld: Bertelsmann. schaft, 57, 11–13. K.-J. Tillmann, & P. Stanat (Eds.), Disparitäten (pp. 161–170). Die Laborschule im Spiegel ihrer Münster: Waxmann. Stern, E. (2006e). Wie viel Hirn Stern, E. (2005g). Pfadfinder im PISA-Ergebnisse: Pädagogisch- braucht die Schule? Chancen Neuronendschungel. Gehirn & didaktische Konzepte und em- Trautwein, U., Köller, O., und Grenzen einer neuropsy- Geist, 9, 53. pirische Evaluation reformpäda- Lehmann, R., & Lüdtke, O. chologischen Lehr-Lern- gogischer Praxis (pp. 213–224). (2007c). Öffnung von Bildungs- Forschung. In R. Caspary (Ed.), Stern, E. (2005h). Vom Gehirn Weinheim: Juventa. wegen, erreichtes Leistungs- Lernen und Gehirn: Der Weg zu zur Kultur: Mit Mathematik die niveau und Vergleichbarkeit von einer neuen Pädagogik (pp. 128– Welt verstehen. In M. Hassel- Trautwein, U., & Baeriswyl, F. Abschlüssen. In U. Trautwein, 141). Freiburg: Herder. horn, H. Marx, & W. Schneider (in press). Wenn leistungsstarke O. Köller, R. Lehmann, & O. (Eds.), Diagnostik von Mathe- Klassenkameraden ein Nachteil Lüdtke (Eds.), Schulleistungen Stern, E. (2006f). Wie viel Hirn matikleistungen (pp. 293–300). sind: Referenzgruppeneffekte von Abiturienten: Regionale, braucht die Schule? Chancen Göttingen: Hogrefe (Tests und bei Übergangsentscheidungen. schulformbezogene und soziale und Grenzen einer neuropsy- Trends: Neue Folge 4). Zeitschrift für Pädagogische Disparitäten (pp. 11–29). chologischen Lehr-Lern- Psychologie. Münster: Waxmann. Forschung. In U. Herrmann (Ed.), Stern, E., & Felbrich, A. (2006). Neurodidaktik: Grundlagen und Erziehungs- und Schulpsycholo- Trautwein, U., & Brunner, M. Trautwein, U., Köller, O., Vorschläge für gehirngerechtes gie. In K. Pawlik (Ed.), Handbuch (2005). Didaktische Prinzipien Lehmann, R., & Lüdtke, O. Lehren und Lernen (pp. 79–86). Psychologie: Wissenschaft – An- und Lehrerhandeln an der (2007d). Vorwort. In U. Weinheim: Beltz. wendung – Berufsfelder Laborschule: Empirische Be- Trautwein, O. Köller, R. Lehmann, (pp. 719–731). Heidelberg: funde zu Unterricht, Bezie- & O. Lüdtke (Eds.), Schulleistun- Stern, E. (2005a). Brain goes to Springer. hungsqualität und Elternsicht. gen von Abiturienten: Regionale, school [Review of the book The In R. Watermann, S. Thurn, K.-J. schulformbezogene und soziale learning brain: Lessons for edu- Stern, E., Felbrich, A., & Tillmann, & P. Stanat (Eds.), Die Disparitäten (pp. 7–9). Münster: cation]. Trends in Cognitive Sci- Schneider, M. (2006). Mathe- Laborschule im Spiegel ihrer Waxmann. ences, 9, 563–565. matik lernen. In D. H. Rost (Ed.), PISA-Ergebnisse: Pädagogisch- Handwörterbuch Pädagogische didaktische Konzepte und em- Trautwein, U., Köller, O., Stern, E. (2005b). Editorial: Psychologie (3rd, rev. ed., pirische Evaluation reformpäda- Lüdtke, O., & Baumert, J. Pedagogy meets neuroscience. pp. 461–469). Weinheim: Beltz. gogischer Praxis (pp. 237–244). (2005). Student tracking and Science, 310, 745. Weinheim: Juventa. the powerful effects of opt-in Stern, E., Grabner, R., & courses on self-concept: Re- Stern, E. (2005c). Je früher, Schumacher, R. (2005). Lehr- Trautwein, U., Köller, O., flected-glory effects do exist desto besser? Über Lernstrate- Lern-Forschung und Neurowis- Lehmann, R., & Lüdtke, O. after all. In H. W. Marsh, R. G. gien von Vorschulkindern. In senschaften: Erwartungen, Be- (Eds.). (2007a). Schulleistungen Craven, & D. M. McInerney Deutscher Hochschulverband funde und Forschungsperspek- von Abiturienten: Regionale, (Eds.), New frontiers for self (Ed.), Glanzlichter der Wissen- tiven. Bonn: BMBF, Referat schulformbezogene und soziale research (pp. 307–327). Green-

124 | Center for Educational Research 07_EUB_066_125.qxd 19.07.2007 15:20 Uhr Seite 125

wich, CT: Information Age Pub- Trautwein, U., Lüdtke, O., überfachlichen Kompetenzen: Tillmann, & P. Stanat (Eds.), Die lishing (International Advances Marsh, H. W., Köller, O., & Befunde national und interna- Laborschule im Spiegel ihrer in Self Research 2). Baumert, J. (2006). Tracking, tional vergleichender Analysen. PISA-Ergebnisse. Pädagogisch- grading, and student motiva- In J. Baumert, P. Stanat, & R. didaktische Konzepte und em- Trautwein, U., & Lüdtke, O. (in tion: Using group composition Watermann (Eds.), Herkunfts- pirische Evaluation reformpäda- press-a). Epistemological beliefs, and status to predict self-con- bedingte Disparitäten im Bil- gogischer Praxis (pp. 73–88). school achievement, and aca- cept and interest in ninth-grade dungswesen: Differenzielle Bil- Weinheim: Juventa. demic choices: A large-scale, mathematics. Journal of Educa- dungsprozesse und Probleme der longitudinal study on the im- tional Psychology, 98, 788–806. Verteilungsgerechtigkeit: Vertie- Watermann, R., & Nagy, G. pact of the certainty dimension. fende Analysen im Rahmen von (2006). Egalitäre Orientierungen Contemporary Educational Psy- Trautwein, U., Lüdtke, O., PISA 2000 (pp. 61–94). Wies- und Geschlecht: Ergebnisse chology. Schnyder, I., & Niggli, A. (2006). baden: VS Verlag für Sozialwis- einer Längsschnittstudie. In Predicting homework effort: senschaften. A. Ittel & H. Mertens (Eds.), Trautwein, U., & Lüdtke, O. (in Support for a domain-specific, Veränderungsmessung und press-b). Predicting global and multilevel homework model. Watermann, R., & Brunner, M. Längsschnittstudien in der topic-specific certainty beliefs: Journal of Educational Psychol- (2005). Durchführung und empirischen Erziehungswissen- Domain-specifity and the role of ogy, 98, 438–456. methodische Grundlagen der schaft (pp. 125–146). Wies- the academic environment. Untersuchung an der Labor- baden: VS Verlag für Sozial- British Journal of Educational Trautwein, U., & Maaz, K. schule Bielefeld. In R. wissenschaften. Psychology. (2006). Soziale Herkunft und Watermann, S. Thurn, K.-J. Till- Bildungserfolg: Die Rolle beruf- mann, & P. Stanat (Eds.), Die Watermann, R., & Stanat, P. Trautwein, U., & Lüdtke, O. licher Gymnasien. Schulman- Laborschule im Spiegel ihrer (2005). Schulentwicklung und (2007). Students' self-reported agement, 37, 29–31. PISA-Ergebnisse: Pädagogisch- multiple Zielerreichung in der effort and time on homework in didaktische Konzepte und em- Laborschule: Zusammenfassung six school subjects: Between- Trautwein, U., Maaz, K., pirische Evaluation reformpäda- und Diskussion der Ergebnisse. student differences and within- Lüdtke, O., Nagy, G., Huse- gogischer Praxis (pp. 37–47). In R. Watermann, S. Thurn, K.-J. student variation. Journal of mann, N., Watermann, R., Weinheim: Juventa. Tillmann, & P. Stanat (Eds.), Die Educational Psychology, 99, et al. (2006). Studieren an der Laborschule im Spiegel ihrer 432–444. Berufsakademie oder an der Watermann, R., & Maaz, K. (in PISA-Ergebnisse: Pädagogisch- Universität, Fachhochschule press). Der Übergang von der didaktische Konzepte und em- Trautwein, U., & Lüdtke, O. oder Pädagogischen Hoch- Schule in die Hochschule: Eine pirische Evaluation reformpäda- (2005). The Big-Fish-Little-Pond schule: Ein Vergleich des Leis- Anwendung der Theorie des gogischer Praxis (pp. 285–295). effect: Future research questions tungsstands, familären Hinter- geplanten Verhaltens. In A. Weinheim: Juventa. and educational implications. grunds, beruflicher Interessen Schellack & S. Große (Eds.), Bil- Zeitschrift für Pädagogische Psy- und der Studienwahlmotive von dungswege: Aufgaben für die Watermann, R., Thurn, S., chologie, 19, 137–140. (künftigen) Studierenden aus Wissenschaft – Herausforderun- Tillmann, K.-J., & Stanat, P. Baden-Württemberg. Zeitschrift gen für die Politik. Münster: (Eds.). (2005a). Die Laborschule Trautwein, U., Lüdtke, O., & für Erziehungswissenschaft, 9, Waxmann. im Spiegel ihrer PISA-Ergebnisse: Husemann, N. (2006). Die Qua- 393–412. Pädagogisch-didaktische Kon- lität der Studienvorbereitung in Watermann, R., & Maaz, K. zepte und empirische Evaluation der gymnasialen Oberstufe: Eine Vetter, P. H., Krauss, S., Steiner, (2006a). Effekte der Öffnung reformpädagogischer Praxis. Längsschnittstudie mit Absol- O., Moises, H. W., Kropp, P., von Wegen zur Hochschulreife Weinheim: Juventa. venten von allgemeinbildenden Möller, W. D., et al. (in press). auf die Studienintention am und beruflichen Gymnasien. In Vascular dementia: Do patients Ende der gymnasialen Ober- Watermann, R., Thurn, S., A. Ittel, L. Stecher, H. Merkens, and family caregivers make ade- stufe. Zeitschrift für Erziehungs- Tillmann, K.-J., & Stanat, P. & J. Zinnecker (Eds.), Jahrbuch quate use of available supports? wissenschaft, 9, 219–239. (2005b). Versuchsschule und Jugendforschung (Vol. 6, Aging and Mental Health. Evaluationsforschung im Dialog. pp. 47–67). Wiesbaden: VS Ver- Watermann, R., & Maaz, K. In R. Watermann, S. Thurn, K.-J. lag für Sozialwissenschaften. Watermann, R. (2005). Die (2006b). Soziale Herkunft und Tillmann, & P. Stanat (Eds.), Die Laborschule als "polis", als Ver- Studienintention am Ende der Laborschule im Spiegel ihrer Trautwein, U., Lüdtke, O., antwortungsgemeinschaft: Em- gymnasialen Oberstufe: Institu- PISA-Ergebnisse: Pädagogisch- Kastens, C., & Köller, O. (2006). pirische Befunde zur politischen tionelle und individuelle Ein- didaktische Konzepte und em- Effort on homework in grades Sozialisation. In R. Watermann, flussgrößen. In W. Georg (Ed.), pirische Evaluation reformpäda- 5–9: Development, motivational S. Thurn, K.-J. Tillmann, & P. Soziale Ungleichheit im Bil- gogischer Praxis (pp. 15–23). antecedents, and the associa- Stanat (Eds.), Die Laborschule im dungssystem: Eine empirisch- Weinheim: Juventa. tion with effort on classwork. Spiegel ihrer PISA-Ergebnisse: theoretische Bestandsaufnahme Child Development, 77, Pädagogisch-didaktische Kon- (pp. 227–263). Konstanz: UVK Zimmer, K., Brunner, M., Lüdtke, 1094–1111. zepte und empirische Evaluation Verlagsgesellschaft. O., Prenzel, M., & Baumert, J. (in reformpädagogischer Praxis press). Die PISA-Spitzengruppe in Trautwein, U., Lüdtke, O., (pp. 189–200). Weinheim: Watermann, R., Maaz, K., & Deutschland: Eine Charakteri- Köller, O., & Baumert, J. (2006). Juventa. Brunner, M. (2005). Die sierung hochkompetenter Ju- Self-esteem, academic self-con- Schülerschaft an der Labor- gendlicher. In K. A. Heller & A. cept, and achievement: How the Watermann, R., & Baumert, J. schule: Empirische Befunde zum Ziegler (Eds.), Talentförderung – learning environment moderates (2006). Entwicklung eines soziokulturellen Hintergrund Expertiseentwicklung – Leis- the dynamics of self-concept. Strukturmodells zum Zusam- und zu den Eingangsvorausset- tungsexzellenz: Vol. 1. Begabt Journal of Personality and Social menhang zwischen sozialer zungen des PISA-Jahrgangs. In sein in Deutschland. Münster: Psychology, 90, 334–349. Herkunft und fachlichen und R. Watermann, S. Thurn, K.-J. LIT-Verlag.

Center for Educational Research | 125 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 126 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 127

Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 128

Contents

Introductory Overview ...... 129

Research Project 1 Intra-Person Dynamics Across the Lifespan ...... 136

Research Project 2 Sensorimotor-Cognitive Couplings ...... 147

Research Project 3 The Berlin Aging Study (BASE): Dynamic Modeling of Longitudinal Change and Mortality ...... 151

Research Project 4 Developmental Reguation of Affect, Motivation, and Abilities (DRAMA) ...... 156

Research Project 5 Interactive Brains, Social Minds ...... 162

Research Project 6 Formal Methods in Lifespan Psychology ...... 166

Research Project 7 Toward a Psychological and Developmental Theory of Sehnsucht (Life Longings) ...... 171

Publications 2005–2006 ...... 175

Scientific Staff (2005–2006)

Julia A. M. Delius, Imke Kruse, Shu-Chen Li, Ulman Lindenberger, Martin Lövdén (Inter- national Research Fellow), Timo von Oertzen, Viktor Müller, Michaela Riediger, Florian Schmiedek (as of 2006: Adjunct Research Scientist, Humboldt University Berlin), Jacqui Smith (as of 2006: University of Michigan, Ann Arbor)

Postdoctoral Research Fellows Christian Chicherio, Natalie C. Ebner, Denis Gerstorf (as of 2005: University of Virginia), Franziska Kopp, Sabine Schäfer, Susanne Scheibe

Predoctoral Research Fellows Karen Bartling, Yvonne Brehmer (LIFE) (as of 2006: Karolinska Institute, Stockholm), Annette Brose (LIFE), Agnieszka Zofia Burzynska (LIFE), Daniel Grühn (as of 2006: University of Geneva), Dorothea Hämmerer, Oliver Huxhold (LIFE), Anna Kleinspehn, Dana Kotter-Grühn (LIFE), Peter Mohr (LIFE), Irene E. Nagel (LIFE), Antje Rauers, Christina Röcke (LIFE) (as of 2006: Brandeis University, Boston), Michael Schellenbach, Yee Lee Shing (LIFE), Julius Verrel (LIFE), Markus Werkle-Bergner (LIFE))

Visiting Scientists (3 months or longer) Christopher Hertzog, Fredda Blanchard-Fields (Georgia Institute of Technology, Atlanta) Lars Bäckman, Agneta Herlitz (Karolinska Institute, Stockholm) Naftali Raz (Wayne State University, Detroit)

128 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 129

But … its eminent modifiability, and its predisposition to self-initiated action, may it develop little or much, and may it differ in amount between different individuals, is among the immutable features of humankind, which can be found wherever humans exist. Johann Nicolaus Tetens, 1777, I, p. 766

Introductory Overview

Founded in 1981 by the late Paul B. Baltes, the Center for Lifespan Psychology at the Max Planck Institute for Human Development has helped to establish lifespan psychology as a distinct conceptual approach within developmental psychology. Recently, the Center has ex- J. N. Tetens (1736–1807), tended its research program into developmental behavioral neuroscience. Work at the Center philosopher of the Enlightenment Era is guided by three propositions: (i) to study lifespan changes in behavior as interactions among maturation, learning, and senescence; (ii) to develop theories and methods that integrate empirical evidence across domains of functioning, timescales, as well as behavioral and neuronal levels of analysis; (iii) to identify mechanisms of development by exploring age- graded differences in plasticity. While the Center continues to pay special attention to the age periods of late adulthood and old age, which offer unique opportunities for innovation, both in theory and practice, it also has intensified its interest in early periods of ontogeny including infancy and early childhood.

Three Guiding Propositions to old age. By identifying the commonalities, Key References For more than two decades, the Center for differences, and interrelations in the on- Lindenberger, U., Li, Lifespan Psychology has promoted a perspec- togeny of sensation, motor control, cognition, S.-C., Lövdén, M., & Schmiedek, F. (in press). tive on behavioral development that seeks to affect, and motivation, both within and The Center for Lifespan integrate age periods, domains of function- across individuals, developmental psycholo- Psychology at the Max ing, timescales, and levels of analysis. In part gists attempt to arrive at more or less com- Planck Institute for Hu- man Development: through these efforts, lifespan psychology prehensive theories of behavioral develop- Overview of conceptual has evolved into a distinct conceptual ap- ment. To provide explanations that qualify as agenda and illustration proach within developmental psychology psychological and developmental, the effects of research activities. International Journal of (e.g., Baltes, 1987; Baltes, Lindenberger, & of external influences, such as parents’ affect Psychology. Staudinger, 2006). The Center’s current re- attunement, teachers’ classroom behavior, or Baltes, P. B., Linden- search agenda can be summarized by three a state’s retirement policies, need to be berger, U., & Staudinger, interrelated theoretical propositions that re- mapped onto mechanisms and organizational U. M. (2006). Life span flect this tradition. In line with the general laws that operate and evolve within the de- theory in developmental psychology. In W. Damon tenets of lifespan psychology, the three veloping person. Hence, as John Nesselroade, & R. M. Lerner (Eds.), propositions emphasize conceptual and Peter Molenaar, and others have emphasized, Handbook of child psy- methodological issues in the study of lifespan developing individuals, rather than groups of chology: Vol. 1. Theoret- ical models of human behavioral development, and thereby provide individuals or domains of functioning within development (pp. 569– a general script for formulating research individuals, form the privileged system of 664). New York: Wiley. questions in more specific domains of analysis and explanation. Baltes, P. B. (1987). The- interest. Individuals organize their exchange with the oretical propositions of lifespan developmental physical and social environment through be- psychology: On the dy- Proposition 1: Lifespan Changes in the havior (see Figure 1). On the one hand, the namics between growth Individual’s Behavior as Interactions changing brain and the changing physical and decline. Develop- mental Psychology, 23, Among Maturation, Learning, and and cultural environment shape behavioral 611–626. Senescence development. On the other hand, behavior The general goal of developmental psychol- alters both the brain and the environment. ogy is to identify mechanisms that generate Hence, environment and brain act both as invariance and variability, constancy and antecedents and consequents of moment-to- change in behavioral repertoires from infancy moment variability and long-term changes in

Center for Lifespan Psychology | 129 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 130

Figure 1. Environment and brain as antecedents and consequents of moment-to- moment variability and long-term changes in patterns of behavior. Lifespan changes in brain-behavior map- Cultural and pings are shaped by in- physical environment teractions among processes related to maturation, learning, and senescence. The identification of key players in the ontogeny of brain-behavior dynamics requires a coalition between formal tools for synthesis Behavior across levels of analysis and timescales as well as empirical methods for studying variability and change in brain and behavior (adapted from Momentary Lindenberger, Li, & Bäckman, 2006b). process dynamics Brain

Neuronal function

Neuroanatomy

Neurochemistry

Lifespan differences/change across years

patterns of behavior. The components of this take place without maturation. Similarly, the system, brain, behavior, and environment, are ways in which senescence takes its toll on constantly coupled and cannot be reduced the brains of aging individuals depend on in- onto each other, as they jointly condition an dividuals’ past and present learning and mat- individual’s life trajectory through recursive urational histories. To complicate matters self-regulation. even more, processes commonly associated In attempts to explain the age-graded evolu- with maturation are not confined to early tion of this system, maturation and senes- ontogeny, and processes related to senescence denote the operation of age-graded cence are not restricted to old and very old brain mechanisms and their effects on age (Raz et al., 2005). For instance, neuro- changes in behavior, which are especially genesis and synaptogenesis, as expressions of pronounced early and late in life. In addition, maturation, continue to exist in the adult learning, at any point during ontogeny, de- and aging brain, and declines in dopaminer- notes changes in brain states induced by be- gic neuromodulation, which indicate senes- havior-environment interactions. Note, how- cence-related changes in brain chemistry, ever, that maturation cannot take place commence in early adulthood (Bäckman, without learning, and that learning cannot Nyberg, Lindenberger, Li, & Farde, 2006).

130 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 131

Thus, maturation, senescence, and learning In this endeavor, psychologists occupy a cen- mutually enrich and constrain each other tral position because they possess a rich and throughout the lifespan, and must be under- adequate repertoire of conceptual, experi- stood and studied as interacting forces driv- mental, and methodological tools to describe ing the brain-behavior-environment system. and modify the organization of behavior.

Brain-Behavior Dynamics Across the Lifespan Neuroscience & Biobehavioral Reviews, 30 (6), 2006 Special Issue Edited by Ulman Lindenberger, Shu-Chen Li, and Lars Bäckman

1. Editorial: Delineating Brain-Behavior Mappings Across the Lifespan: Substantive and Methodological Advances in Developmental Neuroscience Ulman Lindenberger, Shu-Chen Li, and Lars Bäckman

Anatomical Changes 2. Brain Development in Children and Adolescents: Insights From Anatomical Magnetic Resonance Imaging Rhoshel K. Lenroot and Jay N. Giedd 3. Differential Aging of the Brain: Patterns, Cognitive Correlates and Modifiers Naftali Raz and Karen M. Rodrigue 4. Diffusion Tensor Imaging and Aging Edith V. Sullivan and Adolf Pfefferbaum 5. Advances in White Matter Imaging: A Review of In Vivo Magnetic Resonance Methodologies and Their Applicability to the Study of Development and Aging Jeffrey R. Wozniak and Kelvin O. Lim

Neurochemical Changes 6. Neuromodulation of Associative and Organizational Plasticity Across the Life Span: Empirical Evidence and Neurocomputational Modeling Shu-Chen Li, Yvonne Bremer, Yee Lee Shing, Markus Werkle-Bergner, and Ulman Lindenberger 7. The Correlative Triad Among Aging, Dopamine, and Cognition: Current Status and Future Prospects Lars Bäckman, Lars Nyberg, Ulman Lindenberger, Shu-Chen Li, and Lars Farde

Functional Dynamics 8. Imaging Human EEG Dynamics Using Independent Component Analysis Julie Onton, Marissa Westerfield, Jeanne Townsend, and Scott Makeig 9. Simultaneous EEG-fMRI Petra Ritter and Arno Villringer

Lifespan Development of Brain-Behavior Dynamics: Illustrations 10. Cortical EEG Correlates of Successful Memory Encoding: Implications for Lifespan Comparisons Markus Werkle-Bergner, Viktor Müller, Shu-Chen Li, and Ulman Lindenberger 11. The Neuronal Basis and Ontogeny of Empathy and Mind Reading: Review of Literature and Implications for Future Research Tania Singer 12. Cognitive and Neural Plasticity in Aging: General and Task-Specific Limitations Sari Jones, Lars Nyberg, Johan Sandblom, Anna Stigsdotter Neely, Martin Ingvar, Karl Magnus Petersson, and Lars Bäckman 13. SWIFT Explorations of Age Differences in Eye Movements During Reading Jochen Laubrock, Reinhold Kliegl, and Ralf Engbert 14. Brain-Behavior Relations Across the Lifespan: A Commentary Fergus I. M. Craik

Center for Lifespan Psychology | 131 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 132

Key References Proposition 2: Lifespan Theory and tion or of emotion regulation and motiva- Lindenberger, U., & von Methodology Need to Integrate Evidence tional states. Empirically, dense multivariate Oertzen, T. (2006). Vari- Across Domains of Functioning, time-series data are needed to assess short- ability in cognitive term variability and long-term changes in aging: From taxonomy to Timescales, and Levels of Analysis theory. In E. Bialystok & If the lifespan development of behavior is de- across-domain dependencies. F. I. M. Craik (Eds.), Life- fined to originate from recursive interactions Second, there is a need to understand the span cognition: Mecha- among maturation, learning, and senescence, mechanisms that link short-term variations nisms of change (pp. 297–314). Oxford: with the developing individual as the privi- to long-term change (Lindenberger & von Oxford University Press. leged system of analysis, then developmental Oertzen, 2006). Short-term variations are Li, S.-C. (2003). Biocul- psychology is faced with three difficult inte- often reversible and transient, whereas long- tural orchestration of grative tasks. First, there is the need to inte- term changes are often cumulative, progres- developmental plasticity across levels: The inter- grate theorizing and research practice across sive, and permanent. Establishing links be- play of biology and cul- functional domains to attain a comprehen- tween short-term variations and long-term ture in shaping the mind sive picture of individual development. For changes is of eminent heuristic value, as and behavior across the lifespan. Psychological instance, sensorimotor and cognitive func- it helps to identify mechanisms that drive Bulletin, 129, 171–194. tioning are more interdependent in early development into different directions. For childhood and old age than during middle instance, aging cognitive systems show an portions of the lifespan, and developmental increase in maladaptive moment-to-moment changes in either domain are better under- fluctuations or decrease in processing ro- stood if studied in conjunction (Schäfer, bustness, at both behavioral and neuronal Huxhold, & Lindenberger, 2006). Similar ob- levels of analysis. These maladaptive changes servations can be made for many other do- may signal impending long-term changes in mains of functioning whose changes gener- other characteristics of the system (see ally have been studied in isolation, such as Figure 2; Lövdén, Li, Shing, & Lindenberger, in the ontogeny of social interaction and cogni- press). In contrast, other forms of moment-

Person 2 Performance

Person 1

Moment-to-moment variability Change (seconds) (years)

Figure 2. Example for predictions linking moment-to-moment variability to long-term change, and brain changes to behavioral changes. Senescent changes in neuromodulation lead to greater moment-to-moment fluctuations in neural signaling, enhance the prominence of background noise, reduce the distinctiveness of processing pathways and representations, and increase variability of cognitive performance. Aging individuals with greater moment-to-moment process fluctuations at a given point in time are expected to show greater subsequent longitudinal decline in mean levels of functioning than individuals who fluctuate less. Recent empirical evidence supports this prediction (Lövdén, Li et al., in press; adapted from Lindenberger, Li, & Bäckman, 2006b).

132 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 133

to-moment variability indicate an individual’s coefficient modeling (RCM), latent growth ability to bring a wide variety of different curve modeling (LGCM), and related statisti- strategies to the task, and are positively re- cal techniques have served as versatile tools lated to long-term change in both childhood for the analysis of multivariate data with and old age. nested time structures, such as trials, blocks Third, to arrive at mechanistic explanations of trials, days, weeks, and years. Dynamic ex- of behavioral change, there is the need to in- tensions of these methods, such as the dual- tegrate behavioral and neuronal levels of change score model, introduced by John J. analysis. At any given point in the lifespan, McArdle and Fumiaki Hamagami, permit the one-to-one mappings between brain states investigation of directed lead-lag hypotheses and behavioral states are the exception with longitudinal panel data (for examples, rather than the rule, as the brain generally see Gerstorf, Ram, Röcke, Lindenberger, & offers more than one implementation of an Smith, in press; Lövdén, Li et al., in press; adaptive behavioral outcome (Li, 2003). Project 3, Berlin Aging Study). Second, neuro- Therefore, ontogenetic changes in behavioral computational modeling, such as the neuro- repertoires are accompanied by continuous computational theory of cognitive aging pro- changes in multiple brain-behavior mappings. posed by Shu-Chen Li and colleagues (Li, Some of these re-mapping gradients may be Lindenberger, & Sikström, 2001; Li, von relatively universal and age-graded, whereas Oertzen, & Lindenberger, 2006), has facili- others may be more variable, reflecting ge- tated conceptual integration of empirical netic differences, person-specific learning findings and concepts from a wide range of histories, the path-dependent nature of de- behavioral and neuronal research traditions, velopmental dynamics, or a combination of and provides a theoretical basis for major the three. The resulting picture underscores portions of the Center’s research program. the diversity and malleability of the organization of brain and behavior as well as the Proposition 3: The Exploration of Age- Key Reference constraints on diversity and malleability Graded Differences in Behavioral Plasticity Brehmer, Y., Li, S.-C., brought about by (a) universal age-graded is a Powerful Tool for Identifying Müller, V., von Oertzen, mechanisms associated with maturation and T., & Lindenberger, U. Mechanisms of Development (2007). Memory plastic- senescence, (b) general laws of neuronal and Behavioral plasticity or the alteration of de- ity across the lifespan: behavioral organization, and (c) cultural- velopmental trajectories through experience Uncovering children's la- social as well as physical regularities of the is a precious phenomenon (Figure 3). This tent potential. Develop- mental Psychology, 43, environment (Baltes, Lindenberger, & statement holds both from scientific and so- 465–478. Staudinger, 2006). cietal perspectives. Scientifically, inquiries In summary, developmental psychology Figure 3. Lifespan devel- needs theory and Lifespan psychology in interdisciplinary context: opment as biocultural methodology apt to Linking levels of plasticity co-construction. A cen- integrate (a) multiple tral goal of lifespan psychology is to describe, domains of function- explain, and optimize ing, (b) multiple Behavioral human potential and to plasticity timescales, and identify its societal and neural causes and conse- (c) multiple levels of quences. analysis. In recent Development years, the Center for Lifespan Psychology Neural and Societal genetic plasticity plasticity has relied on two methodologies that seem well-suited to Linkage requires: Cross-level research methods Co-constructive theories and models these conceptual de- Joint use of principles of distal and proximal causation mands. First, random

Center for Lifespan Psychology | 133 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 134

into the plasticity of human behavior are a Overview of Research Projects at the rich source of developmental information. Center for Lifespan Psychology Through the assessment of “changes in The empirical and conceptual work at the change,” they offer the promise to observe Center for Lifespan Psychology is currently the operation and proximal consequences of structured into seven research projects or developmental mechanisms. In particular, teams of scientific investigators (see Table 1). cognitive intervention studies, in which re- The research activities pursued in these proj- search participants of different ages are in- ects cover a wide array of developmental structed and trained to perform one or more topics. To provide a few examples for illustra- cognitive tasks, come with important validity tion, recent studies have addressed the fol- benefits, such as (a) an increase in experi- lowing questions: (a) How do relations be- mental control; (b) the identification of age tween body and mind change from childhood differences near asymptotic performance lev- to adulthood, and from adulthood to old age? els; and (c) the assessment of transfer and (b) How and why do functions, such as intel- maintenance effects. If neurochemical, ligence and memory, vary within and across neuroanatomical, or neurofunctional imaging individuals, and how and why do they change measures are assessed before, during, and with age? (c) How and to what end do indi- after training, intervention studies also offer viduals acquire and maintain a sense of per- new insights into relations between behav- sonal control, and how do they plan and ioral and neuronal levels of plasticity. Thus, manage their lives? (d) How do aging individ- by partly taking control over behavior- uals cope with rapid technological change, environment interactions, the mechanisms of and how can human engineering technolo- learning can be studied in the context of gies facilitate the transition to old age? maturation and senescence. When longitudi- (e) How do young children learn to coordi- nal information is available, intervention nate their behavior with others, and how studies bridge the gap between short-term does interpersonal action coordination affect alterations in performance and long-term social and cognitive development? developmental trajectories (e.g., Lövdén, Li et al., in press). Sofja-Kovalevskaja Award From the larger perspective of societal evolu- In 2006, Martin Lövdén received the Sofja tion, cognitive intervention studies explore Kovalevskaja Award of the Alexander von the range of possible development or what Humboldt Foundation. Financed by the could be possible, in principle, if conditions Federal Ministry for Education and Research, were different. The resulting knowledge the one million Euro award enables young about the plasticity of developmental trajec- scientists from outside Germany to finance tories is essential for improving human wel- their own research groups at a German uni- fare. Hence, investigations of age changes in versity or nonuniversity research institution the plasticity of development carry the of their choice. Martin Lövdén’s independent potential to explain and ameliorate human research group will be affiliated with the development. For these reasons, age-compar- Center for Lifespan Psychology. The group ative intervention studies with focus on will conduct a series of extensive training behavioral and neuronal manifestations of studies using magnetic resonance imaging to plasticity are the foundation stone of em- examine training-induced changes in behav- pirical research at the Center for Lifespan ior and brain among younger and older adults Psychology. in the functional domains of spatial navigation and working memory.

134 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 135

Table 1 The Center for Lifespan Psychology: Overview of Research Projects

Name of the project Scientific investigators Predoctoral research fellows

Intra-Person Dynamics Shu-Chen Li**; Christian Chicherio*, Hauke Yvonne Brehmer, Annette Brose, Dorothea Hämmerer, Across the Lifespan Heekeren, Martin Lövdén, Viktor Müller, Florian Oliver Huxhold, Irene Nagel, Christina Röcke, Yee Lee Schmiedek, Timo von Oertzen, Ulman Shing, Markus Werkle-Bergner Lindenberger Sensorimotor-Cognitive Martin Lövdén**; Oliver Huxhold*; Sabine Michael Schellenbach, Julius Verrel Couplings Schäfer*, Ulman Lindenberger The Berlin Aging Study Ulman Lindenberger**, Jacqui Smith**; Julia Daniel Grühn (BASE) Delius, Denis Gerstorf*, Shu-Chen Li, Martin Lövdén Developmental Regulation Michaela Riediger**; Natalie Ebner*, Sabine Antje Rauers of Affect, Motivation, and Schäfer*, Ulman Lindenberger Abilities (DRAMA) Interactive Brains, Ulman Lindenberger**; Franziska Kopp*, Shu- Karen Bartling, Anna Kleinspehn Social Minds Chen Li, Viktor Müller, Michaela Riediger Formal Methods and Theory Timo von Oertzen**; Shu-Chen Li, Ulman N.N. in Lifespan Psychology Lindenberger Toward a Psychological † Paul Baltes**; Susanne Scheibe* Dana Kotter-Grühn and Developmental Theory of Sehnsucht (life-longings)

Note. The table refers to projects and project members as of Fall 2006; for updates, visit www.mpib-berlin.mpg.de ** Principal investigator(s); * Postdoctoral fellow.

Center for Lifespan Psychology | 135 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 136

Scientific Research Project 1 Investigators Intra-Person Dynamics Across the Lifespan

Shu-Chen Li All is flux; nothing stays still. Hauke Heekeren Heraclite, ca. 500 B.C. (also Head of the Independent Junior Research Group, Conceptual Overview see p. 180) Behavioral development comprises both short-term variability and long-term change, and is Martin Lövdén embedded in cultural, environmental, and different levels of neuronal contexts. Environment Viktor Müller and brain both act as antecedents and consequences of moment-to-moment variability and Florian Schmiedek long-term changes in behavior. The overarching objective of this project is to explore theories (until 11/2006) and research designs that articulate behavioral development across timescales, levels of Jacqui Smith analysis, and functional domains. Empirically, this emphasis on integration across timescales, (until 09/2006) domains, and levels requires a drastic increase in observation density within individuals. Timo von Oertzen Ulman Lindenberger Intra-Person Dynamics: Forms and individual’s ability to regain earlier func- Functions tional levels after perturbations arising from Florian Schmiedek, In examining relations between short-term either internal processing fluctuations (e.g., Department of Psy- variability and long-term age changes or attention slips) or changes in the external chology, Humboldt age differences, different forms and func- environment (e.g., more demanding tasks). University Berlin tions of variability can be set apart. Specifi- Processing fluctuation, or lack of processing (12/2006) (adjunct cally, one may distinguish among plasticity, robustness (e.g., Lövdén, Li et al., in press; Li, research scientist) diversity, adaptability, fluctuation, and tem- Lindenberger et al., 2004), reflects stochas- poral coupling (see Figure 1 and Table 1). tic fluctuations around a modal response. Christian Chicherio Plasticity, in this context, refers to various Processing fluctuations can be observed (postdoctoral fellow) forms of progressive performance alter- more easily when the functional value of ations, such as learning induced by instruc- other forms of variability is low to begin Yvonne Brehmer tion, practice, and training. Diversity refers with, as is often the case for standard reac- (until 12/2006) to variations in responses to environmental tion time tasks, or when the functional val- Annette Brose demands, such as exploration of behavioral ues have been reduced, as is the case when Dorothea Hämmerer strategies during initial phases of complex individuals have consolidated the use of a Oliver Huxhold skill acquisition. Adaptability indicates an particular strategy and are operating near Irene E. Nagel Christina Röcke (until 12/2006) Intra-person variability in cognition Yee Lee Shing Functional Markus Werkle- diversity Bergner (predoctoral fellows)

Functional Functional plasticity fluctuation Functional adaptability Level of functioning Perturbed state

Figure 1. Types of intra- person variability in Acquisition Acquired functioning cognitive functioning.

136 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 137

Table 1 Taxonomy of Within-Person Variability in Cognitive Functioning Across the Lifespan

Scope Variations in a single function Transformations in functional organization Timescales (e.g., local, univariate) (e.g., global, multivariate)

Microgenetic Relatively reversible variations in one function Relatively reversible variations in functional organization (e.g., usually Examples: Examples: across trials, • processing fluctuation (processing lability • shifts in resource allocation, coordination, and sessions, or or lack of processing robustness) compensatory behavior during multitasking weeks) • neural and behavioral plasticity (short-term • context-driven variations in mental set and functional learning potential) organization (e.g., posture control with eyes open or • within-task strategic diversity (richness of closed) within-task behavioral repertoire) • situational choice and preference behavior • adaptability/resilience to environmental perturbations • cyclic (e.g., state) variations in any specific function

Ontogenetic Relatively permanent (e.g., cumulative, progressive) Relatively permanent (e.g., cumulative, progressive) (e.g., usually changes in one function alterations in functional organization across months, Examples: Examples: years, or • physical growth • ability differentiation from adulthood to old age decades) • progressive (e.g., trait) changes in any specific • ability dedifferentiation from childhood to early adulthood cognitive function • corticogenesis and functional specification of brain areas • long-term learning and skill acquisition during maturation and learning • functional reintegration of brain circuitry in old age

Note. This taxonomy is not meant to be exhaustive. For instance, societal sources of variability are not systematically considered. All listed forms of variability can be studied at behavioral and neuronal levels of analysis. Examples are drawn from both levels. A major challenge for lifespan psychology is to identify mechanisms that link local to global variations, microgenetic variations to ontogenetic change, and neuronal mechanisms to behavior. Theories that link neuronal mechanisms in a single function acting at a microgenetic timescale to global ontogenetic transformations in behavior are high in parsimony and explanatory power (adapted from Lindenberger & von Oertzen, 2006; cf. Li, Huxhold, & Schmiedek, 2004).

maximum levels of functioning. Finally, differences in neuromodulatory efficacy and temporal coupling refers to associations be- their effect on patterns of brain activation tween two or more forms of processing and cognitive performance. Altogether, the within or across domains of functioning, project endorses a multilevel multimethod Key References such as concurrent covariation, lead-lag approach by combining behavioral and Li, S.-C., & Baltes, P. B. relations, and synchronization, at identical, neuronal observations with experimental, (2006). Cognitive developmental research from different, or hierarchically nested correlational, and computational methods. lifespan perspectives: timescales. The challenge of inte- Subproject I: Intra-Person Variability and gration. In E. Bialystok & F. I. M. Craik (Eds.), Overview of Subprojects Ability Structures Lifespan cognition: The Intra-Person Dynamics Project currently A first microlongitudinal study, completed in Mechanisms of change consists of four subprojects. Subproject I November 2005, covered 45 daily measure- (pp. 344–363). Oxford: Oxford University Press. investigates adult age differences in intra- ment occasions to examine adult age differ- Li, S.-C., Lindenberger, person variability and ability structures. Sub- ences in intra-person fluctuations in four do- U., Hommel, B., project II studies lifespan age differences in mains of functioning: postural control, spa- Aschersleben, G., Prinz, the plasticity and components of episodic tial working memory, positive and negative W., & Baltes, P. B. (2004). Transformations learning and memory. Subproject III investi- affect, and task-specific motivation. Eighteen in the couplings among gates lifespan age differences in electrophys- younger adults (20–30 years of age) and 19 intellectual abilities and iological correlates of monitoring mecha- older adults (70–80 years of age) participated constituent cognitive processes across the life nisms in various types of cognitive conflicts. in the study across 9 weeks. With respect to span. Psychological Finally, subproject IV focuses on adult age cognitive and sensorimotor functions, older Science, 15, 155–163.

Center for Lifespan Psychology | 137 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 138

Key References adults showed greater intra-person fluctua- positive and negative affect than younger Lövdén, M., & Linden- tions than younger adults. For instance, older adults (Figure 4A). This pattern of findings is berger, U. (2005). De- adults’ reaction times in a spatial working consistent with the notion advanced by Laura velopment of intellec- memory task had a higher mean and a wider Carstensen and others that emotion regula- tual abilities in old age: From age gradients to distribution than younger adults’ reaction tion improves with advancing adult age. individuals. In O. times, and varied more from day-to-day in However, alternative explanations, such as Wilhelm & R. W. Engle older adults than in younger adults (see age-linked differences in social contexts or (Eds.), Handbook of understanding and meas- Figure 2). arousal, cannot be ruled out. In addition, the uring intelligence We also examined whether intra-person fluc- correlation between age and affect fluctua- (pp. 203–221). Thousand tuations in sensorimotor functioning (e.g., tions was more stronger than the correlation Oaks, CA: Sage. postural control) and cognitive functioning between personality factors, such as extra- Li, S.-C., Lindenberger, U., & Sikström, S. (e.g., spatial working memory) are more version and neuroticism, and affect fluctua- (2001). Aging cognition: strongly coupled in older than in younger tions (Figure 4B), demonstrating the impor- From neuromodulation adults. Individuals who fluctuated more in tance of a lifespan perspective on dynamic to representation. Trends in Cognitive postural control (i.e., older males) showed aspects of well-being. Sciences, 5, 479–486. stronger couplings between daily fluctuations As a continuation and expansion of the goals in postural control and daily fluctuation in of the first study, the project carried out a working memory than individuals who fluc- second study to explore differences and com- tuated less in postural control (Dissertation monalities between covariance structures of Oliver Huxhold; see Figure 3A). In addition, intellectual abilities measured either across older adults, but not younger adults, showed individuals at a given occasion or across oc- increasing sway with increasing cognitive de- casions within a given individual. Most exist- mands. This finding suggests that older ing research on intellectual abilities assumes adults need to invest more cognitive re- that covariance structures based on inter- sources into the execution of simple sensori- individual differences generalize to intra- motor tasks than younger adults (Huxhold, person structures. As Jacques Lautrey, John Li, Schmiedek, & Lindenberger, 2006; see Fig- Nesselroade, Peter Molenaar, and others have ure 3B; see also Project 2, Sensorimotor- argued for a long time, differences between Cognitive Couplings). intra-person and inter-person structures are Regarding emotional functioning (Disserta- to be expected. Conceptually, the malleability tion Christina Röcke), older adults showed of functional organization at both behavioral significantly less day-to-day fluctuations in and neuronal levels and the diversity of de-

Intra-person RT distribution across occasions (younger adult) Intra-person RT distribution across occasions (older adult)

250 250 200 200 150 150 100 100

Frequency 50 Frequency 50 0 0 –50 –50 15 0 500 0 1,000 10 500 15 1,500 5 1,000 5 10 RT (msec) 2,000 1,500 Occasions RT (msec) 2,000 Occasions (each bin is across 3 days) (each bin is across 3 days)

Figure 2. Short-term development of reaction-time distributions in one younger and one older adult. Each occasion represents three consecutive days of the 45-day assessment. The older adult’s performance is more variable and show less change over time.

138 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 139 ) 2 (A) (B) 210 Younger women 8 Younger men 190 ) 6 Older women 2 170 4 Older men 150 2 130 Older adults 0 110 Younger adults –2 90 –4 Sway area (mm 70 –6 50 –8 –50ms 50ms Standing Standing & Standing & Standing perception choice RT & 2-back

Moment-to-moment fluctuation (area/mm Processing fluctuation in working memory RT Cognitive demand

Figure 3. (A) The coupling between daily fluctuations in postural control and working memory is stronger in older men whose postural control function is less robust. (B) Increasing cognitive demands in a dual-task increases postural sway more in older adults than in younger adults (Huxhold et al., 2006).

velopmental trajectories and life experiences measures of affect, motivation, and mood render any strict congruence between intra- were also assessed on a daily basis (Disserta- person and inter-person structures unlikely tion Annette Brose). In addition, all partici- (cf. Li & Lindenberger, 2002). pants completed comprehensive pretests and The main empirical study, preceded by exten- posttests, with baseline measures of cogni- sive pilot work and labelled the Cognition tive abilities, and transfer tasks for the prac- Ergodicity Study of the MPI for Human Devel- ticed abilities. A subsample of about 25 opment (COGITO ERGO SUM), was started in younger and 25 older participants also took 2006. One hundred younger (20–30 years) part in structural and functional magnetic and 100 older (65–80 years) adults partici- resonance imaging sessions and electroen- pated in 100 daily sessions, working each day cephalographic recordings at pretest and on a set of 12 cognitive tasks comprising posttest. Data collection will be completed in perceptual speed, episodic memory, and summer 2007. working memory (see Figure 5). Self-report

(A) (B) 1.0 1.0 0.9 0.9

0.8 change) 0.8 0.7 2 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 Trend-controlled variability Trend-controlled 0.0 0.0 Positive affect Negative affect components ( R Variance Positive affect Negative affect

Younger adults Shared Unique age group Older adults Unexplained Unique personality

Figure 4. (A) Older adults show less day-to-day fluctuations in both positive and negative affects than younger adults. (B) Age is more highly correlated with individual differences in affect fluctuation than personality measures (Dissertation Christina Röcke).

Center for Lifespan Psychology | 139 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 140

Figure 5. Design of COGITO study. Pretest Longitudinal phase Posttest Younger adults (20–30 years) 10 × 2.5 h fMRI/EEG 100 daily sessions with 10 × 2 h fMRI/EEG 12 cognitive tasks: Behavioral N ~ 25 – working memory Behavioral N ~ 25 testing – episodic memory testing – perceptual speed N = 100 N = 100 N = 100

Older adults (65+ years) 10 × 2.5 h fMRI/EEG 100 daily sessions with 10 × 2 h fMRI/EEG 12 cognitive tasks: Behavioral N ~ 25 – working memory Behavioral N ~ 25 testing – episodic memory testing – perceptual speed N = 100 N = 100 N = 100

The data of this study will allow researchers between-person structures of younger adults. to systematically examine differences and For related reasons, the study will also shed commonalities between covariance structures new light on the dedifferentiation hypothesis of intellectual abilities measured either of old-age intelligence (cf. Lövdén & Linden- (a) across individuals at a given occasion or berger, 2005). According to this hypothesis, (b) across occasions within a given individual. increasing correlations among cognitive abil- Most existing research and theorizing on ities with advancing age reflect the operation cognitive abilities assumes that covariance of domain-general resource limitations. The structures based on between-person differ- critical question here is whether earlier ences generalize to intra-person structures. observations of increasing correlations Surprisingly, it is not yet known whether abil- among cognitive abilities with advancing ity structures representing between-person age, when assessed across persons, are differences are good approximations of matched by increasing correlations among ability structures representing day-to-day cognitive abilities with advancing age, when variations in cognitive performance within assessed within persons across days. Extend- individuals. To find out about this issue, ing this question to the neuronal level of developmental researchers need to abandon analysis, we can ask whether older individu- a quid-pro-quo research practice, in which als who show low levels of average perform- between-person differences stand-in for ance and high correlations among different within-person variations (cf. Lindenberger & cognitive abilities tend to be those who show von Oertzen, 2006; Lövdén & Lindenberger, diffuse cortical activation patterns relative 2005). to older individuals with high average levels To complicate matters, the degree of congru- of performance and low correlations among ence between within-person and between- cognitive abilities. person structures may vary by age. For in- In sum, the data of this study will provide a stance, given that alterations in cognitive full data cube of persons × timepoints × vari- functioning among older adults reflect variables as envisioned by Raymand B. Cattell. able combinations of age-related, pathology- This will open up new data-analytic ap- related, and death-related mechanisms, the proaches, and allow new insights into be- average similarity of the within-person struc- tween-person differences in intra-person tures to the between-person structures of old variability and change. adults may be smaller than the average similarity of the within-person structures to the

140 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 141

Subproject II: Lifespan Age Differences in practice (see Figure 6A; Dissertation Yvonne Key References Episodic Memory Plasticity Brehmer; Brehmer et al., 2007). Older adults Brehmer, Y., Li, S.-C., This subproject examines lifespan age differ- showed considerable instruction-related Müller, V., von ences in plasticity and components of performance gains (baseline plasticity), but Oertzen, T., & Linden- berger, U. (2007). Mem- episodic memory. The subproject is partially did not profit much from further training ory plasticity across the funded by a research grant from the German and practice (developmental plasticity). In lifespan: Uncovering Research Foundation (Deutsche Forschungs- contrast, younger children initially showed children's latent potential. Developmental Psy- gemeinschaft, Forschergruppe 448, Binding: smaller instruction-related performance chology, 43, 465–478. Functional architecture, neuronal correlates, gains, but considerably larger practice- Werkle-Bergner, M., and ontogeny). It pursues two interrelated related gains than older adults. The resulting Müller, V., Li, S.-C., & goals: (a) to investigate age differences in memory plasticity advantage of middle child- Lindenberger, U. (2006). Cortical EEG the plasticity of skilled episodic memory per- hood over late adulthood provides direct em- correlates of successful formance from middle childhood to later pirical support for central assumptions about memory encoding: Im- adulthood; (b) to estimate the relative contri- lifespan changes in behavioral plasticity. plications for lifespan comparisons. Neuro- bution of strategic and associative compo- To examine lifespan age differences in the science & Biobehavioral nents to lifespan differences in episodic maintenance of skilled memory performance, Reviews, 30, 839–854. memory. Lifespan age differences in memory a longitudinal follow-up study was carried Li, S.-C., Naveh- plasticity were initially investigated in a first out 11 months after completion of the first Benjamin, M., & Lindenberger, U. multisession memory training study involving study. Whereas younger and older adults (2005). Aging neuro- 23 younger children (9–10 years), 27 older were able to maintain their earlier levels of modulation impairs as- children (11–12 years), 29 younger adults performance, younger and older children ac- sociative binding: A neurocomputational (20–25 years), and 29 older adults (65–78 tually improved their performance beyond account. Psychological years). Participants in all age groups were in- the levels they had originally attained Science, 16, 445–450. structed and trained in a simplified variant (Brehmer et al., submitted; see Figure 6B). of the Method of Loci, an imagery-based This performance improvement points to mnemonic strategy. Whereas all age groups a powerful coalition between learning and benefited from mnemonic training, substan- maturation during middle childhood. tial age differences in gains were observed as In order to further study strategic and asso- a function of instruction versus training ciative components of lifespan differences in

(A) Averaged timed recall score (TRS) (B) Age differences in skill maintenance 11-month interval

4 7 6 3 5 4 2 3 1 2 Number correct per time

of baseline performance 1

0 Effect size based on pooled SD 0 Pre-instruction Post-instruction Post-training Post- Post- Pre- Post- baseline baseline plasticity developmental instruction practice instruction instruction plasticity Experiment 1 Follow-up Session

Younger adults Older childrenYounger children Older adults

Figure 6. (A) Lifespan age differences in episodic memory plasticity (Brehmer et al., 2007). (B) Maintenance of skilled performance (Brehmer et al., submitted).

Center for Lifespan Psychology | 141 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 142

episodic memory, a new series of experiments both prefrontal and mediotemporal regions was initiated to progress to a mechanistic of the brain. explanation of lifespan differences in asso- A new series of experiments tested these pre- ciative and strategic components of episodic dictions using a paired-associates recognition memory. With the strategic component of memory paradigm, similar to paradigms used episodic memory, we refer to the selection, by Moshe Naveh-Benjamin from the Univer- organization, and elaboration of episodic fea- sity of Missouri and Fergus Craik from the tures during encoding and retrieval. In con- University of Toronto. The study involved a trast, the associative component refers to total of 170 participants aged 10 to 12, 13 to mechanisms that bind features into a coher- 15, 20 to 25, and 70 to 75 years. We utilized ent memory representation or compound a fully crossed within-participant repeated trace. Due to the late maturation of pre- measures design, in which the associative frontal regions and associated neuronal component was varied by preexisting word pathways, we assume that the strategic com- pair associations, whereas the strategic component of learning and memory is less effi- ponent was varied by encoding instruction. cient in middle childhood than in early and Specifically, we manipulated two levels of young adulthood. In contrast, the associative preexisting associative strength between component, which primarily involves medio- word pairs by language types. For German temporal structures, should be fully func- speakers, German-German word pairs have tional by middle childhood, so that differ- higher associative strength than pairs with ences to younger adults in this component one German and one foreign language word should be small. With respect to older adults, (Malay language was used in the current we expect impairments in both strategic and study). We manipulated the extent of explicit associative components relative to younger strategic encoding with three levels of in- adults, reflecting senescent alterations in struction (i.e., ranging from incidental item

German-German word pairs Children Teenagers Younger adults Older adults 1.0 0.8 0.6 0.4 0.2

Hit–false alarms 0.0 –0.2 Item Pair Post- Item Pair Post- Item Pair Post- Item Pair Post- strategy strategy strategy strategy

German-Malay word pairs 1.0 0.8 0.6 0.4 0.2

Hit–false alarms 0.0 –0.2 Item Pair Post- Post- Item Pair Post- Post- Item Pair Post- Post- Item Pair Post- Post- strategy practice strategy practice strategy practice strategy practice Encoding instruction

Figure 7. Individual (colored lines) and mean performance (in black bold line) within each age group across levels of encoding instruction in the German-German (top panel) and German-Malay (bottom panel) condition (Dissertations Yee Lee Shing and Markus Werkle-Bergner).

142 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 143

instruction to explicit pair instruction and episodic memory (cf. Werkle-Bergner, Müller, elaborative strategy instruction). In the elab- Li, & Lindenberger, 2006; Dissertations Yee orative strategy condition, participants were Lee Shing and Markus Werkle-Bergner). instructed to elaborate on the word pairs with visual imagery that integrates the two Subproject III: Lifespan Age Differences in words of the pair (e.g., to imagine a dog sit- Conflict Monitoring Mechanisms ting on the back seat of a car for the word This subproject examines lifespan age differ- pair “dog-car”). For the German-Malay con- ences in conflict monitoring mechanisms. It dition, we also included a practice phase (five was started in August 2006, and is partially sessions) for participants to practice the use funded by a research grant from the German of the elaborative strategy due to the higher Research Foundation (Deutsche Forschungs- difficulty of the condition. gemeinschaft, Forschergruppe 778, Conflicts For both associative strength conditions, as Signals in Cognitive Systems). Mechanisms children’s initial performance was similar to for monitoring conflicts between one’s re- that of the older adults, whereas teenagers sponses (response conflicts), between habits and younger adults performed at comparable and deliberate actions (overriding conflicts), levels (see Figure 7 for intra- and inter-group or between actual outcomes and desired variability; note also that younger adults goals (outcome conflicts) help individuals to were provided less time to encode the words dynamically adapt to changes in the envi- than the other three groups). In the German- ronment by regulating and changing their German condition, children benefited the behavioral repertoires. Neuroimaging and most from the strategy instruction, and electrophysiological studies show that the tended to perform above the level of older anterior cingulate cortex (ACC) and related adults. In the German-Malay condition, regions are activated during conflict moni- children, teenagers, and younger adults bene- toring. fited from the strategy instruction, and con- Little is known about how conflict-monitor- tinued to improve their performance after ing mechanisms change across the lifespan. five sessions of practice. In contrast, older Older adults and children perform more adults did not show much improvement slowly and less accurate than young adults across different encoding instructions and on tasks involving response error monitoring continued to show the lowest average per- and overriding conflicts (e.g. Li et al., 2004). formance of all four groups. Thus, it seems Relatedly, event-related potentials are re- that children’s latent potential in the asso- duced in amplitude in children and older ciative component could be brought to the adults relative to young adults. Age differ- fore by combining strategic support with ences in processing outcome conflict have task-relevant practice. In contrast, older not yet been investigated so far. Using sev- adults were less able to benefit from either of eral well-established experimental paradigms the training components. Furthermore, older adapted for lifespan comparisons, this sub- adults were particularly disadvantaged when project has begun to examine lifespan age associative demands were high (i.e., in the differences in electrophysiological correlates German-Malay condition), probably due to of monitoring mechanisms during response, deficits in both strategic elaboration and as- overriding, and outcome conflicts. Prelimi- sociative binding. In summary, these findings nary results suggest that children and are consistent with the guiding hypothesis of younger adults show larger outcome conflict a lifespan dissociation between strategic and related ERPs than older adults (Dissertation associative components of episodic memory Dorothea Hämmerer). (Dissertations Yee Lee Shing and Markus Werkle-Bergner). Currently, we are developing EEG paradigms to examine age differences in binding mechanisms at the neuronal level in the domains of visual perception and

Center for Lifespan Psychology | 143 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 144

Key References Subproject IV: Neuromodulation of the largest cognitive improvement and the Bäckman, L., Nyberg, L., Cognitive Dynamics most pronounce increase in neural efficiency Lindenberger, U., Li, In collaboration with the Berlin Neuroimag- following the administration of a dopamine S.-C., & Farde, L. (2006). The correlative triad ing Center, this large-scale subproject com- agonist. among aging, dopamine, bines genomic, pharmacological, and brain Given the distribution of COMT polymor- and cognition: Current imaging approaches to investigate the triadic phisms (val/val, val/met, met/met) in the status and future prospects. Neuroscience relation between aging, dopaminergic modu- population and the need to exclude individu- & Biobehavioral Reviews, lation and cognition (Bäckman, Nyberg, als not suitable for a neuroimaging study 30, 791–807. Lindenberger, Li, & Farde, 2006). Specifically, with a pharmacological component, the main Li, S.-C., Brehmer, Y., we investigate the main effects and interac- study in this subproject, which was initiated Shing, Y.-L., Werkle- tions of three factors that affect dopaminer- in 2006, proceeds in two phases. In the first Bergner, M., & Lindenberger, U. (2006). gic neuromodulatory efficacy: (a) chronologi- phase, more than 1,000 younger (20–30 Neuromodulation of as- cal age, (b) the Catechol-O-Methyltrans- years old) and older (60–70 years old) indi- sociative and organizational plasticity across ferase (COMT) genetic polymorphism, and viduals are being screened for COMT poly- the life span: Empirical (c) pharmacological intervention. Effects are morphisms and administered a comprehen- evidence and neurocom- examined at behavioral, neurofunctional, sive battery of cognitive tests (see Figure 8 putational modeling. Neuroscience and Bio- and neurostructural levels of analysis (see for domains of cognitive functions assessed). behavioral Reviews, 30, Figure 8). In the second phase, which will start in 775–790. Based on theoretical models and data in sup- summer 2007, we plan to test a total of 96 port of a curvilinear relation between participants in a double-blind placebo- dopaminergic activity and cognitive perform- controlled cross-over design involving three ance, this subproject will examine the fol- imaging sessions. The participants will be lowing hypotheses (Dissertation Irene Nagel): equally divided by age group and COMT poly- (a) Older age and val/val COMT genotype are morphism (val/val vs. met/met), in three im- associated with lower cognitive performance, aging sessions: (a) functional imaging session higher within-person variability, and more with placebo treatment, (b) functional imag- diffuse (e.g., bilateral) patterns of brain acti- ing session with D-amphetamine, and (c) vation; (b) older val/val carriers will show dis- structural imaging including diffusion tensor proportionately reduced cognitive perform- imaging. By early 2007, about 500 partici- ance and inefficient brain responses (e.g., pants had completed the first phase. Prelimi- overactivation for a given level of perform- nary analyses suggest expected associations ance); (c) older val/val carriers will exhibit between cognitive performance and COMT

Figure 8. Combined genomic, pharmacological, Themes of the AGE-COG-DA study at a glance and brain imaging investigations of interactive relations between aging, Age group cognition, and dopaminergic modulation.

Aging In domains of: working memory episodic memory processing speed Genetic variation processing robustness

Dopamine Cognition

Pharmacological At levels of: intervention behavior brain function brain structure

144 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 145

Figure 9. Effects of COMT Backward recall genotype on backward serial recall performance 1.0 COMT genotype Met/met n = 18 as observed in a prelimi- 0.9 Val/met n = 41 nary sample of younger 0.8 Val/val n = 14 adults. The negative effect of the val allele on 0.7 cognitive performance is 0.6 most pronounced in the 0.5 middle segment of the

Accuracy 0.4 serial position curve, 0.3 probably reflecting dif- 0.2 ferential susceptibility to 0.1 interference. 0.0 123456789101112 Word position

Figure 10. Adult age differences in a spatial working memory task 1.00 1,000 performed in the imaging session. Increasing 0.80 800 working memory load reduces accuracy, but increases reaction time. 0.60 600 Furthermore, older adults performed worse and slower than younger 0.40 400 adults. (Dissertation

Mean accuracy Mean correct RT in ms Irene Nagel; data are based on 11 younger and 0.20 200 11 older adults).

0.00 0 Younger Older Younger Older Adult age group Adult age group

1-back 2-back 3-back

polymorphisms (Figure 9). Furthermore, initial suggest that older adults tend to show more behavioral (Figure 10) and functional brain diffuse bilateralized brain activation patterns imaging (Figure 11) results based on pilot in prefrontal regions (Dissertation Irene work with a spatial working memory task Nagel).

Center for Lifespan Psychology | 145 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 146

Figure 11. Brain activation patterns during a spatial working memory Left Right task. Activation are derived by subtracting a high-load condition from the fixation baseline. Brighter colors indicate higher activation. The Younger fronto-posterior network, usually found during spatial working memory tasks, is active. Older adults tend to show more activation in prefrontal cortex than younger adults, and the activation appears to be more bilateral (Disserta- tion Irene Nagel; data Older are based on 11 younger and 11 older adults).

p > 2.0 (p < 0.005 uncorrected)

2.0 6.0

146 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 147

Research Project 2 Scientific Sensorimotor-Cognitive Couplings Investigators

Conceptual Overview Martin Lövdén This project investigates lifespan changes in the interactions between sensorimotor and cog- Ulman Lindenberger nitive aspects of behavior. Everyday life often requires integration of multiple sensory inputs and concurrent coordination of sensorimotor and cognitive demands. Examples are walking Oliver Huxhold while trying to memorize a shopping list, maintaining one’s balance on a bus while trying to Sabine Schäfer read an advertisement, or trying to remember the way to a friend’s house while driving in the (postdoctoral hectic morning traffic. Everyday observation further suggests that older adults, and young fellows) children, need to invest more attention into sensorimotor aspects of their behavior than teenagers and young adults. For example, when facing an obstacle on a narrow path, older Michael adults may tend to stop talking and resume their conversations after the obstacle has been Schellenbach overcome, whereas the same obstacle will affect younger adults’ conversation to a lesser ex- Julius Verrel tent. How do individuals of different ages adapt to multiple sensorimotor and cognitive de- (predoctoral mands? How does the interaction between sensorimotor and cognitive dimensions of behavior fellows) change across the lifespan? This project seeks to provide answers to these questions by studying sensorimotor and cognitive behavior in multiple-task settings with a high degree of everyday validity. A related goal of the project is to propose and test criteria for effective technological assistance in old age.

Dual-Task Costs in the Domain of Walking impression of actually walking through this Earlier studies from this project demon- environment. The task for participants might strated that older adults invest considerable be, for example, to find and remember the cognitive resources to compensate for the way from the entrance of a museum to the decreased efficiencies of their sensorimotor bistro. This laboratory furthermore features functions. For instance, Lindenberger, Marsiske, and Baltes (2000) had participants from three age groups walk different tracks while memorizing word lists. They found that speed and accuracy of walking were reduced when participants had to simultaneously walk and memorize, particularly in older adults. Compared to younger adults, older adults also showed more pronounced performance reductions in the cognitive domain when walking on the track. In our recent work, we expanded the study of the old-age quandary between increasing cognitive control demands of sensorimotor functioning and decreasing efficiency of relevant control operations to the domain of spatial navigation. To this end, we developed a virtual environment paradigm with a walking component (Figure 1). A scenary, designed to give participants the impression of walking through, for example, an art museum or a zoo, is projected in front of a treadmill. The movement of the treadmill is synchro- Figure 1. The motion laboratory used for studies on dual-task costs in the domain nized to the visual flow of the virtual envi- of walking and for examining effects of assistive technology on walking stability ronment such that participants have the and spatial navigation performance.

Center for Lifespan Psychology | 147 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 148

Key References an advanced motion capture system and in- perfect performance. Walking support atten- Schäfer, S., Huxhold, tegrated synchronized assessment of EEG and uated age-related decrements in navigational O., & Lindenberger, U. EMG components. To visualize the partici- learning (see Figure 3), and walking with (2006). Healthy mind in pant’s movement (e.g., while walking), mark- navigation load increased older adults’, but healthy body? A review of sensorimotor-cogni- ers reflecting infrared light are attached to not younger adults’, trunk-angle variability. tive interdependencies in the participant’s body. In turn, cameras cap- Thus, walking demands influenced the navi- old age. European Re- ture the position of the markers, which are gation performance of older but not younger view of Aging and Physi- cal Activity, 3, 45–54. postprocessed offline according to bio- adults. Lövdén, M., Schellen- mechanical models. This procedure allows for Thus far, studies of the interaction between bach, M., Grossman- the visualization of the participant’s move- cognitive tasks and walking behavior point to Hutter, B., Krüger, A., & ments and further analyses. Figure 2 shows greater dual-task costs in older adults, pre- Lindenberger, U. (2005). Environmental topogra- captures of different motions that have been sumably due to increased demands of motor phy and postural control taken in this laboratory: A single participant control on cognitive control and reductions demands shape aging- has performed different typical motions (e.g., in cognitive control capacity. However, a re- associated decrements in spatial navigation per- walking, dancing, playing tennis). cent study on postural control during upright formance. Psychology The first study with this paradigm (Lövdén, standing, undertaken in collaboration with and Aging, 20, 683–694. Schellenbach, Grossman-Hutter, Krüger, & with the Intra-Person Dynamics Project (see Lindenberger, U., Lindenberger, 2005) tested our fundamental Figure 3, p. 139), suggests that this may not Marsiske, M., & Baltes, P. B. (2000). Memorizing hypothesis that aging-induced cognitive per- be the whole story (Huxhold et al., 2006). In- while walking: Increase meation of sensorimotor functions con- stead, a dual-process model may be needed in dual-task costs from tributes to adult age differences in spatial to capture the interactions between cognitive young adulthood to old age. Psychology and navigation performance. Sixteen 20- to 30- control and motor control. Specifically, a Aging, 15, 417–436. year-old and sixteen 60- to 70-year-old men U-shaped relationship between motor control were required to find and remember the way and difficulty of concurrent cognitive activi- to the bistro in museums under conditions of ties during balance was found; that is, motor walking with support (holding on to a hand- control was improved by concurrent cogni- rail) or without support until they reached tive tasks with low difficulty and attenuated

Figure 2. Processed motion captures of an individual performing various different movements. The position of reflective markers attached to the body of the participant is cap- tured by infrared cameras and postprocessed according to biomechan- ical models to arrive at a dynamic visualization of the participant’s movements.

148 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 149

by additional increases in difficulty. Presum- ably, cognitive activities of lower difficulty 1,600 may promote an external focus of attention that allow the motor system to self-organize 1,400 Younger adults and smoothly execute movement. At the 1,200 Older adults same time, higher levels of cognitive task dif- 1,000 ficulty may hamper motor control performance through cross-domain resource compe- 800 tition, much like those found in the walking 600 studies reported above. The point at which 400 performance improvements due to the first Distance covered (meters) 200 process is surpassed by decrements induced 0 by the second process in a given individual Walking with support Walking without support should be predictable by the individual’s Navigation condition sensorimotor and cognitive resources and the demands of motor control on cognitive control. Figure 3. Walking support reduces adult age differences in spatial navigation performance (Lövdén, Schellenbach et al., 2005). In a recent study Lövdén, Schäfer, Pohlmeyer, and Lindenberger tested the dual-process account in the domain of walking by paramet- their gait variability with increasing working rically manipulating working memory load, memory load. In the cognitive domain, how- using the n-back task, in 32 younger adults ever, both children’s and young adults’ work- with a mean age of 25 years and 32 older ing memory performance was higher when adults with a mean age of 74 years who were walking than when sitting on a chair. Thus, at asked to walk on a treadmill at self-selected least in children and young adults, cognitive speed. Effects of age and age by working performance may benefit from periodic forms memory load, but not dual-tasking, were of body movement, such as continuous walk- found for cognitive performance. Stride-to- ing across a wide range of difficulty levels. stride variability was reduced when participants simultaneously performed an easy working memory task (1-back), compared to walking without a task. Further increments in working memory load (i.e., from 1-back up to 4-back) caused reductions in variability, but not in means, of stride time and stride length 1.7 Older adults Younger adults in younger adults, but not in older adults. In line with these observations, a principal component analysis based on gait patterns indi- 1.5 cated a U-shaped pattern for residual whole- body variability in older adults (see Figure 4; 1.3

Verrel, Lövdén, Schäfer, & Lindenberger, in Residual variance (%) prep.). We conclude that normal aging is associated with alterations in the trade-offs 1.1 between two continuous control processes No task 1-back 2-back 3-back 4-back involving positive effects of external focus of Working memory load (n-back) attention and negative effects of resource competion, respectively. Our current work on Figure 4. Residual variance of whole-body motion after accounting for the first this issue also includes lifespan comparisons. four principal components as a function working memory load and adult age. In a first study, Schäfer, Lövdén, Wieckhorst, Whole-body motion variability decreases with low cognitive demands, and then increases again. This increase occurs earlier for older adults, pointing to increasing Pohlmeyer, and Lindenberger found that chil- cross-domain resource competition with advancing adult age (Verrel et al., in dren, similar to older adults, tend to increase prep.).

Center for Lifespan Psychology | 149 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 150

Dual-Task Costs in the Domain of Balance ued to show performance improvements in Balance performance can be measured by the balance domain under dual-task condi- dynamic posturography. For this purpose, tions even when participants were instructed participants stand on a force platform that to focus their attention more strongly on the measures their stability at any given point in cognitive task. This pattern of resource allo- time. Previous research in our laboratory has cation is interpreted as an incidence of adap- shown that older adults tend to prioritize the tive resource allocation, since it helps chil- balance task in a challenging dual-task situa- dren, who are operating closer to their stabil- tion, by showing less pronounced dual-task ity boundaries than young adults, to stabilize related performance decrements (dual-task their body’s equilibrium in a challenging costs) in the balance domain as opposed to situation. the cognitive domain (e.g., Rapp, Krampe, & Key Reference Baltes, 2006). In her dissertation thesis, Future Perspectives Lindenberger, U., & Sabine Schäfer extended this research to the In our future work, we will examine the ef- Lövdén, M. (2006). Co- other end of the lifespan, by investigating fects of assistive technology on spatial navi- constructing human engineering technologies in how children differ from young adults in gation while walking. For instance, we hy- old age: Lifespan psychol- their ability to balance on a special balance pothesize that providing older adults with ogy as a conceptual foun- device, the ankle-disc board, while concur- appropriate spatial navigation support, which dation. In P. B. Baltes, P. A. Reuter-Lorenz, & F. Rösler rently memorizing word lists or working on a is not requiring more cognitive resources for (Eds.), Lifespan develop- working memory task. While both children operation that it releases, enhances their ment and the brain: The and young adults showed performance ability to find and remember the way to a perspective of biocultural co-constructivism decrements (dual-task costs) in the cognitive place in the environment as well as their (pp. 350–375). Cambridge, domain, young adults also increased their walking stability (Lindenberger & Lövdén, UK: Cambridge University body sway on the ankle-disc board when 2006). Moreover, we will address the Press. concurrently working on a demanding cogni- processes involved in the increased demands tive task. Children, on the other hand, swayed of motor control on cognitive control, even less when concurrently performing a and the possibility of separating flexibility cognitive task, resulting in negative dual-task and stability of motor behavior, such as in costs (see Figure 5; Schäfer, Krampe, Linden- walking. berger, & Baltes, submitted). Children contin-

Figure 5. In young adults, dual-task costs are comparably high in the cog- 20 nitive and the balance domain. Children show a 15 trade-off pattern be- 10 tween dual-task costs in the two domains, with 5 performance improvements in the balance 0 task under dual-task –5 conditions, assumably

because they are operat- Dual-task costs (%) –10 ing closer to their stabil- Cognition ity boundaries when bal- –15 Balance ancing on the ankle-disc –20 board (Schäfer et al., submitted). Children Young adults

150 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 151

Research Project 3 Scientific The Berlin Aging Study (BASE): Dynamic Modeling of Longitudinal Change Investigators and Mortality Ulman Lindenberger During the 20th century, average life expectancy nearly doubled. More and more individuals in Jacqui Smith current cohorts of older individuals experience additional years of life between the ages of 70 Julia Delius and 100+. What do these added years mean in terms of levels of functioning and life quality Shu-Chen Li for most people? Are there constraints on aging successfully in the last years of life? To exam- Martin Lövdén ine these issues, members of the Center of Lifespan Psychology have been investigating age- and death-related changes in psychological functioning from age 70 to 100+ in the context of Denis Gerstorf the Berlin Aging Study (Baltes & Mayer, 1999). (postdoctoral fellow, until 09/2005) At present, longitudinal data in BASE are The analyses of the BASE data set during the available for seven measurement occasions 2005–2006 period can be classified into Daniel Grühn spanning over 15 years. Mortality-related three broad categories. First, the BASE re- (predoctoral fellow, information is updated at regular intervals. search team made increasing use of a spe- until 09/2005) As of March 2007, 85 % (n = 439) of the cific variant of latent growth curve modeling, 516 individuals who had participated in the the multivariate dual-change score model 14-session multidisciplinary assessment introduced by John J. McArdle and Fumiaki at the first occasion about 16 years ago were Hamagami, in order to test dynamic hypothe- no longer alive. ses about lead-lag relations between and

Overview of the Berlin Aging Study (BASE) www.base-berlin.mpg.de The multidisciplinary Berlin Aging Study (BASE), initially directed by Paul B. Baltes and Karl Ulrich Mayer, was started in 1989 under the sponsorship of the former West Berlin Academy of Sciences and Tech- nology and its Committee on Age and Societal Development as well as the Federal Ministry for Family Affairs, Senior Citizens, Women, and Youth. Later, the study came under the auspices of the Berlin- Brandenburg Academy of Sciences. As of 2006, the study involves 7 measurement occasions spaced over 15 years. In addition, several sub- samples have been recruited for intensive study. The distinguishing features of BASE include (1) a focus on the very old (70 to 100+ years), (2) a locally representative sample, stratified by age and sex, and (3) a broad-based interdisciplinarity (involving two research groups from the Free University Berlin, In- ternal Medicine and Psychiatry, and two from the MPIB, Sociology and Psychology). In addition to discipline-specific topics, four integrative theoretical orientations guide the study: (1) differential aging, (2) continuity versus discontinuity of aging, (3) the range and limits of plasticity and reserve capacity, Key References and (4) aging as a systemic phenomenon. Smith, J. (2006a). Berlin The initial focus of BASE (1990–1993) was to obtain a heterogeneous sample, stratified by age and sex, Aging Study. In R. Schulz, of individuals aged 70 to 100+ years who completed a 14-session intensive protocol that involved de- L. S. Noelker, K. Rock- tailed measures from each of the four participating disciplines. The original sample participating in this wood, & R. L. Sprott initial intensive protocol consisted of 258 men and 258 women from the western districts of Berlin. Up (Eds.), Encyclopedia of to 2005, six longitudinal follow-ups of the survivors from this initial sample involving different amounts aging: A comprehensive of assessment had been completed at approximately 2-year intervals. A single-session multidisciplinary resource in gerontology assessment was collected in 1993 and 1994 (n = 361), reduced versions of the intensive protocol (six and geriatrics (4th ed., sessions) were collected in the periods 1995 and 1996 (n = 206) and 1997–1998 (n = 132), and a repeat Vol. 1, pp. 110–112). of parts of the psychology battery together with multidisciplinary outcome variables (e.g., screening for New York: Springer. dementia, assessment of well-being) in 2000 (n = 88), 2004 (n = 47), and 2005 (n = 38). In addition, & mortality information about the entire BASE sample was updated at regular intervals. Baltes, P. B., Mayer, (1999). The Berlin The initial sample of 516 individuals formed the basis of the cross-sectional analyses reported in a K. U. Aging Study: Aging from German monograph first published in 1996 (Mayer & Baltes, 1996, 1999), in a featured section of Psy- 70 to 100. New York: chology and Aging (1997), and an English monograph published with Cambridge University Press (Baltes Cambridge University. & Mayer, 1999, 2000). Six papers reporting two-wave longitudinal findings were published in November 2002 in a special section of the Journal of Gerontology: Psychological Sciences (57B, P471–P571). Spe- Lindenberger, U., & cific interests of the Psychology Unit of BASE include issues of sample selectivity and representative- Baltes, P. B. (1994). Sen- ness, cognitive aging, subgroup profiles of psychological functioning, the Fourth Age, gender differences, sory functioning and mortality prediction, self-related change, well-being, and models of successful aging, such as selective intelligence in old age: optimization with compensation. A strong connection. Psychology and Aging, 9, 339–355.

Center for Lifespan Psychology | 151 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 152

within domains of functioning. Second, and tioning takes clear developmental precedence in line with the project’s emphasis on selec- over the other. tive attrition and predictors of mortality, we Gerstorf, Lövdén, Röcke, Smith, and Linden- continued our attempts to separate age- berger (2007) examined cross-domain associ- associated, pathology-associated, and death- ations between perceptual speed and well- related contributions to developmental being. Reports of well-being were found to change. Third, we continue to examine indi- influence subsequent decline in perceptual vidual differences in aging patterns. In the speed. No evidence was found for a directed following, we selectively summarize research effect in the other direction. Potential covari- from each of the three categories. ates, such as initial health constraints, personality, or social participation, did not ac- Key Reference The Dynamics of Longitudinal Change: count for these differential lead-lag associa- Lövdén, M., Ghisletta, P., Identifying Lead-Lag Relations tions. These results are of general importance & Lindenberger, U. A central question in cognitive aging re- for understanding patterns of aging across (2005). Social participation attenuates decline search concerns the relation between social different functional domains, as they suggest in perceptual speed in participation and cognitive change. Does an that well-being is not only a consequence of old and very old age. engaged and active lifestyle in old age allevi- but also a source for successful (cognitive) Psychology and Aging, 20, 423–434. ate cognitive decline, does high cognitive aging. functioning in old age increase the possibility Finally, Lövdén, Li, Shing, and Lindenberger of maintaining an engaged and active (in press) examined the hypothesis that lifestyle, or both? Lövdén, Ghisletta, and losses in neuronal signaling fidelity cause Lindenberger (2005) addressed these ques- senescent changes in behavior by testing tions by applying the bivariate dual-change whether trial-to-trial reaction time variability score model to BASE longitudinal data. The in one test of perceptual speed (Identical authors found that prior scores of social par- Pictures) antecedes and predicts longitudinal ticipation influence subsequent changes in decline in levels of performance on another perceptual speed, while the opposite does not test of perceptual speed (Digit Letter Substi- hold. These results lend direct support to the tution) and a test of ideational fluency (Cate- hypothesis that an engaged and active gories). Higher trial-to-trial variability pre- lifestyle in old and very old age may alleviate ceded and predicted greater decline in levels cognitive decline. of perceptual speed and ideational fluency Earlier cross-sectional analyses of the BASE performance, whereas the reverse direction data set revealed a strong association be- of influence was not reliably different from tween intellectual and sensory functioning in zero (see Figure 1). In accord with neurocom- old age (Lindenberger & Baltes, 1994). How- putational theories of cognitive aging (see ever, based on cross-sectional data, the lead- Project 6), this finding suggests a develop- lag relations between sensory and cognitive mental cascade from senescent changes in decline remain unclear. Therefore, Ghisletta variability to senescent changes in central and Lindenberger (2005) applied advanced tendency. structural equation models to longitudinal The extensive use of the dual-change score BASE data in order to explore the structural model in these and other ongoing longitudi- dynamics among perceptual speed, verbal nal analyses of the BASE data set mandates knowledge, close visual acuity, and distance close attention to the model’s statistical visual acuity. The authors found latent re- properties. Methodological work on this issue gression paths influencing longitudinal is currently being undertaken in collaboration changes within and across intellectual and with Paolo Ghisletta from the University sensory domains. Their findings support the of Geneva and Christopher Hertzog from the hypothesis that intellectual and sensory do- Georgia Institute of Technology, Atlanta (see mains are dynamically linked in old and very Project 6). old age. At the same time, they lend little support to the view that one domain of func-

152 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 153

(A) Categories (B) Trial-to-trial variability (CoV) 70 70

60 60

50 50 -scores) -scores) 40 40

30 30 Initial level of categories + 0.5 SD Initial level of categories + 0.5 SD Performance ( T Performance 20 Initial level of variability ( T Performance 20 Initial level of variability Initial level of variability – 0.5 SD Initial level of variability – 0.5 SD 10 10 12345678 12345678 Assessment occasion Assessment occasion

Figure 1. Is high trial-to-trial variability (coefficient of variation, CoV) preceding and predicting decline in performance on the Categories test, a measure of ideational fluency? The two panels of this figure show the model-implied means from the bivariate dual change score model for Categories performance (Panel A) and for trial-to-trial variability (Panel B) as a function of measurement occasion and initial (t1) levels of performance (–0.5 SD, estimated mean, and +0.5 SD) on either trial-to-trial variability (Panel A) or categories (Panel B). Lower initial trial-to-trial variability is associated with a less negative development of Categories performance (Panel A). In contrast, variations in initial level of Categories performance have little influence on the development of trial-to-trial variability (Panel B) (adapted from Lövdén, Li et al., in press).

Disentangling Age- and Mortality-Related Ram, Christina Röcke, Ulman Lindenberger, Key Reference Changes and Jacqui Smith explored terminal decline in Ghisletta, P., McArdle, In the context of a special section on death reported life satisfaction. Life satisfaction as J. J., & Lindenberger, U. (2006). Longitudinal and cognition in the European Psychologist a function of distance to death was associ- cognition-survival rela- edited by Lars Bäckman and Stuart ated with greater interindividual differences tions in old and very old MacDonald, Ghisletta, McArdle, and Linden- and steeper average decline than life satis- age: 13-year data from the Berlin Aging Study. berger (2006) made combined use of latent faction as a function of distance from birth European Psychologist, growth curve modeling and survival analysis (i.e., age). Furthermore, the authors identified 11, 204–223. to estimate the influence of level and change a point about 4 years before death at which in cognition on age at death in old and very decline showed a two-fold increase in steep- old age. When cognitive performance was ness relative to the preterminal phase. For in- accounted for, age, sex, and motor perform- dividuals 85 years and older, they observed a ance predicted survival, while socioeconomic three-fold increase. Established mortality status and broad personality factors did not, predictors including sex, comorbidities, de- and sensory performance did only at times. mentia, and cognition accounted for only Also, when cognitive variables were analyzed small portions of the observed associations independently of each other, both level and between life satisfaction and distance to change in speed and fluency, as well as level death. Thus, it appears that late-life changes in memory and knowledge, predicted survival. in subjective well-being are related to mech- However, when all cognitive variables were anisms predicting death (see Figure 2). analyzed simultaneously, none of them were significantly associated to survival. Taken to- Individual Differences in Psychological Key Reference gether, these findings suggest that survival is Aging Profiles Gerstorf, D., Herlitz, A., related to cognitive development in old and Gerstorf, Herlitz, and Smith (2006) examined & Smith, J. (2006). Sta- bility of sex differences very old age in a global rather than ability- whether patterns of sex differences on tasks in cognition in advanced specific manner. of perceptual speed, episodic memory, verbal old age: The role of edu- In a further attempt to separate death- fluency, and verbal knowledge are main- cation and attrition. Journal of Gerontology: related from age-associated changes in psy- tained during advanced old age. Using longi- Psychological Sciences, chological functioning, Denis Gerstorf, Nilam tudinal BASE data from participants screened 61B, P245–P249.

Center for Lifespan Psychology | 153 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 154

Figure 2. Prototypical (thick line) and individ- 70 ual changes (thin lines) in life satisfaction over distance to death, as revealed by the optimal multiphase growth model with a transition 60 point about 4 years prior to death. Such a two- phase model described the structure of intraindividual changes in life satisfaction in the BASE 50 data set better than a single-phase model covering the complete 15- year period. The prototypical rate of decline steepened from the preterminal phase 40 (–0.59 T-score units per Life satisfaction ( T -score) year) to the terminal phase (–1.16 T-score units per year) by a factor of 2 (adapted from Gerstorf et al., in press). 30

20 15 10 5 0 Distance to death (years)

for dementia, the authors estimated sex- ship, subgroup-to-subgroup change was specific age trajectories of cognitive change associated with greater decline and later and explored the roles of education and se- mortality. lective attrition. The authors found that both To further examine the link between sensory women and men declined virtually in parallel, and intellectual functioning, Ulman Linden- with no evidence of differential change. After berger and Paolo Ghisletta examined covari- statistically controlling for differences in ed- ances in longitudinal change for eight cogni- ucation, women’s average performance ex- tive measures assessing four intellectual ceeded that of men on tasks from all four do- abilities as well as close vision, distance mains of functioning (see Figure 3). vision, and hearing. Cognitive declines were Gerstorf, Smith, and Baltes (2006) extended highly correlated, with a single factor ac- cross-sectional work in BASE to 6-year longi- counting for 60 % of the variance (65 % tudinal cluster analyses (n = 132). At base- when controlling for age at the first occa- line, three different subgroups had been sion, distance to death, and risk for demen- identified, with distinct within-person pro- tia). In contrast, correlations between cogni- files across cognitive, personality, and social- tive and sensory declines were less promi- integration constructs. Over time, about two nent. While confirming the need to search for thirds of the participants remained in the general mechanisms of behavioral senes- same subgroups. Baseline subgroups differed cence, these results also suggest that sen- in level and slope of subsequent change and sory-cognitive links may be less pronounced in two long-term outcomes, well-being and when observed longitudinally than when ob- mortality. Independent of subgroup member- served cross-sectionally.

154 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 155

Christiane Hoppmann, Denis Gerstorf, Jacqui observation period. These findings speak to Smith, and Petra Klumb investigated links of important associations between beliefs about possible selves in three domains—health, possible selves and activities in advanced old everyday cognition, and social relations—to age. Life-history-related differences in possi- daily activities. In the domains of health and ble selves were explored by Hoppmann and social relations, hoped-for selves were asso- Smith (2007). ciated with higher probabilities of performing Future analyses in BASE will focus on selec- daily activities in these domains. Also, indi- tive mortality, longitudinal patterns, and the viduals who engaged in hope-related activi- in-depth assessments taken in 2005. In addi- ties reported higher positive affect, and were tion, we plan to conduct an eighth longitudi- more likely to survive a subsequent 10-year nal assessment wave in 2008.

Figure 3. Estimated intel- 60 60 lectual ability age gradi- Digital Letter Identical Pictures ents for men and women in the Berlin Aging Study 50 50 (BASE), residualized for education using regression analyses. Solid lines T -score T -score 40 40 represent men and dashed lines represent women. At the zero- 30 30 order level, there were 70 80 90 100 70 80 90 100 sex differences in inter- Age Age cepts of cognitive functioning favoring women 60 60 on the Paired Associates Paired Associates Memory for Text and Memory for Text tests and those favoring 50 50 men on the Vocabulary test. Once we took sex T -score

T -score differences in education 40 40 into account, the advantage for men vanished, and a reliable advantage 30 30 of women over men was 70 80 90 100 70 80 90 100 Age Age found for five out of the eight cognitive tests: Digit Letter, Paired Asso- 60 60 ciates, Memory for Text, Categories Word Beginnings Word Beginnings, and Spot-a-Word. For exam- 50 50 ple, on the Paired Associ- ates test, the average T -score T -score performance of women 40 40 exceeded the average performance of men by 4 T-score units. The mag- 30 30 nitude of sex differences 70 80 90 100 70 80 90 100 Age Age remained constant throughout the age range studied in BASE. 60 60 We standardized the T- Spot-a-Word Vocabulary scores (mean = 50, SD = 10) to the T1 parent 50 50 BASE sample to ensure a common metric while T -score T -score maintaining both the 40 40 psychometric properties of the scores and the 30 30 longitudinal changes in means and variances 70 80 90 100 70 80 90 100 Age Age (adapted from Gerstorf, Herlitz, & Smith, 2006).

Center for Lifespan Psychology | 155 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 156

Scientific Research Project 4 Investigators Developmental Regulation of Affect, Motivation, and Abilities (DRAMA)

Michaela Riediger Overview Ulman Lindenberger The DRAMA Project investigates lifespan changes in the regulation of goals and preferences. This includes the pursuit and coordination of personal goals, the use of shared knowledge in Natalie C. Ebner the goal-directed behavior of couples, and the implications of lifespan changes in cognitive Sabine Schäfer resources on schema reliance and task selection. To strike a balance between external and in- (postdoctoral ternal validity and to assess both intentional and implicit facets of self-regulation, the project fellows) uses a variety of empirical methods, ranging from laboratory experiments over self- and peer reports to microlongitudinal studies embedded into participants’ everyday lives. The concep- Antje Rauers tual framework of the project owes much to the selection, optimization, and compensation (predoctoral fellow) (SOC) metamodel of lifespan development introduced by the late Margret M. and Paul B. Baltes. According to this model, successful development requires the regulation of four universal developmental mechanisms: elective selection, loss-based selection, optimization, and compensation (see Figure 1).

Currently, the DRAMA project is structured Subproject I: Goal Regulation in into three subprojects. Subproject I studies Individuals and Dyads the regulation of personal goals and other Lifespan theory posits that people actively forms of goal-directed behavior in individuals shape their own life course in the context of and dyads. Subproject II examines whether malleable cultural, historical, and biological younger and older adults differentially rely on constraints. Selecting, pursuing, and regulat- schematic knowledge about the lifespan ing such personalized developmental goals when processing developmental goals and play important roles in intentional self- emotions. Subproject III experimentally in- development, particularly in adolescence and vestigates SOC mechanisms from a lifespan adulthood. perspective. Younger adulthood typically offers access to resources that allow the acquisition of skills

The selection, optimization, and compensation metamodel of lifespan development (SOC)

Central proposition Adaptive development results from the interaction of three universal developmental regulatory processes:

1. Selection (elective and loss-based) Focusing one’s resources on a subset of potentially available options, either in response to new demands or tasks (elective selection) or in response to actual or anticipated losses (loss-based selection) → Directionality aspect of development

2. Optimization Acquisition, refinement, and coordinated application of resources directed at the achievement of higher functional levels → Growth aspects of development

3. Compensation Figure 1. The selection, Efforts to maintain a given level of functioning despite actual or optimization, and anticipated decline in, or loss of, previously available resources compensation (SOC) → Regulation of loss in development metamodel of lifespan development (Baltes & Baltes, 1990).

156 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 157

and the improvement of abilities, whereas A second series of studies carried out explored Key References later parts of adulthood offer fewer opportu- facilitative and interfering relations among Ebner, N. C., Freund, nities for growth (Ebner & Freund, in press; multiple developmental goals within the same A. M., & Baltes, P. B. (2006). Developmental Freund & Ebner, 2005). In a first series of person (cf. Riediger, 2006). Using experience changes in personal goal studies carried out in 2003 and 2004, we ex- sampling and observational methods, we orientation from young amined individuals’ tendency to frame devel- found that intergoal interference is associated to late adulthood: From striving for gains to opmental goals in terms of (a) achieving pos- with impairments in subjective well-being, maintenance and pre- itive states (e.g., ”I want to become healthy”), and that intergoal facilitation is associated vention of losses. Psy- (b) maintaining achieved states (e.g., ”I want with enhanced behavioral involvement in goal chology and Aging, 21, to stay healthy”), or (c) avoiding negative pursuit. Also, older adults tended to report 664–678. states (e.g., ”I don’t want to become sick”). more mutually facilitative goals than younger Riediger, M., Li, S.-C., & Lindenberger, U. (2006). Results from both self-report and experimen- and middle-aged adults, resulting from in- Selection, optimization, tal studies confirmed the expectation, based creasing motivational selectivity from middle and compensation as de- on lifespan theory (Baltes, Lindenberger, & to early-late adulthood. Thus, in line with the velopmental mechanisms of adaptive resource Staudinger, 2006), that goal orientations SOC metamodel of successful lifespan devel- allocation: Review and shift from growth toward maintenance and opment (see Figure 1), older adults’ ability to preview. In J. E. Birren & loss-prevention throughout adulthood (Ebner, invest resources into fewer more central and K. W. Schaie (Eds.), The handbooks of aging: Freund, & Baltes, 2006; see Figure 2). These more similar goals may promote their subjec- Vol. 2. Handbook of the studies also show that loss-prevention orien- tive well-being and everyday competence. psychology of aging (6th tation is negatively related to well-being in Specifically, intergoal facilitation may con- ed., pp. 289–313). Amsterdam: Elsevier. younger, but not in middle-aged and older tribute to high levels of engagement in goal Baltes, P. B., & Baltes, adults, and that maintenance orientation is pursuit among older adults in the face of age- M. M. (1990). Psycholog- positively associated with well-being in older associated declines in external and internal ical perspectives on suc- but not in younger and middle-aged adults resources (Riediger, Freund, & Baltes, 2005; cessful aging: The model of selective optimization (Ebner et al., 2006). Riediger & Freund, 2006; see Figure 3). with compensation. In P. B. Baltes & M. M. Baltes (Eds.), Successful aging: Perspectives from the behavioral sciences 8 (pp. 1–34). New York: Cambridge University Press. 7

2 Goal orientation (mean across self-reported goals) Figure 2. Adult age- 1 group differences in mo- Younger adults Middle-aged adults Older adults tivational orientation of self-reported personal goals: Goal orientations Growth Maintenance Loss-prevention shift from growth toward maintenance and Note. Error bars represent two standard errors. loss-prevention throughout adulthood (adapted from Ebner et al., 2006).

Center for Lifespan Psychology | 157 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 158

Key References Developmental psychologists generally agree positively associated with evaluations of part- Riediger, M., & Freund, that individual development is embedded into nership quality. Furthermore, individual A. M. (2006). Focusing social systems. However, research on develop- salience of assumed dyadic goals, as indicated and restricting: Two as- mental regulation has rarely examined individ- by centrality and longer term stability of re- pects of motivational selectivity in adulthood. uals’ social contexts (for overviews, see Freund ported dyadic goals, predicted individuals’ own Psychology and Aging, & Riediger, 2005, 2006). The DRAMA project and their partners’ positive evaluations of 21, 173–185. seeks to take a direct look at developmental partnership quality as well as improvements in Riediger, M., Freund, goals and goal-directed behavior in social these evaluations over time. A. M., & Baltes, P. B. (2005). Managing life contexts. In an initial study completed in In a second line of research, we examine the through personal goals: 2005, Michaela Riediger and Antje Rauers lon- use and usefulness of dyadic knowledge Intergoal facilitation and gitudinally investigated the role of dyadic goal (“common ground”) among younger and older intensity of goal pursuit in younger and older representations for partnership development adults. We assume that the use of common adulthood. Journal of in 69 couples. We defined dyadic (“we”) goals ground facilitates mutual understanding, but Gerontology: Psychologi- as mental representations of states that “we also draws upon cognitive resources. There- cal Sciences, 60B, P84–P91. as a couple” seek to obtain, maintain, or avoid fore, we predict: (a) that younger adults use in the future. Results cast some doubt on the common ground more often than older notion that dyadic goals are collective-level adults; but (b) that older adults, whenever phenomena. For instance, partners were not they use common ground, profit more from it aware of how well they corresponded with than younger adults. their partners in ideas about dyadic goals, and We currently test these predictions in a Figure 3. Throughout the extent of dyadic-goal correspondence be- quasi-experimental study with a modified adulthood, motivational tween partners was unrelated to partnership- version of the collaborative-cognition game selectivity increases (in terms of both restricting quality evaluations. Dyadic-goal representa- Taboo (Dissertation Antje Rauers). Forty to few goals and focus- tions, however, nevertheless contributed to younger and 42 older couples participated in ing on central and simi- positive partnership development, but prima- the study. Data collection started in 2006, lar goals), contributing to higher intergoal facil- rily on the basis of their individual-level char- and will end in Spring 2007. itation and higher levels acteristics. Individuals’ subjective judgment of The third line of research in this subproject of involvement in goal assumed dyadic-goal correspondence was uses experience-sampling methods to explore pursuit (adapted from Riediger & Freund, 2006).

Few goals Instrumental Strategies .56** .96** .73**

.33** T1: Restricting .36 ns Age 2 .32* R = .19 .40** T1: Facilitation T2: Goal pursuit Age2 .42** R 2 = .50 R 2 = .16 .20* T1: Focusing .55** .94** .84* .76** R 2 = .20 .85** .68** Energy Time Daily Similar goals Central goals

χ2(df = 29) = 48.1*; CFI = .97; NNFI = .94; RMSEA = .06 (90% CI: .03 – .09). Note. Standardized solution ** p < .01; * p < .05; ns p > .05.

158 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 159

the regulation of affect, motives, and abilities This study is undertaken in collaboration with in everyday life. Alexandra M. Freund from the the Socioeconomic Panel (SOEP), and is University of Zurich and Michaela Riediger in- funded by the Federal Ministry for Education, vestigated whether negative affective experi- Science, Research, and Technology. Gert ences help to activate goal-directed action. Wagner from the German Institute for Eco- Results suggest that negative affective experi- nomic Research is a co-investigator in this ences play a role in regulating action, espe- project. The main goal of this study is to cially when the need for regulation is high. For chart everyday life processes of affective example, we found that the negative affective self-regulation from adolescence to older valence of goal-detrimental activities was par- adulthood—their frequency and form, effec- ticularly pronounced when participants dealt tiveness, and attentional requirements. Self- with overcoming difficulties. In addition, neg- report and cognitive measures will be ative affective tone predicted subsequent assessed in everyday contexts using JAVA- decrease in goal-detrimental behaviors. programmable mobile phones. We also investigated age-related differences in the implications of motivational conflict Subproject II: Normative Beliefs About experiences for affective well-being in daily the Lifespan as a Schema in Learning and life. Results showed that affective reactivity Memory to motivational conflicts increased, but that Lifespan theoreticians have argued repeat- the prevalence of motivational conflicts de- edly that normative beliefs about the lifespan creased with age. The decreased prevalence influence our perceptions and actions in vari- of motivational conflicts accounted for age- ous ways. For instance, adults of all ages are related improvements in average emotional more likely to expect that individuals are try- well-being, suggesting that older adults may ing to avoid losses when they are older than be motivated to situations that evoke nega- when they are younger (Freund & Ebner, tive affect or motivational conflicts. 2005). At the same time, research on adult A second study using experience-sampling age differences in episodic memory by Marcia procedures has been initiated in late 2006. Johnson, Moshe Naveh-Benjamin, and others

Figure 4. Age-associated 100 increase in schema reliance: Older adults are 90 more likely than younger adults to erroneously at- 80 tribute growth goal ori- 70 entations to young faces and loss-prevention goal 60 orientations to old faces. 50 40 30 20 10 orientations of correctly recognized faces Percentage of erroneous attribution goal Percentage 0 Younger adults Older adults

Erroneous attribution of growth orientation to younger faces Erroneous attribution of loss-prevention orientation to older faces

Note. Error bars represent two standard errors.

Center for Lifespan Psychology | 159 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 160

suggests that older adults may be less able to faces presented with a growth orientation form, consolidate, and retrieve associations during encoding, the likelihood to erro- that are at odds with their schematic beliefs. neously associate these faces with loss-pre- Therefore, they should rely to a greater ex- vention during recognition was again very tent on lifespan schemas when learning and high among older adults, and lower among retrieving schema-relevant new associations. younger adults (see Figure 4). Natalie Ebner, Michaela Riediger, and Ulman In addition to goal orientation, the second Lindenberger tested these predictions in two study also varied the emotional valence of recent studies, one completed in 2006 and the face (happy vs. sad) in a fully crossed de- the other ongoing. In both studies, we used a sign. The design of this study required the recognition-memory paradigm in which par- creation of a new set of emotional facial ticipants are presented adult faces of various stimuli, as available stimulus materials do not ages along with one of two developmental fully cross the age of the face with emotional goal orientations (growth vs. loss-preven- expression. In collaboration with professional tion), and are later asked to recognize the photographers and digital media designers, initially presented faces and the developmen- we compiled a database of naturalistic pho- tal goal orientations with which they were tographs of younger, middle-aged, and older associated. The results of the first study were women and men, each showing two in- consistent with our predictions. Specifically, stances of six different facial expressions for younger faces presented with a loss-pre- (neutrality, sadness, disgust, fear, anger, and vention goal during encoding, the likelihood happiness; see Figure 5). This database, called to erroneously associate these faces with FACES, will be made available for research growth during recognition was very high purposes as soon as results from an ongoing among older adults, and considerably lower validation study have been analyzed and sub- among younger adults. Conversely, for older mitted for publication.

Figure 5. A lifespan database of six facial expressions (FACES)—sample photographs.

160 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 161

Subproject III: Experimental Investigation focus on deviations between self-selected of Selection, Optimization, and and maximally manageable number of Compensation Mechanisms From simultaneous tasks, which we refer to as se- Childhood to Old Age lection margins (see Figure 6). In an ongoing The general goal of this subproject is to study started in 2006, we are currently col- better understand lifespan changes in devel- lecting data on children, teenagers, younger opmental regulation by investigating the adults, and older adults playing a modified mechanisms of selection, optimization, and version of the BINGO task, in which called- compensation in age-comparative quasi- out numbers have to be found on computer- experiments. The subproject is partially ized game cards. Overestimating one’s per- funded by a research grant from the German formance potential and choosing too many Research Foundation (Deutsche Forschungs- cards reduces the chance to reach a high gemeinschaft, Forschergruppe 778, Conflicts score (progressive selection margin) just like as Signals in Cognitive Systems). Shu-Chen Li underestimating one’s potential and choos- is a scientific investigator in this project, ing too few cards (conservative selection together with Sabine Schäfer, Michaela margin). We assume that children and older Riediger, and Ulman Lindenberger. adults will show more variable performances So far, the subproject has examined the se- and selection margins, whereas younger lection mechanism of the SOC model. Given adults will be most accurate in estimating the predominantly positive, relatively bal- their maximum manageable task difficulty. anced, and predominantly negative gain- We furthermore expect that children and loss ratio of developmental resources in teenagers will show progressive selection childhood, adulthood, and old age, respec- margins and that older adults will show rel- tively, we expected lifespan age differences atively conservative selection margins. Such in the operation of the selection mecha- lifespan differences in selection margins nism. In particular, we propose the concept may mirror differences in developmental of selection margins to study the develop- trajectories, such as the experience and an- ment of adaptive resource allocation ticipation of improvement in childhood, and processes in multiple-task situations. We of losses in later adulthood.

Figure 6. Schematic diagram of selective mar- Maximum manageable difficulty gins defined as discrep- ancies between the number of multiple tasks an individual could maximally manage given the available processing resources and the number of tasks he or she actually selects to work on.

Conservative Progressive selection margin selection margin Performance { {

Number of simultaneous tasks

Center for Lifespan Psychology | 161 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 162

Scientific Research Project 5 Investigators Interactive Brains, Social Minds

Ulman Lindenberger In everyday life, people often need to coordinate their actions with each other. Some common Shu-Chen Li examples are walking with someone at a set pace, playing collective sports, dancing, playing Viktor Müller music in duet or a group as well as a wide range of social bonding behaviors, such as eye-gaze Michaela Riediger coordination and joint attention between mother and infant or between partners. Little if any is known about brain mechanisms implementing interpersonally coordinated behavior. The In- Franziska Kopp teractive Brains, Social Minds project investigates lifespan changes in behavioral and neuronal (postdoctoral fellow) mechanisms that permit individuals to coordinate their behavior with each other in time and space. The empirical focus is on temporal aspects of interpersonal action coordination as Karen Bartling assessed by simultaneous behavioral and EEG recordings. General and ontogenetically early as- Anna Kleinspehn pects of social activities, such as joint gaze, imitation, and turn-taking are assumed to require (predoctoral fellows) interpersonal action coordination. The project has begun to develop several experimental paradigms to study the ontogeny of interpersonal action coordination across the lifespan.

Conceptual Introduction requires the perception, representation, and Natural systems often show a tendency to anticipation of other persons’ actions. Theo- synchronize (e.g., swarms of bees, schools of ries of ideomotor processes specify how ac- fish, or clapping of audiences). As a part of tions of others selectively influence one’s own this general phenomenon, the primary focus actions, as can be observed, for example, in of the project is on interpersonal synchroniza- mimicry or imitation. Informed by the work of tion in the context of goal-directed actions Steve Baker, Jean Decety, Chris Frith, between two or more human agents. We as- Wolfgang Prinz, Daniel Wolpert, and others, sume that interpersonal action coordination Figure 1 depicts a forward model of interper-

Figure 1. Simplified dyadic forward model of Predicted sensory feedback interpersonal action co- (single and joint actions) ordination. (The single- Representation Forward Forward person part of the model of shared dynamic output is grey-shaded.) forward model model model

Representation Forward Forward C of other’s dynamic output forward model model model

Forward Forward Efference copy dynamic output model model

Sensory C discrepancy

Motor Motor Sensory command system system

Sensory system External Actual sensory feedback influences (reafferences; single and joint action)

162 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 163

sonal action coordination. The model moves ers, such as eye gaze, facial expression, or Key References from the single-person to the two-person pointing, and that these cues facilitate Drewing, K., Aschers- case by assuming two additional representa- focusing of attention and object processing. leben, G., & Li, S.-C. tional layers beyond the representation of the We hypothesize that oscillatory activity in (2006). Sensorimotor synchronization across person’s own actions. The first additional layer the infant and in the adult brain may the life span. Interna- represents the assumption that individuals synchronize in situations when the infant tional Journal of Behav- continuously update and revise representa- follows social cues of the adult toward an ioral Development, 30, 280–287. tions about the other individual’s forward external object or event. The interbrain Striano, T., Kopp, F., model and sensory feedback. To achieve syn- coupling is hypothesized to be particularly Grossmann, T., & Reid, chronization, individuals need to coordinate prominent in situations when the interactive V. M. (2006). Eye contact the forward model regarding their own ac- partner is the infant’s parent (see also sub- influences neural processing of emotional ex- tions with the forward model regarding the project below). pressions in 4-month-old other person’s actions. The second additional The human brain undergoes major anatomi- infants. Social Cognitive layer, also shown in Figure 1, tentatively cap- cal and functional changes during the first and Affective Neuro- science, 1, 87–94. tures the possibility that highly skilled indi- years of life. Therefore, the task to relate os- Baltes, P. B., & viduals may represent jointly performed activ- cillatory activity in adults, compared to in- Staudinger, U. M. (1996). ities as a unified supra-personal action with a fants, will not be easy. To address these Interactive minds in a joint forward model and partially joint sen- methodological issues, we assessed the EEG life-span perspective: Prologue. In P. B. Baltes sory feedbacks. We assume that this privi- in 4-month-, 8-month-, and 12-month-old & U. M. Staudinger leged format may emerge from the two lower infants and in younger adults in a passive au- (Eds.), Interactive minds: layers in the course of acquiring skills that ditory oddball task. Standard 1000 Hz tones Life-span perspectives on the social foundation of place high demands on interpersonal synchro- and deviant 800 Hz tones were presented cognition (pp. 1–32). nization, such as acrobatics, dancing, or while participants were watching a movie New York: Cambridge chamber music. A central objective of the (Figure 2). Results based on this standard EEG University Press. project is to explore the temporal dynamics of paradigm will ascertain age-graded differ- these crossed reciprocal feedback loops from ences in general properties of oscillatory a lifespan perspective. activity, such as frequency bands, power, and phase, and will help in the implementation Interpersonal Action Coordination in and analysis of EEG measures taken during Principal Investigator Mother-Infant Interactions mother-infant interactions. Franziska Kopp Research in the project’s infant laboratory, which has been set up in 2006, will focus on Lifespan Development of Interpersonal behavioral and neural correlates of mother- Action Coordination infant interactions. Already in their first year This subproject investigates the development of life, infants perceive others as social part- of interpersonal behavioral synchronization ners and learn through imitation, contin- across the lifespan. Using a dyadic drumming gency detection, or joint attention. It is paradigm, we investigated lifespan age dif- known that infants use social cues from oth- ferences in dyadic action synchrony in four

Figure 2. EEG recording while an infant is watching a movie during the passive auditory oddball task.

Center for Lifespan Psychology | 163 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 164

Figure 3. Dyadic drum- nate their actions by assessing activities that ming paradigm for impose high synchronization demands. Little studying interpersonal action coordination. if any is known about brain mechanisms implementing interpersonally coordinated behavior. These mechanisms most likely will have to meet two constraints: (a) They need to be sufficiently fast to permit fluidity in interpersonal action coordination; (b) they need to integrate and regulate sensory, motor, and brain activity to generate and sustain action coordination between two or more persons. Synchronous oscillatory brain activi- age groups covering early childhood to old ties appear to meet both criteria. First, brain age (5-, 12-, 20- to 30-, and 70- to 80-year- oscillations bind spatially distributed, but olds; Dissertation Anna Kleinspehn). Partici- functionally related, information at the level pants were asked to synchronize their drum- of individual neurons, cell assemblies, and ming either with metronome frequencies of cortical areas. Onset times and frequency human-like variability or with four different ranges of coherent or synchronized oscilla- interaction partners in same- and mixed-age tions are sufficiently fast to permit, in princi- dyads (see Figure 3). Preliminary results in- ple, the speed and precision of information dicate that the precision of interpersonal exchange required by interpersonal action synchronization is highest in younger adults, coordination. followed by older adults, older children, By simultaneously recording the EEG of two and younger children (see Figure 4). Further people, we measured brain electrophysiologi- analyses will focus on the role of flexibility cal activities from eight pairs of guitarists and relevant social abilities in the develop- while playing a short melody together over ment of interpersonal action synchronization, about 60 trials (see Figure 5). Synchronous and on the association between dyadic asyn- brain activities within and between brains chrony and the subjective experience of the were investigated and analyzed by Phase interaction partner. Locking Index (PLI) and Interbrain Phase Coherence (IPC), respectively. We found that Principal Investigator Brain Mechanisms of Interpersonal Action phase synchronization both within and be- Viktor Müller Coordination tween brains increased significantly during This subproject seeks to identify cortical the periods of (i) preparatory metronome mechanisms of individuals’ ability to coordi- tempo setting and (ii) coordinated play onset

18,000 16,000 Younger child—younger child Older child—younger child 14,000 Older adult—younger child 12,000 Younger adult—younger child 10,000 Older child—older child 8,000 Older adult—older child Younger adult—older child 6,000 Older adult—younger adult Figure 4. Mean dyadic 4,000

Mean dyadic asynchronomy Older adult—older adult asynchrony of different 2,000 Younger adult—younger adult age-group combinations: Young children show 0 highest asynchrony, es- 12345678 pecially when drumming Trials in same-age dyads.

164 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 165

(Figure 6). Phase alignment extracted from Figure 5. One pair of gui- within-brain dynamics was related to behav- tarists with electrode caps during the experi- ioral play onset asynchrony between gui- mental session. tarists. We tentatively propose that oscillatory couplings between brains provide the neurophysiologic basis for interpersonally coordinated voluntary actions, which may serve as a basic mechanism for social interaction.

(A) Synchronization within the brains Guitarist A Guitarist B 20 20

15 15

10 10

Frequency (Hz) 5 Frequency (Hz) 5

0 0 –1 0 1 2 –1 0 1 2 Time (s) Time (s)

(B) Synchronization between the brains AB BA 20 20

15 15

10 10

Frequency (Hz) 5 Frequency (Hz) 5

0 0 –1 0 1 2 –1 0 1 2 Time (s) Time (s) min max n.s.

Figure 6. Phase synchronization within and between the brains during the period of guitar playing. (A) Time-frequency diagrams of the average Phase Locking Index (PLI) for guitarists A and B separately. PLI was averaged across six fronto-central electrodes showing PLI maxima (F3, Fz, F4, C3, Cz, and C4). Only significant PLI values (p < 0.003) are highlighted. Time zero is time locked to play onset of the leading guitarist A. The leading guitarist’s finger gesture to start playing together is depicted with the red arrow. The yellow arrows marked single guitar notes as recorded. High phase synchronization in both guitarists took place not only at play onset but also at the time point of gesture to start, and at the individual guitar strikes. (B) Time-frequency diagram of the average IPC averaged across six electrode pairs. On the left diagram (A > B), the selected electrode pairs represent interphase coherence between one electrode of guitarist A (Cz) to the six fronto-central electrodes of subject B. On the right diagram (B > A), the selected electrode pairs represent phase coherence between one electrode of guitarist B (Cz) to the six fronto-central electrodes of subject A. Only significant IPC values (p < 0.003) are highlighted. High phase synchronization between the guitarists took place at play onset.

Center for Lifespan Psychology | 165 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 166

Scientific Research Project 6 Investigators Formal Methods in Lifespan Psychology

Timo von Oertzen Since its foundation by the late Paul B. Baltes in 1981, the Center for Lifespan Psychology has (as of 02/2006) sought to promote conceptual and methodological innovation within developmental psychol- Shu-Chen Li ogy and in interdisciplinary context. Over the years, the critical examination of relations Ulman Lindenberger among theory, method, and data has evolved into a distinct feature of the Center. The Meth- ods Project of the Center continues this tradition with an added emphasis on computer algebra and computational neuroscience.

Currently, the Methods Project has three behavioral change, and discourage its use for main research foci. First, it investigates the this purpose (see also Lindenberger & Pötter, properties of statistical methods commonly 1998). used in developmental psychology, with an A second line of formal analysis and simula- emphasis on structural equation modeling tion aimed at the issue of statistical power and maximum likelihood estimation. Second, in structural equation modeling (Hertzog, the project develops neurocomputational Lindenberger, Ghisletta, & von Oertzen, models to integrate lifespan theorizing across 2006). Variants of structural equation mod- behavioral and neuronal levels of analysis. eling, such as latent growth curve models, Third, and more recently, the project has are widely used in developmental research, started to work on brain signal analysis. but surprisingly little is known about their statistical power. For instance, researchers Principal Investigator Statistical Modeling are generally unaware about how many indi- Timo von Oertzen Together with Paolo Ghisletta (University of viduals or occasions they would need to de- Geneva), Christopher Hertzog (Georgia Insti- tect individual differences in change of a tute of Technology, Atlanta), and Jack McAr- given magnitude, or to detect covariations dle (University of Southern California), this between two change processes across indi- Key References part of the project uses formal mathematical viduals. Hertzog, C., Linden- analysis and Monte Carlo simulations to ex- Figure 1, adapted from Hertzog et al. (2006), berger, U., Ghisletta, P., & von Oertzen, T. amine statistical properties of methods that displays the power to detect the covariance (2006). On the power of are widely used in developmental research. of change for two linear change processes, multivariate latent growth curve models to A first line of formal analysis concerned the mapping reliability (x-axis) against power detect correlated relation between correlated individual differ- (y-axis) for different numbers of longitudinal change. Psychological ences in longitudinal change and correlations occasions. Even with large samples (N = 200 Methods, 11, 244–252. among variables as they are observed in or 500) and several longitudinal occasions (4 Lindenberger, U., & cross-sectional age-heterogeneous data sets. or 5), statistical power to detect covariance Pötter, U. (1998). The complex nature of By decomposing cross-sectional correlations of slopes was moderate to low unless growth unique and shared in terms of bivariate linear growth, we for- curve reliability at study onset was above .80. effects in hierarchical mally showed that longitudinal change in a Thus, studies using latent growth curve mod- linear regression: Con- sequences for cross- mediator variable that shares large amounts els may fail to detect slope correlations due sectional developmental of age-related variance with the target vari- to low power rather than the lack of rela- research. Psychological able does not have to correlate with longitu- tionship of change between variables. In Methods, 3, 218–230. dinal change in the target variable. Con- combination with other results on the power versely, longitudinal change in a mediator to detect variances of change, our results variable that does not share any age-related allow researchers to make more informed variance with the target may nevertheless design decisions when planning a longitudi- correlate highly with longitudinal change in nal study. Current extensions of this work in- the target variable. We conclude that media- clude the gain in power associated with the tor analyses performed on the basis of cross- use of multiple indicators at each point in sectional data provide invalid approximations time, and the programming of a statistical to the dimensionality and causal structure of tool that allows researchers to perform

166 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 167

Figure 1. The statistical power to detect covari- (N = 500; r = .25) (N = 500; r = .50) (N = 500; r = .75) ance of change in latent 1.0 1.0 20 1.0 20 10 growth curve models. 20 10 0.8 0.8 0.8 Satorra-Saris power 6 10 plots mapping reliability 0.6 0.6 0.6 5 6 4 (x-axis) against power Power 0.4 0.4 5 0.4 6 3 5 4 (y-axis) for different 0.2 4 0.2 3 0.2 3 numbers of longitudinal 0.0 0.0 0.0 occasions (organized 0.5 0.6 0.7 0.8 0.9 1.0 0.5 0.6 0.7 0.8 0.9 1.0 0.5 0.6 0.7 0.8 0.9 1.0 from left to right as 20, 10, 6, 5, 4, and 3 occa- (N = 200; r = .25) (N = 200; r = .50) (N = 200; r = .75) sions). The top row cor- 20 1.0 1.0 20 1.0 responds to N = 500, the 10 bottom row to N = 200. 0.8 0.8 10 0.8 20 The columns represent 0.6 0.6 0.6 slope-slope correlations 10 6 Power 0.4 0.4 6 0.4 (from left to right) of .25, 6 5 5 5 4 4 4 3 0.2 3 0.2 3 0.2 .50, and .75. 0.0 0.0 0.0 0.5 0.6 0.7 0.8 0.9 1.0 0.5 0.6 0.7 0.8 0.9 1.0 0.5 0.6 0.7 0.8 0.9 1.0 Growth curve reliability Growth curve reliability Growth curve reliability

power calculation for a wide range of devel- neuromodulation, which then leads to noisier Key References opmental research designs. neuronal signaling, less separable processing Li, S.-C., von Oertzen, To further facilitate the computation of sta- pathways, and less differentiated cortical T., & Lindenberger, U. tistical power in structural equation model- representations (see also Project 1, pp. 136- (2006). A neurocomputational model of ing, Timo von Oertzen is currently exploring 146). stochastic resonance and power equivalence, or the relation between During the 2005–2006 periods, this line of aging. Neurocomputing, two structural equation models that have the inquiry was extended in two directions. First, 69, 1553–1560. same statistical power to detect a given ef- together with Moshe Naveh-Benjamin (Uni- Li, S.-C., Lindenberger, U., & Sikström, S. fect. The main purpose of this work is to re- versity Missouri-Columbia), the project inves- (2001). Aging cognition: duce complex models into simple models tigated the relation between age changes in From neuromodulation with equivalent power. Power-equivalent neuromodulation and associative binding to representation. Trends in Cognitive Sciences, 5, transformations would allow researchers to deficits in old age (Li, Naveh-Benjamin, & 479–486. analytically arrive at “iso-power lines” pro- Lindenberger, 2005; see also Project 1, jecting design parameters, such as the num- p. 136). Our results support the conjecture ber of measurement occasions, the number of that neuromodulatory processes act as a indicators per measurement, and the reliabil- basic memory binding mechanism by affect- ity of measurement onto simpler computa- ing the efficiency of distributed conjunctive tionally more tractable models. coding. Currently, we are developing multi- module networks to study interactions be- Neurocomputational Modeling tween reward-related phasic dopaminergic Principal In a series of neurocomputational investiga- modulation in the midbrain and aging- Investigators tions, Shu-Chen Li and colleagues have pro- related deficits in tonic dopaminergic modu- Timo von Oertzen moted the use of neural networks as a com- lation in prefrontal and hippocampal regions Shu-Chen Li putational platform for cross-level integra- in tasks involving attention, working memory, tion of lifespan differences in cognitive and and episodic memory. neuronal processes (e.g., Li, Lindenberger, & Second, we studied the effects of aging- Sikström, 2001). The neurocomputational ap- related increments in intrinsic neuronal proach proposed by Li et al. (2001) represents noise on stochastic resonance, a fundamen- central aspects of cognitive senescence as a tal property of physical and biological sys- developmental cascade of suboptimal neu- tems, in which noise acts as an amplifier of ronal gain control that is caused by deficient weak signals (see Figure 2). We showed that

Center for Lifespan Psychology | 167 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 168

Figure 2. Stochastic resonance as a means to enhance the processing of weak signals. In (A) Panel (A), a grey scale picture of Big Ben is shown on the left, with all pixels below 100 re- moved. The figure is downscaled such that the building can hardly be recognized. In the two other pictures, noise is added such that the picture first shows clearly (B) (middle), and then fades again (right). In Threshold crossing spikes Panel (B), the operation of stochastic resonance is illustrated. A subthreshold signal is not recognizable unless an optimal amount of noise raises it above threshold. With too much noise, the signal blurs again and Subthreshold signal Noise + Signal can be detected less well. Panel (C) displays detection accuracy as a (C) function of noise. 1.5

1.0

0.5 Mutual information (bits) 0.0 0.00 0.02 0.04 0.06 0.08 0.1 Noise intensity

Principal Investigator systems with greater intrinsic neuronal Brain Signal Modeling Timo von Oertzen noise and less plasticity continue to exhibit This part of the project was initiated in sum- stochastic resonance at both single-neuron mer 2006. Here, we seek to refine methods in and network levels. However, the stochastic brain signal analysis as a means to improve resonance effect is smaller and, somewhat our understanding of lifespan changes in counter-intuitively, requires greater brain-behavior relations (e.g., Lindenberger, amounts of external noise for its operation Li, & Bäckman, 2006b). Currently, three lines (Li, von Oertzen, & Lindenberger, 2006; see of inquiry are being pursued. First, in collabo- Figure 3). Empirical evidence in line with ration with Oliver Stegle from Cambridge these predictions were independently ob- University, Timo von Oertzen is working on tained in an age-comparative study on ag- Bayesian methods for the localization of ing and vibrotactile perception by Cari Wells sources of neuronal activity on the basis of and colleagues from the University of British electro-encephalic recordings (EEG). Of par- Columbia. ticular interest is the simultaneous estima-

168 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 169

Figure 3. Stochastic reso- (A) nance (SR) in neural net- 0.20 works with different levels of stochastic gain (G) simulates the effects 0.16 of senescent changes in dopaminergic neuromod- 0.12 ulation on perceptual processing. (A) Stochas- tic resonance functions 0.08 at different levels of gain. Peaks of stochastic 0.04 resonance functions with smaller stochastic gains are lower and right ) 0.00 shifted. (B) Short-term 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 plasticity and stochastic no noise resonance. Stochastic (B) resonance functions af- 0.20 ter exposure to sub-

–accuracy threshold stimuli and external noise for different noise 0.16 levels of gain. (C) Long- term plasticity and 0.12 stochastic resonance. Stochastic resonance functions assessed after 0.08 extensive exposure to suprathreshold stimuli 0.04 and external noise.

0.00 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2

0.16

0.12

0.08

0.04

0.00 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Level of input noise (σ) G[1.0, 1.2] G[0.6, 0.8] G[0.4, 0.6]

tion of multiple sources defined by frequency, Second, we are investigating spiking neuron power, or spatial position for two or more models and synfire chains. Building on our brains of individuals performing the same work on stochastic resonance, we explore task. This comprises the brains of two indi- ways of modeling age-associated processing viduals who are performing a joint action, changes in integrate-and-fire neurons, and such as playing music together, as a special assess the effect of these changes on sto- case (see Project 5, p. 162). chastic resonance and the signaling robust-

Center for Lifespan Psychology | 169 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 170

ness of synfire chains. We plan to use such Third, we plan to relate our work on power models to simulate the emergence of electri- equivalence to EEG data analysis, with the cal fields in synchronized neuronal assem- goal to pool EEG data in ways that optimize blies, with the goal to link integrate-and-fire the statistical power to detect hypothesized models to simulated and empirical EEG data. effects.

The Center for Lifespan Psychology in February 2007

Left to right: Julia Delius, Michaela Riediger, Imke Kruse, Dorothea Hämmerer, Anna Kleinspehn, Karen Bartling, Franziska Kopp, Antje Rauers, Natalie Ebner, Ulman Lindenberger, Timo von Oertzen, Irene Nagel, Yee Lee Shing, Sabine Schäfer, Shu-Chen Li, Michael Schellenbach, Martin Lövdén, Julius Verrel, Viktor Müller, Christian Chicherio, Nilam Ram, Markus Werkle-Bergner.

170 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 171

Research Project 7 Scientific Toward a Psychological and Developmental Theory of Sehnsucht Investigator (Life Longings) † Paul B. Baltes One of the exciting events in lifespan psychology is the identification of novel constructs that appear worthwhile of investigation when considering the life course as a whole. With the Susanne Scheibe construct of “Sehnsucht” (German for longing, yearning), this project aims at capturing indi- (postdoctoral fellow) vidual and collective thoughts and feelings about one’s optimal or utopian life. The concept highlights two central topics of life in the modern world, a permanent search for perfection Dana Kotter-Grühn and a permanent sense of incompleteness, objectively and subjectively (Baltes, 1997). “Sehn- (predoctoral fellow) sucht” is difficult to translate into English. After much deliberation, we have chosen to char- acterize the emerging research program as the “psychology of life longings.” We view life longings as emotional and mental representations of personal peaks of life, akin to personal utopias. In principle, however, these personal utopias are unattainable and, therefore, ambivalent in emotional quality.

Introduction ings for development. Third, we tested the Key References In German everyday life, the concept of applicability of our theory of life longings to Baltes, P. B. (in press). "Sehnsucht” is salient. In fact, “Sehnsucht” cultural contexts other than Germany, such Entwurf einer Lebens- was the third most often nominated word in as the United States. spannen-Psychologie der Sehnsucht: Utopie eines a recent contest of “the most beautiful vollkommenen und per- German word” (Spiegel Online, 2004). Build- Delineating Life Longings From Other fekten Lebens. Psycholo- ing on everyday dictionary-like definitions of Concepts gische Rundschau. “Sehnsucht” on the one hand, and applying The most notable concept related to life Scheibe, S., Freund, A. M., & Baltes, P. B. principles of lifespan psychology (e.g., Baltes, longings is that of personal goals. Like life (2007). Toward a devel- 1997) on the other, we derived a family of longings, goals are directed at positive out- oopmental psychology of six characteristics to capture the structure of comes and give directionality to life. Goals Sehnsucht (life-longings): The optimal life longings (see Table 1). Empirical data operate primarily at the behavioral level: (utopian) life. Develop- confirmed this conceptualization (Scheibe, People strive at attaining their goals by for- mental Psychology, 43, Freund, & Baltes, in press). mulating specific implementation intentions 778–795. To assess life longings, we generally ask study and by engaging in goal-relevant behaviors. Baltes, P. B. (1997). On the incomplete architec- participants to undertake a guided “mental Life longings, in contrast, are utopian and ture of human ontogeny: journey through life” during which they are unattainable; they work mainly at the level Selection, optimization, asked to generate a list of personal life long- of imagination and fantasy (Boesch, 1998). and compensation as foundation of develop- ings. They then evaluate each of their most Confirming this view, we found that personal mental theory. American important life longings on a questionnaire goals were reported to be more concrete, Psychologist, 52, designed to measure the six structural char- controllable, and more closely linked to 366–380. acteristics and other important aspects of life everyday behavior than life longings. Personal longings. Sample items are given in Table 1. life longings, in contrast, were rated as more Reliability indices and confirmatory factor emotionally bittersweet than goals (e.g., as analyses indicate that life longings can be simultaneously painful and pleasurable; satisfactorily measured by self-report (Disser- Diploma Thesis Sabine Mayser). tation Susanne Scheibe). We further differentiated life longings from When introducing a new concept into the lit- regret, goals, and the conception of one’s erature, it is crucial to show its uniqueness in ideal self by examining people’s beliefs about comparison to an already existing well-artic- the typical states and traits associated with ulated concept. Therefore, one goal of our re- their activation. Table 2 lists the most typical search has been to investigate commonalities states and traits assigned to the four con- and differences with concepts, such as goals, cepts. Consistent with the defining charac- regrets, or ideal selves. A second goal has teristic of emotional ambivalence, life long- been to study possible functions of life longings were rated equally high on both positive

Center for Lifespan Psychology | 171 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 172

Table 1 A Family of Six Characteristics to Capture the Structure of Life Longings: Explanations and Sample Items From Questionnaire Designed to Measure Life Longings

Structural characteristics Sample items from of life longings Explanation questionnaire

Unrealizable Consistent with the idea that individuals hold I am longing for something that personal utopia subjective beliefs about their own optimal is too perfect to be true. development, life longings involve personal utopias of ideal life realities and optimal life courses. These are unattainable in principle.

Feeling of Life longings involve feelings of incomplete- My longing means that something incompleteness ness and imperfection of life. They reflect the essential is missing in my life. notion that development is a lifelong process that never reaches completion.

Ontogenetic Life longings focus on the life course as a In my longing, memories of my past tritime focus whole, that is, on the personal past, present, and fantasies about my future are and future. connected.

Ambivalent Life longings reflect the conception that My feeling of longing is both emotions development always entails both gains and painful and pleasurable. losses: They have an ambivalent, bittersweet, emotional quality.

Life reflection and Life longings elicit reflections and evaluations My longing makes me think a lot evaluation of life and one’s standing relative to ideals or about the meaning and sense of to others who serve as guideposts for optimal my life. development.

Symbolic richness Life longings comprise representations that What I am longing for is heavily are rich in symbolic meaning. Specific targets filled with meaning. of life longings are linked to encompassing mental-emotional representations for which they stand.

In collaboration with and negative states, and were linked with The Regulative Function of Life Longings Alexandra Freund, positive (e.g., openness, agreeableness, hav- in Adult Development and Aging University of Zurich ing a sense of purpose and directedness) and What is the role of life longings in planning, negative personality traits (e.g., introversion, managing, and experiencing one’s life? Do lack of environmental mastery, low self- they give direction or help to cope with loss acceptance). Goals and the ideal self were and unattainability? Are they indicators of rated higher on positive than negative states, positive or negative states and outcomes? and were linked with a well-adjusted person- Answers to such questions require analysis of ality profile. In contrast, regret was rated antecedent-consequent relationships, consid- higher on negative than positive adjectives, erations of different outcome measures, and and was linked to a maladaptive personality the recognition that such associations and profile (with the exception that people often causal connections may be nonlinear. having regrets were judged to have a sense In a first project, we found that life longings of purpose and directedness). These findings are perceived as functional (facilitative) in support the divergent validity of life longings, development (Dissertation Susanne Scheibe; demonstrating its uniqueness in relation to Scheibe, Freund, & Baltes, 2007). Participants other already existing concepts. reported that their life longings (1) provided

172 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 173

Table 2 Most Typical States and Traits Assigned to Life Longings and Other Concepts of Developmental Regulation (N = 109 younger and older adults)

Life longings Goals Regrets Ideal self

Five most typical states A person whose thoughts and feelings CURRENTLY revolve around life longings/ goals/regrets/his or her ideal self, typically feels …

dreamy ambitious sad ambitious sensual creative depressed creative empathic energetic frustrated persistent sad powerful silent inspired silent persistent helpless energetic

Three most typical traits A person whose thoughts and feelings OFTEN revolve around life longings/ goals/ regrets/ his or her ideal self, typically is …

someone who has an someone who has someone who feels someone who finds it active imagination an active imagination that the demands of important to have new everyday life often experiences that challenge get him/her down how he/she thinks about him-/herself and the world

someone who feels someone who does someone who feels someone who has an disappointed about his/ a thorough job disappointed about his/ active imagination her achievements in life her achievements in life

someone who feels that someone who feels in someone who gets someone who does a the demands of charge of the situation nervous easily thorough job everyday life often get in which he/she lives him/her down

a sense of directionality for development and a regulatory strategy to manage the nonful- (2) helped in regulating losses and unattain- fillment of important life goals (Dissertation able aspects of life. Seemingly contrary to Dana Kotter-Grühn). Here, we focused on this finding was a negative relationship be- the widely shared goal to have children. tween life longings and overall well-being Childless women aged 35 to 55 years rated (e.g., positive emotions, life satisfaction). In- the extent to which their wish for children dividuals with high-level expressions of life fulfilled the six structural characteristics of longings reported lower well-being. However, life longings as well as typical characteris- when persons also reported a strong sense of tics of goals (e.g., intention to pursue the control over the experience of life longings, wish actively). Results were consistent with the negative relationship was reduced. It can the idea that women were at different be speculated that life longings are linked points in the transformation process from with critical self-reflection about the devel- goals to life longings. The first group (22 % opmental progression toward personal ideals of the sample) described their wish for chil- of life. As such, strong life longings may lead dren as a strong goal, but only as a weak to lower well-being, but possibly promote life longing. Another group (30 %) reported personal growth (Scheibe, Kunzmann, & a strong goal and a strong life longing to Baltes, 2007, in press). have children. A third group (20 %) reported A second project investigated whether life a strong life longing, but only a weak goal. longings develop from unattainable goals, The remaining women reported a weak goal and whether this transformation represents and life longing.

Center for Lifespan Psychology | 173 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 174

Figure 1. Life satisfaction and depressivity in 6 groups of women with Strong goal + weak life longing either a strong or weak Strong goal + strong life longing goal to have children, 5 combined with either a Weak goal + strong life longing strong or weak life long- Weak goal + weak life longing ing to have children. 4 Women with a weak goal and life longing to have children had the highest 3 life satisfaction and lowest depressivity, whereas women with a strong 2 goal and life longing to have children had the lowest life satisfaction 1 and highest depressivity. Life satisfaction Depressivity The differences between these two groups were statistically reliable.

Does this transformation process lead to pos- English-language version of our question- itive outcomes? Replicating our earlier study, naire that is suitable for assessing life long- a stronger expression of life longing was re- ings in English-speaking samples. Initial lated to lower well-being. However, women evidence from an internet questionnaire with a strong expression of life longing to study with German and American partici- have children reported high well-being if pants (Diploma Thesis Maja Wiest) and from they used additional self-regulatory strate- a laboratory study in Atlanta, undertaken in gies, such as reengagement in alternative life collaboration with Fredda Blanchard-Fields, goals. Looking at the groups specified above, support the view that life longings can be as- women with a strong goal and life longing to sessed in cultural contexts outside Germany. have children had lower well-being than The reliabilities of the scales and the statisti- women with neither a goal nor life longing to cal fit of the six-dimensional structure of life have children (Figure 1). Women with a longings were satisfactory in the North strong life longing, but a weak goal, did not American samples, and factor loadings were differ from the other groups. Together, these similar across the two cultural contexts. At results suggest that pursuing an unattainable the level of latent constructs, low correla- goal as a life longing may have positive out- tions between personal utopia and the re- comes. However, this positive effect emerges maining five characteristics in the North only under certain conditions, and therefore American sample suggest that unattainability needs to be investigated in more detail. may not be linked to life longing experiences to the same degree as it is in Germany. Also, In cooperation with Going Beyond Germany: Sehnsucht in the relative to the German sample, participants Fredda Blanchard- American Culture in the North American sample reported fewer Fields, Georgia In German culture, the concept of “Sehn- life longings, and were more likely to report Institute of sucht” is salient, and its core connotations that the concept of life longings is difficult Technology, Atlanta have evolved during the cultural era of to understand and not often used in everyday Romanticism. The German word "Sehnsucht" life. At the same time, the two samples did is difficult to translate into English. Its basic not differ in the self-reported ease with characteristics, however, have been derived which they were able to identify their per- on the basis of lifespan theory, and are sonal life longings nor in their salience (in- thought to reflect basic universal aspects of tensity, frequency). lifespan development. To inquire the generality and cultural specificity of life longings, we developed an

174 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 175

Publications 2005–2006 (last update: May 2007)

Anokhin, A. P., Müller, V., der Sehnsucht: Utopie eines voll- The perspective of biocultural abilität und Plastizität. In M. Lindenberger, U., Heath, A. C., & kommenen und perfekten Lebens. co-constructivism. Cambridge, Hasselhorn & W. Schneider (Eds.), Myers, E. (2006). Genetic influ- Psychologische Rundschau. UK: Cambridge University Press. Handbuch der Psychologie: ences on dynamic complexity of Vol. Entwicklungspsychologie. brain oscillations. Neuroscience Baltes, P. B. (2007). Alter(n) als Göttingen: Hogrefe. Letters, 397, 93–98. Balanceakt: Im Schnittpunkt von Fortschritt und Würde. In P. Gruss Brehmer, Y., & Lindenberger, U. Bäckman, L., & Farde, L. (2005). (Ed.), Die Zukunft des Alterns: Die (in press). Kognitive Leistungs- The role of dopamine systems in Antwort der Wissenschaft reserven im höheren Erwachse- cognitive aging. In R. Cabeza, L. (pp. 15–34). München: Beck. nenalter: Befunde der Interven- Nyberg, & D. C. Park (Eds.), Cogni- tionsforschung. In F. Petermann & tive neuroscience of aging: Link- Baltes, P. B. (2006a). Alter(n) als W. Schneider (Eds.), Enzyklopädie ing cognitive and cerebral aging Balanceakt im Schnittpunkt von der Psychologie, Vol. 7. Entwick- (pp. 58–84). New York: Oxford Fortschritt und Würde: Vortrag lungspsychologie. Göttingen: University Press. gehalten anlässlich der Öffent- Hogrefe. lichen Jahrestagung des Natio- Bäckman, L., Jones, S., Berger, nalen Ethikrates. Berlin, 27. Ok- Charness, N., Tuffiash, M., A.-K., Jonsson Laukka, E., & Small, tober 2005. In E. Loewy (Ed.), Krampe, R. T., Reingold, E., & B. J. (2005). Cognitive impairment Schwerpunkt: Selbstbestimmtes Vasyukova, E. (2005). The role of in preclinical Alzheimer's disease: Sterben (pp. 210–225). Nürn- deliberate practice in chess A meta-analysis. Neuropsychol- berg: Gesellschaft für kritische expertise. Applied Cognitive ogy, 19, 520–531. Philosophie (Aufklärung und Kri- Psychology, 19, 151–165. tik, Sonderheft 11). Baltes, P. B., Reuter-Lorenz, P. A., Bäckman, L., Nyberg, L., & Farde, & Rösler, F. (2006). Preface and Delius, J. A. M., & von Hammer- L. (in press). Dopamine and cogni- Baltes, P. B. (2006b). Facing our acknowledgments. In P. B. Baltes, stein, N. (2005). International tive aging: State-of-the-art and limits: Human dignity in the very P. A. Reuter-Lorenz, & F. Rösler Max Planck Research Schools: future directions. In Q. Jing & H. old. Daedalus, 135, 32–39. (Eds.), Lifespan development and Neue Wege der Graduiertenaus- Zhang (Eds.), Proceedings from the brain: The perspective of bio- bildung. In U. Teichler & R. Tippelt the 28th International Congress Baltes, P. B. (Ed.). (2005). Theo- cultural co-constructivism (Eds.), Hochschullandschaft im of Psychology. London: Psychol- retical approaches to lifespan (pp. XIII–XIV). Cambridge, UK: Wandel (pp. 170–185). Wein- ogy Press. development: Interdisciplinary Cambridge University Press. heim: Beltz (Zeitschrift für Päda- perspectives [Special Issue]. Re- gogik, Beiheft 50). Bäckman, L., Nyberg, L., & Farde, search in Human Development, 2 Baltes, P. B., Rösler, F., & Reuter- L. (2007). Dopamine and cognitive (1–2). Mahwah, NJ: Erlbaum. Lorenz, P. A. (2006). Prologue: Delius, J. D., & Delius, J. A. M. aging: A strong relationship. In Biocultural co-constructivism as a (2006). Intelligences and brains: Q. Jing, M. R. Rosenzweig, G. Baltes, P. B., Freund, A. M., & Li, theoretical metascript. In P. B. An evolutionary bird's eye view. d’Ydewalle, H. Zhang, H.-C. Chen, S.-C. (2005). The psychological Baltes, P. A. Reuter-Lorenz, & F. In E. A. Wasserman & T. R. Zentall & K. Zhang (Eds.), Progress in science of human ageing. In M. Rösler (Eds.), Lifespan develop- (Eds.), Comparative cognition: Ex- psychological science around Johnson, V. L. Bengston, P. G. ment and the brain: The perspec- perimental explorations of animal the world: Congress proceedings. Coleman, & T. B. L. Kirkwood tive of biocultural co-construc- intelligence (pp. 555–579). New XVIII International Congress of (Eds.), The Cambridge handbook tivism (pp. 3–39). Cambridge, UK: York: Oxford University Press. Psychology, Beijing, 2004: Vol. 1. of age and ageing (pp. 47–71). Cambridge University Press. Neural cognitive and develop- Cambridge, UK: Cambridge Uni- mental issues (pp. 455–469). versity Press. Baltes, P. B., & Smith, J. (in press). Hove: Psychology Press. The fascination of wisdom: Its na- Baltes, P. B., & Kunzmann, U. (in ture, ontogeny, and function. Per- Bäckman, L., Nyberg, L., press). Wisdom and aging: The spectives on Psychological Science. Lindenberger, U., Li, S.-C., & road toward excellence in mind Farde, L. (2006). The correlative and character. In D. C. Park & N. Brandtstädter, J., & Linden- triad among aging, dopamine, Schwarz (Eds.), Cognitive aging: berger, U. (Eds.). (in press). Ent- and cognition: Current status and A primer. Hove: Psychology Press. wicklungspsychologie der future prospects. Neuroscience & Lebensspanne: Ein Lehrbuch. Biobehavioral Reviews, 30, Baltes, P. B., Lindenberger, U., Stuttgart: Kohlhammer. 791–807. & Staudinger, U. M. (2006). Life span theory in developmental Brehmer, Y., Li, S.-C., Müller, Bäckman, L., & Small, B. J. (in psychology. In W. Damon & R. M. V., von Oertzen, T., & Linden- press). Cognitive deficits in pre- Lerner (Eds.), Handbook of child berger, U. (2007). Memory plas- clinical Alzheimer's disease and psychology: Vol. 1. Theoretical ticity across the lifespan: Uncov- vascular dementia: Patterns of models of human development ering children's latent potential. findings from the Kungsholmen (pp. 569–664). New York: Wiley. Developmental Psychology, 43, Projekt. Physiology & Behavior. 465–478. Baltes, P. B., Reuter-Lorenz, P. A., Derwinger, A., Stigsdotter Neely, Baltes, P. B. (in press). Entwurf & Rösler, F. (Eds.). (2006a). Life- Brehmer, Y., & Lindenberger, U. A., MacDonald, S., & Bäckman, L. einer Lebensspannen-Psychologie span development and the brain: (in press-a). Intraindividuelle Vari- (2005). Forgetting numbers in old

Center for Lifespan Psychology | 175 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 176

age: Strategy and learning speed Freund, A. M., & Baltes, P. B. berger, U. (2007). Well-being af- list paradigm? Psychology and matter. Gerontology, 51, 277–284. (2005). Entwicklungsaufgaben als fects changes in perceptual speed Aging. Organisationsstrukturen von Ent- in advanced old age: Longitudinal Drewing, K., Aschersleben, G., & wicklung und Entwicklungsopti- evidence for a dynamic link. Grühn, D., Smith, J., & Baltes, Li, S.-C. (2006). Sensorimotor mierung. In N. Birbaumer et al. Developmental Psychology, 43, P. B. (2005). No aging bias synchronization across the life (Series Eds.), S.-H. Filipp & U. M. 705–718. favoring memory for positive span. International Journal of Staudinger (Vol. Eds.), Enzyklo- material: Evidence from a Behavioral Development, 30, pädie der Psychologie: Vol. C, V, Gerstorf, D., Ram, N., Röcke, heterogeneity-homogeneity list 280–287. 6. Entwicklungspsychologie C., Lindenberger, U., & Smith, paradigm using emotionally des mittleren und höheren Er- J. (in press). Decline in life satis- toned words. Psychology and Ebner, N. C., & Freund, A. M. (in wachsenenalters (pp. 35–78). faction in old age: Longitudinal Aging, 20, 579–588. press). Annäherungs- vs. Vermei- Göttingen: Hogrefe. evidence for links to distance-to- dungsmotivation. In V. Brand- death. Psychology and Aging. Herlitz, A., & Kabir, Z. A. (2006). stätter & J. H. Otto (Eds.), Hand- Freund, A. M., & Ebner, N. C. Sex differences in cognition buch der Allgemeinen Psycholo- (2005). The aging self: Shifting Gerstorf, D., Smith, J., & among illiterate Bangladeshis: gie: Motivation und Emotion. from promoting gains to balancing Baltes, P. B. (2006). A systemic- A comparison with literate Göttingen: Hogrefe. losses. In W. Greve, K. Rothermund, wholistic approach to differential Bangladeshis and Swedes. Scan- & D. Wentura (Eds.), The adaptive aging: Longitudinal findings from dinavian Journal of Psychology, Ebner, N. C., & Freund, A. M. self: Personal continuity and inten- the Berlin Aging Study. Psychol- 47, 441–447. (2007). Personality theories of tional self-development (pp. 185– ogy and Aging, 21, 645–663. successful aging. In J. A. Black- 202). Göttingen: Hogrefe. Hertzog, C., Lindenberger, U., burn & C. N. Dulmus (Eds.), Hand- Ghisletta, P., Bickel, J.-F., & Ghisletta, P., & von Oertzen, T. book of gerontology: Evidence- Freund, A. M., & Riediger, M. Lövdén, M. (2006). Does activity (2006). On the power of multi- based approaches to theory, (2006). Goals as building blocks of engagement protect against variate latent growth curve mod- practice, and policy (pp. 87– personality and development in cognitive decline in old age? els to detect correlated change. 116). Haboken, NJ: Wiley. adulthood. In D. K. Mroczek & T. Methodological and analytical Psychological Methods, 11, D. Little (Eds.), Handbook of per- considerations. Journal of Geron- 244–252. Ebner, N. C., Freund, A. M., & sonality development (pp. 353– tology: Psychological Sciences, Baltes, P. B. (2006). Develop- 372). Mahwah, NJ: Erlbaum. 61B, P253–P261. Hoppmann, C., Gerstorf, D., mental changes in personal goal Smith, J., & Klumb, P. L. (2007). orientation from young to late Freund, A. M., & Riediger, M. Ghisletta, P., & Lindenberger, U. Linking possible selves and behav- adulthood: From striving for gains (2005). Gerontopsychologie: Er- (2005). Exploring structural dy- ior: Do domain-specific hopes and to maintenance and prevention of folgreiches Altern. In J. Bengel namics within and between sen- fears translate into daily activities losses. Psychology and Aging, 21, et al. (Series Eds.), H. Weber & T. sory and intellectual functioning in very old age? Journal of 664–678. Rammsayer (Vol. Eds.), Handbuch in old and very old age: Longitu- Gerontology: Psychological der Psychologie: Vol. 2. Hand- dinal evidence from the Berlin Sciences, 62B, P104–P111. Erixon-Lindroth, N., Farde, L., buch der Persönlichkeitspsycho- Aging Study. Intelligence, 33, Robins-Wahlin, T.-B., Sovago, J., logie und differentiellen Psycho- 555–587. Hoppmann, C., & Smith, J. (2007). Halldin, C., & Bäckman, L. logie (pp. 564–571). Göttingen: Life-history related differences in (2005). The role of the striatal Hogrefe. Ghisletta, P., McArdle, J. J., & possible selves in very old age. In- dopamine transporter in cognitive Lindenberger, U. (2006). Longi- ternational Journal of Aging and aging. Psychiatry Research: de Frias, C. M., Lövdén, M., tudinal cognition-survival rela- Human Development, 64, 109–127. Neuroimaging, 138, 1–12. Lindenberger, U., & Nilsson, L.- tions in old and very old age: 13- G. (2007). Revisiting the dediffer- year data from the Berlin Aging Huxhold, O., Li, S.-C., Fiedler, K., Kliegl, R., Lindenberger, entiation hypothesis with longitu- Study. European Psychologist, 11, Schmiedek, F., & Lindenberger, U., Mausfeld, R., Mummendey, A., dinal multi-cohort data. Intelli- 204–223. U. (2006). Dual-tasking postural & Prinz, W. (2005). Psychologie gence, 35, 381–392. control: Aging and the effects of im 21. Jahrhundert: Führende Glück, J., & Baltes, P. B. (2006). cognitive demand in conjunction deutsche Psychologen über Lage Gerstorf, D., Herlitz, A., & Using the concept of wisdom to with focus of attention. Brain und Zukunft ihres Fachs und die Smith, J. (2006). Stability of sex enhance the expression of wis- Research Bulletin, 69, 294–305. Rolle der psychologischen Grund- differences in cognition in ad- dom knowledge: Not the philoso- lagenforschung. Gehirn & Geist vanced old age: The role of edu- pher's dream but differential ef- Jones, S., Jonssn Laukka, E., & (7–8), 56–60. cation and attrition. Journal fects of developmental prepared- Bäckman, L. (2006). Differential of Gerontology: Psychological ness. Psychology and Aging, 21, verbal fluency impairment in the Fischer, H., Sandblom, J., Gavazzeni, Sciences, 61B, P245–P249. 679–690. preclinical stages of Alzheimer’s J., Fransson, P., Wright, C. I., & disease and vascular dementia. Bäckman, L. (2005). Age-differen- Gerstorf, D., & Lindenberger, U. Greve, W., & Ebner, N. C. (2007). Cortex, 42, 347–355. tial patterns of brain activation (2006). Vergessen [Forgetting]. In Intentional self-development—ge- during perception of angry faces. J. Bengel et al. (Series Eds.), J. netically framed. European Jour- Jones, S., Livner, A., & Bäckman, Neuroscience Letters, 386, 99–104. Funke & P. A. Frensch (Vol. Eds.), nal of Developmental Sciences, L. (2006). Patterns of prospective Handbuch der Psychologie: Vol. 5. 1, 82–94. and retrospective memory impair- Freund, A. M. (2006). Age-differ- Handbuch der allgemeinen Psy- ment in preclinical Alzheimer's ential motivational consequences chologie: Kognition (pp. 371–378). Grühn, D., Scheibe, S., & disease. Neuropsychology, 20, of optimization versus compensa- Göttingen: Hogrefe. Baltes, P. B. (in press). Reduced 144–152. tion focus in younger and older negativity effect in older adults? adults. Psychology and Aging, 21, Gerstorf, D., Lövdén, M., Memory for emotional pictures: Jones, S., Nyberg, L., Sandblom, J., 240–252. Röcke, C., Smith, J., & Linden- The heterogeneity–homogeneity Stigsdotter Neely, A., Ingvar, M.,

176 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 177

Petersson, K. M., & Bäckman, L. chologie: Vol. C, V, 3. Soziale, Kunzmann, U., Stange, A., & distinctiveness and conjunctive (2006). Cognitive and neural plas- emotionale und Persönlichkeits- Jordan, J. (2005). Positive affec- binding: Computational explo- ticity in aging: General and task- entwicklung (pp. 617–675). tivity and lifestyle in adulthood: rations of possible links. In H. D. specific limitations. Neuroscience Göttingen: Hogrefe. Do you do what you feel? Person- Zimmer, A. Mecklinger, & U. & Biobehavioral Reviews, 30, ality and Social Psychology Lindenberger (Eds.), Handbook of 864–871. Kreppner, K. (2005f). [Review of Bulletin, 31, 574–588. binding and memory: Perspec- the book Väter und Kinder: Lang- tives from cognitive neuro- Jopp, D., & Smith, J. (2006). Re- zeitstudien über väterliche Für- Li, K. Z. H., Krampe, R. T., & science (pp. 291–312). Oxford: sources and life-management sorge und die sozioemotionale Bondar, A. (2005). An ecological Oxford University Press. strategies as determinants of suc- Entwicklung von Kindern]. Zeit- approach to studying aging and cessful aging: On the protective schrift für Pädagogik, 51, dual-task performance. In R. W. Li, S.-C., Naveh-Benjamin, M., & effect of selection, optimization, 300–304. Engle, G. Sedek, U. von Hecker, & Lindenberger, U. (2005). Aging and compensation. Psychology D. N. McIntosh (Eds.), Cognitive neuromodulation impairs associa- and Aging, 21, 253–265. Kreppner, K. (2005g). Wilhelm T. limitations in aging and psy- tive binding: A neurocomputa- Preyer (1841–1897). In B. Hopkins chopathology (pp. 190–218). tional account. Psychological Krampe, R. T. (2006). Musical ex- (Ed.), The Cambridge Encyclope- Cambridge, UK: Cambridge Uni- Science, 16, 445–450. pertise from a lifespan perspec- dia of child development versity Press. tive. In H. Gembris (Ed.), Musical (pp. 523–524). Cambridge, UK: Li, S.-C., von Oertzen, T., & development from a lifespan Cambridge University Press. Li, S.-C. (2006). Biocultural co- Lindenberger, U. (2006). A perspective (pp. 91–105). Frank- construction of lifespan develop- neurocomputational model of furt a. M.: Lang. Kruse, I. (in press-a). Driven from ment. In P. B. Baltes, P. A. Reuter- stochastic resonance and aging. the outside: The EU’s impact on Lorenz, & F. Rösler (Eds.), Lifespan Neurocomputing, 69, Kray, J., & Lindenberger, U. (in Albanian immigration and asylum development and the brain: The 1553–1560. press). Fluide Intelligenz. In J. policies. In T. Faist & A. Ette (Eds.), perspective of biocultural co- Brandtstädter & U. Lindenberger The europeanization of national constructivism (pp. 40–57). Lindenberger, U. (in press). His- (Eds.), Entwicklungspsychologie immigration policies and politics. Cambridge, UK: Cambridge Uni- torische Grundlagen: Johann der Lebensspanne: Ein Lehrbuch. New York: Palgrave Macmillan. versity Press. Nicolaus Tetens als Wegbereiter Stuttgart: Kohlhammer. des Lebensspannen-Ansatzes in Kruse, I. (in press-b). Immigration Li, S.-C. (2005). Neurocomputa- der Entwicklungspsychologie. In J. Kreppner, K. (2005a). Analyzing reform in Germany: The domestic tional perspectives linking neuro- Brandtstädter & U. Lindenberger family interaction patterns from debate under the red-green gov- modulation, processing noise, (Eds.), Entwicklungspsychologie videotapes over time. In V. L. ernment. In O. Schmidtke (Ed.), representational distinctiveness, der Lebensspanne: Ein Lehrbuch. Bengtson, A. C. Acock, K. R. Allen, Governing migration in an age of and cognitive aging. In R. Cabeza, Stuttgart: Kohlhammer. P. Dilworth-Anderson, & D. M. denationalization. London: Rout- L. Nyberg, & D. Park (Eds.), Cogni- Klein (Eds.), Sourcebook of family ledge. tive neuroscience of aging: Link- Lindenberger, U. (2007). Tech- theory & research (pp. 80–82). ing cognitive and cerebral aging nologie im Alter: Chancen aus Thousand Oaks, CA: Sage. Kunzmann, U., & Baltes, P. B. (pp. 354–379). Oxford: Oxford Sicht der Verhaltenswissen- (2005). The psychology of wis- University Press. schaften. In P. Gruss (Ed.), Die Kreppner, K. (2005b). Erleben von dom: Theoretical and empirical Zukunft des Alterns: Die Antwort Veränderung: Pubertät in der challenges. In R. J. Sternberg & J. Li, S.-C., & Baltes, P. B. (2006). der Wissenschaft (pp. 221– Familie. In U. Lehmkuhl (Ed.), Die Jordan (Eds.), A handbook of wis- Cognitive developmental research 239). München: Beck. Bedeutung der Zeit: Zeiterleben dom: Psychological perspectives from lifespan perspectives: The und Zeiterfahrung aus Sicht der (pp. 110–135). Cambridge, UK: challenge of integration. In E. Individualpsychologie (pp. 95– Cambridge University Press. Bialystok & F. I. M. Craik (Eds.), 131). Göttingen: Vandenhoeck & Lifespan cognition: Mechanisms Ruprecht (Beiträge zur Individu- Kunzmann, U., & Grühn, D. of change (pp. 344–363). Oxford: alpsychologie 30). (2005). Age differences in emo- Oxford University Press. tional reactivity: The sample case Kreppner, K. (2005c). Family as- of sadness. Psychology and Ag- Li, S.-C., Brehmer, Y., Shing, sessment and methodological ing, 20, 47–59. Y.-L., Werkle-Bergner, M., & issues: Discussion. European Lindenberger, U. (2006). Neuro- Journal of Psychological Assess- Kunzmann, U., Kupperbusch, C. S., modulation of associative and or- ment, 21, 249–254. & Levenson, R. W. (2005). Behav- ganizational plasticity across the ioral inhibition and amplification life span: Empirical evidence and Kreppner, K. (2005d). Heinz during emotional arousal: A com- neurocomputational modeling. Werner and the Psychological In- parison of two age groups. Psy- Neuroscience and Biobehavioral stitute in Hamburg. In J. Valsiner chology and Aging, 20, 144–158. Reviews, 30, 775–790. (Ed.), Heinz Werner and developmental science (pp. 55–74). Kunzmann, U., & Stange, A. Li, S.-C., & Freund, A. M. (2005). New York: Kluwer Academic/ (2006). Wisdom as a classical Advances in lifespan psychology: Lindenberger, U., & Brehmer, Plenum Publishers. human strength: Psychological A focus on biocultural and per- Y., Kliegl, R., & Baltes, P. B. (in conceptualizations and empirical sonal influences. Research in press). Benefits of graphic design Kreppner, K. (2005e). Persönlich- inquiry. In A. D. Ong & M. H. M. Human Development, 2, 1–23. expertise in old age. In C. Lange- keitsentwicklung in sozialen van Dulmen (Eds.), Oxford hand- Küttner & A. A. Winter (Eds.), Beziehungen. In N. Birbaumer book of methods in positive psy- Li, S.-C., & Lindenberger, U. Drawing and non-verbal intelli- et al. (Series Eds.), J. B. Asendorpf chology (pp. 306–322). Oxford: (2006). Ageing deficit in neuro- gence. Cambridge: Cambridge (Vol. Ed.), Enzyklopädie der Psy- Oxford University Press. modulation of representational University Press.

Center for Lifespan Psychology | 177 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 178

Lindenberger, U., & Chicherio, C. age: Lifespan psychology as a Lövdén, M., & Lindenberger, U. Nilsson, L.-G., Adolfsson, R., (in press). Développement intel- conceptual foundation. In P. B. (2007). Intelligence. In J. E. Birren Bäckman, L., Cruts, M., Nyberg, lectuel au cours du cycle de vie: Baltes, P. A. Reuter-Lorenz, & F. (Ed.), Encyclopedia of gerontol- L., Small, B. J., & Van Broeckoven, Sources de variabilité et niveaux Rösler (Eds.), Lifespan develop- ogy: Age, aging and the aged C. (2006). The influence of APOE d'analyse. L'Année Pschologique. ment and the brain: The perspec- (2nd ed., Vol. 1, pp. 763–770). status on episodic and semantic tive of biocultural co-construc- Amsterdam: Elsevier. memory: Data from a population- Lindenberger, U., & Kray, J. tivism (pp. 350–375). Cambridge, based study. Neuropsychology, (2005). Kognitive Entwicklung. UK: Cambridge University Press. Lövdén, M., & Lindenberger, U. 20, 645–657. In N. Birbaumer et al. (Series (2005). Development of intellec- Eds.), S.-H. Filipp & U. M. Lindenberger, U., & von tual abilities in old age: From age Oberauer, K., Wilhelm, O., & Staudinger (Vol. Eds.), Enzyk- Oertzen, T. (2006). Variability in gradients to individuals. In O. Schmiedek, F. (2005). Experi- lopädie der Psychologie: Vol. C, cognitive aging: From taxonomy Wilhelm & R. W. Engle (Eds.), mental strategies in multivariate V, 6. Entwicklungspsychologie to theory. In E. Bialystok & F. I. M. Handbook of understanding and research. In A. Beauducel, B. Biehl, des mittleren und höheren Craik (Eds.), Lifespan cognition: measuring intelligence (pp. 203– M. Bosniak, W. Conrad, G. Erwachsenenalters (pp. 299– Mechanisms of change (pp. 297– 221). Thousand Oaks, CA: Sage. Schönberger, & D. Wagener (Eds.), 341). Göttingen: Hogrefe. 314). Oxford: Oxford University Multivariate research strategies: Press. Lövdén, M., Schellenbach, M., Festschrift in Honor of Werner Lindenberger, U., Li, S.-C., & Grossman-Hutter, B., Krüger, A., & W. Wittmann (pp. 119–149). Bäckman, L. (2006a). Delineat- Lindenberger, U., & Schäfer, S. Lindenberger, U. (2005). Envi- Aachen: Shaker. ing brain-behavior mappings (in press). Erwachsenenalter und ronmental topography and pos- across the lifespan: Substantive Alter. In R. Oerter & L. Montada tural control demands shape von Oertzen, T. (2006). Exact and methodological advances (Eds.), Entwicklungspsychologie. aging-associated decrements in computation of polynomial zeros in developmental neuroscience. Weinheim: Beltz. spatial navigation performance. expressible by square roots. Editorial. Neuroscience & Psychology and Aging, 20, Algorithmica, 46, 119–136. Biobehavioral Reviews, 30, Lindenberger, U., & Werkle- 683–694. 713–717. Bergner, M. (2005). [Review of Rapp, M. A., Krampe, R. T., & the article Reward-related fMRI Lövdén, M., & Wahlin, A. (2005). Baltes, P. B. (2006). Adaptive Lindenberger, U., Li, S.-C., & activation of dopaminergic mid- The sensory-cognition association task prioritization in aging: Selec- Bäckman, L. (Eds.). (2006b). brain is associated with enhanced in adulthood: Different magni- tive resource allocation to pos- Methodological and conceptual hippocampus-dependent long- tudes for processing speed, inhibi- tural control is preserved in advances in the study of brain- term memory formation]. Neuro- tion, episodic memory, and false Alzheimer’s disease. American behavior dynamics: A multivari- forum, 11 (2), 67–69. memory? Scandinavian Journal Journal of Geriatric Psychiatry, ate lifespan perspective [Special of Psychology, 46, 253–262. 14, 52–61. Issue]. Neuroscience & Biobehav- Lövdén, M., Bergman, L., ioral Reviews, 30. Amsterdam: Adolfsson, R., Lindenberger, U., MacDonald, S. W. S., Stigsdotter Ratcliff, R., Schmiedek, F., & Elsevier. & Nilsson, L.-G. (2005). Studying Neely, A., Derwinger, A., & McKoon, G. (in press). A diffu- individual aging in an interindi- Bäckman, L. (2006). Rate of ac- sion model explanation of the vidual context: Typical paths of quisition, adult age, and basic worst performance rule for re- age-related, dementia-related, cognitive abilities predict forget- action time and IQ. Intelligence. and mortality-related cognitive ting: New views on a classic prob- development in old age. Psychol- lem. Journal of Experimental Psy- Raz, N., Lindenberger, U., ogy and Aging, 20, 303–316. chology: General, 135, 368–390. Ghisletta, P., Rodrigue, K. M., Kennedy, K. M., & Acker, J. D. (in Lövdén, M., Ghisletta, P., & Marcus, B., Höft, S., & Riediger, press). Neuroanatomical correlates Lindenberger, U. (2005). Social M. (2006). Integrity tests and the of fluid intelligence in healthy participation attenuates decline five-factor model of personality: adults and persons with vascular in perceptual speed in old and A review and empirical test of risk factors. Cerebral Cortex. very old age. Psychology and two alternative positions. Inter- Aging, 20, 423–434. national Journal of Selection Raz, N., Lindenberger, U., and Assessment, 14, 113–130. Rodrigue, K. M., Kennedy, K. M., Lövdén, M., Herlitz, A., Head, D., Williamson, A. et al. Schellenbach, M., Grossman- McArdle, J. J., Small, B. J., (2005). Regional brain changes in Hutter, B., Krüger, A., & Bäckman, L., & Fratiglioni, L. aging healthy adults: General Lindenberger, U. (in press). (2005). Longitudinal models of trends, individual differences and Lindenberger, U., Li, S.-C., Quantitative and qualitative sex growth and survival applied to modifiers. Cerebral Cortex, 15, Lövdén, M., & Schmiedek, F. (in differences in spatial navigation. the early detection of Alzheimer's unpaginated. press). The Center for Lifespan Scandinavian Journal of Psy- disease. Journal of Geriatric Psychology at the Max Planck In- chology. Psychiatry and Neurology, 18, Raz, N., & Nagel, I. E. (in press). stitute for Human Development: 234–241. Der Einfluss des Hirnalterungs- Overview of conceptual agenda Lövdén, M., Li, S.-C., Shing, prozesses auf die Kognition: Eine and illustration of research activi- Y.-L., & Lindenberger, U. (in Nagel, I. E., Werkle-Bergner, Integration struktureller und ties. International Journal of press). Trial-to-trial variability M., Li, S.-C., & Lindenberger, funktioneller Forschungsergeb- Psychology. precedes and predicts decline in U. (2007). Perception. In J. E. nisse. In J. Brandtstädter & U. cognitive performance in old Birren (Ed.), Encyclopedia of Lindenberger (Eds.), Entwick- Lindenberger, U., & Lövdén, M. age: Longitudinal data from the gerontology: Age, aging and the lungspsychologie der Lebens- (2006). Co-constructing human Berlin Aging Study. Neuropsy- aged (2nd ed., Vol. 2, pp. 334– spanne: Ein Lehrbuch. Stuttgart: engineering technologies in old chologia. 342). Amsterdam: Elsevier. Kohlhammer.

178 | Center for Lifespan Psychology 08_LIP_126_179.qxd 19.07.2007 9:18 Uhr Seite 179

Rehnman, J., & Herlitz, A. Brandtstädter & U. Lindenberger of reaction time distributions and of emotional expressions in 4- (2007). Women remember more (Eds.), Entwicklungspsychologie their relations to working memory month-old infants. Social Cogni- faces than men do. Acta Psycho- der Lebensspanne: Ein Lehrbuch. and intelligence. Journal of Ex- tive and Affective Neuroscience, logica, 124, 344–355. Stuttgart: Kohlhammer. perimental Psychology: General. 1, 87–94.

Rehnman, J., & Herlitz, A. Schäfer, S., Huxhold, O., & Smith, J. (2006a). Berlin Aging Thilers, P. P., MacDonald, S. W. S., (2006). Higher face recognition Lindenberger, U. (2006). Healthy Study. In R. Schulz, L. S. Noelker, & Herlitz, A. (2006). The associa- ability in girls: Magnified by mind in healthy body? A review of K. Rockwood, & R. L. Sprott (Eds.), tion between endogenous free own-sex and own-ethnicity bias. sensorimotor-cognitive interde- Encyclopedia of aging: A com- testosterone and cognitive per- Memory, 14, 289–296. pendencies in old age. European prehensive resource in gerontol- formance: A population-based Review of Aging and Physical ogy and geriatrics (4th ed., Vol. 1, study in 35 to 90 year-old men Riediger, M. (2007). Interference Activity, 3, 45–54. pp. 110–112). New York: Springer. and women. Psychoneuro- and facilitation among personal endocrinology, 31, 565–576. goals: Age differences and associ- Scheibe, S., Freund, A. M., & Smith, J. (2006b). Developmen- ations with well-being and be- Baltes, P. B. (2007). Toward a de- tal psychology. In R. Schulz, L. S. Werkle-Bergner, M., Müller, havior. In B. R. Little, K. Salmela- velopmental psychology of Sehn- Noelker, K. Rockwood, & R. L. V., Li, S.-C., & Lindenberger, U. Aro, & S. D. Philipps (Eds.), Per- sucht (life-longings): The optimal Sprott (Eds.), Encyclopedia of ag- (2006). Cortical EEG correlates of sonal project pursuit: Goals, (utopian) life. Developmental ing: A comprehensive resource in successful memory encoding: Im- action, and human flourishing Psychology, 43, 778–795. gerontology and geriatrics (4th plications for lifespan compar- (pp. 119–143). Mahwah, NJ: ed., Vol. 1, pp. 306–308). isons. Neuroscience & Biobehav- Erlbaum. Scheibe, S., Kunzmann, U., & New York: Springer. ioral Reviews, 30, 839–854. Baltes, P. B. (in press). New terri- Riediger, M., & Ebner, N. C. (in tories of positive lifespan devel- Smith, J., & Baltes, P. B. (in Yang, L., Krampe, R. T., & Baltes, press). A broader perspective on opment: Wisdom and life long- press). Healthy aging from 70 to P. B. (2006). Basic forms of cog- three lifespan theories: Comment ings. In C. R. Snyder & S. J. Lopez 100+ in Germany: Lessons from nitive plasticity extended into the to Boerner and Jopp. Human (Eds.), Handbook of positive psy- the Berlin Aging Study. In M. oldest-old: Retest learning, age, Development. chology. New York: Oxford Uni- Robinson, W. Novelli, & C. E. and cognitive functioning. Psy- versity Press. Pearson (Eds.), Global health and chology and Aging, 21, 372–378. Riediger, M., & Freund, A. M. global aging. New York: Jossey- (2006). Focusing and restricting: Scheibe, S., Kunzmann, U., & Bass/Wiley. Zimmer, H. D., Mecklinger, A., & Two aspects of motivational se- Baltes, P. B. (2007). Wisdom, life Lindenberger, U. (Eds.). (2006a). lectivity in adulthood. Psychology longing, and optimal development. Smith, J., & Delius, J. A. M. Handbook of binding and mem- and Aging, 21, 173–185. In J. A. Blackburn & C. N. Dulmus (2006). BASE: Die Berliner Alters- ory: Perspectives from cognitive (Eds.), Handbook of gerontology: studie. In W. D. Oswald, U. Lehr, C. neuroscience. Oxford: Oxford Riediger, M., Freund, A. M., & Evidence-based approaches to Sieber, & J. Kornhuber (Eds.), University Press. Baltes, P. B. (2005). Managing theory, practice, and policy Gerontologie: Medizinische, psy- life through personal goals: Inter- (pp. 117–142). New York: Wiley. chologische und sozialwissen- goal facilitation and intensity of schaftliche Grundbegriffe (3rd, goal pursuit in younger and older Schmiedek, F. (2005). Item re- complete rev. ed., pp. 114–119). adulthood. Journal of Gerontol- sponse theory and the measure- Stuttgart: Kohlhammer. ogy: Psychological Sciences, 60B, ment of cognitive processes. In P84–P91. O. Wilhelm & R. W. Engle (Eds.), Smith, J., & Delius, J. A. M. Handbook of understanding and (2005). Psychologische Funktions- Riediger, M., Li, S.-C., & measuring intelligence (pp. 265– fähigkeit im Alter: Potenziale und Lindenberger, U. (2006). Selec- 277). Thousand Oaks, CA: Sage. Grenzen. Quintessenz, 56 (2), tion, optimization, and compen- 159–169. sation as developmental mecha- Schmiedek, F., & Li, S.-C. nisms of adaptive resource alloca- (2006). Die Lernkurve. In J. Bengel Smith, J., & Jopp, D. (2005). tion: Review and preview. In J. E. et al. (Series Eds.), J. Funke & P. A. Geschlechterdifferenzen im Birren & K. W. Schaie (Eds.), The Frensch (Vol. Eds.), Handbuch der Alter(n) [Gender differences in old handbooks of aging: Vol. 2. Psychologie: Vol. 5. Handbuch age and aging]. In N. Birbaumer Handbook of the psychology of der allgemeinen Psychologie: et al. (Series Eds.), S.-H. Filipp & aging (6th ed., pp. 289–313). Kognition (pp. 213–219). Göttin- U. M. Staudinger (Vol. Eds.), Enzy- Amsterdam: Elsevier. gen: Hogrefe. klopädie der Psychologie: Vol. C, V, 6. Entwicklungspsychologie Rönnlund, M., Nyberg, L., Bäck- Schmiedek, F., & Lindenberger, des mittleren und höheren man, L., & Nilsson, L.-G. (2005). U. (in press). Methodologische Erwachsenenalters (pp. 501– Zimmer, H. D., Mecklinger, A., & Stability, growth, and decline in Grundlagen. In J. Brandtstädter & 524). Göttingen: Hogrefe. Lindenberger, U. (2006b). Levels adult life span development of U. Lindenberger (Eds.), Entwick- of binding: Types, mechanisms, declarative memory: Cross- lungspsychologie der Lebens- Staudinger, U. M., & Linden- and functions of binding in re- sectional and longitudinal data spanne: Ein Lehrbuch. Stuttgart: berger, U. (2007). Nachruf auf membering. In H. D. Zimmer, A. from a population-based study. Kohlhammer. Paul B. Baltes. Psychologische Mecklinger, & U. Lindenberger Psychology and Aging, 20, 3–18. Rundschau, 58, 149–151. (Eds.), Handbook of binding and Schmiedek, F., Oberauer, K., memory: Perspectives from cog- Schäfer, S., & Bäckman, L. (in Wilhelm, O., Süß, H.-M., & Striano, T., Kopp, F., Grossmann, nitive neuroscience (pp. 3–22). press). Normales und patholo- Wittmann, W. W. (in press). Indi- T., & Reid, V. M. (2006). Eye con- Oxford: Oxford University Press. gisches kognitives Altern. In J. vidual differences in components tact influences neural processing

Center for Lifespan Psychology | 179 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 180 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 181

Independent Junior Research Group

Neurocognition of Decision Making 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 182

Contents

Research Agenda: Multimodal Approach to the Neurocognition of Decision Making . . . . . 183

Neurocognition of Perceptual Decision Making ...... 184

Reward-Based Decision Making ...... 187

Decision Making in Social Contexts ...... 191

Publications 2005–2006 ...... 198

Scientific Staff (October 2005-September 2010)

Markus Bauer, Guido Biele, Isabel Dziobek, Hauke R. Heekeren

Predoctoral Research Fellows Markus Bahnemann, Nikos Green, Lea Krugel, Katja Mériau (NaföG), Peter Mohr (LIFE), Kristin Prehn (NaföG), Sandra Preißler (BNIC), Christina Scheibe (NaföG), Hermine Wenzlaff

182 | Independent Research Group 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 183

Research Agenda: Multimodal Approach to the Neurocognition of Decision Making

Decision making can be defined as the process of choosing a preferred option or course of action from among a set of alternatives. There is a long history of decision-making research in psychology and economics that has resulted in the development of formal models of behavior, which are inspired by behavioral data or the computational demands of a task. An example for the former are sequential sampling models of decison making, an example for the latter are reinforcement learning models for repeated choice tasks. However, cognitive functions, such as decision making, cannot be completely understood on the basis of mathematical models and behavioral data alone; we have to investigate how mental (cognitive) and neuronal processes map onto each other (cf. Schall, 2004). Therefore, a central goal of the independent junior research group “Neurocognition of Decision Making” is to explicitly link brain function and behavior using formal models of decision-making behavior.

In pursuit of this goal, we investigate deci- tional aspects and the interaction between sion making in different domains. First, at the the two. basis of a number of different decisions we We believe that the investigation of the neu- are facing in everyday life stands perceptual rocognition of decision making requires a decision making: the process of translating multimodal methodological approach that sensory input into some kind of motor output integrates information from an array of (cf. Figure 1). Many of our decisions are in- methods ranging from cognitive modeling fluenced by the potential outcomes associ- based on behavioral data to simultaneous ated with different options, hence, reward- functional Magnetic Resonance Imaging based decision making is another important (fMRI) and EEG experiments (cf. Figure 1). topic for our group. Decision making in social On the following pages, we briefly describe contexts relies not only on perceptual and research on the three topics in some more reward-related processes but also includes detail. Each section begins with a brief intro- more complex cognitive processes and emo- duction, which is followed by short descrip- tions of individual projects.

Figure 1. A multimodal approach to the neurocognition of decision Neurocognition of decision making making.

Sensory Decision processing Motor input output Process model Prior Reward Emotion Cognition probability

Topics Perceptual Reward-based Social decision decision making decision making making

Behavior Cognitive modeling Multimodal Psychophysiology approach EEG MEG Functional MRI Structural MRI

Independent Research Group | 183 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 184

Neurocognition of Perceptual Decision perceptual decisions are made by integrating Making the sensory evidence represented by lower Perceptual decision making entails processing level sensory neurons (Gold & Shadlen, 2002; of sensory signals, formation of a decision, Heekeren et al., 2004). For example, when and the planning and execution of a motor monkeys must decide whether a noisy field of response. For example, in a motion-direction dots is moving upward or downward, a deci- discrimination task, motion signals need to sion can be formed by computing the differ- be interpreted and translated into a saccadic ence in responses between lower level neu- eye movement. In a face-house discrimina- rons sensitive to upward motion and those tion task, degraded images of faces and sensitive to downward motion. In this model, houses have to be interpreted and translated higher level brain regions computing decision into a button press with the right or the left variables thus have to accumulate evidence hand. over time. The dynamics that are involved in Over the last 40 years, decision-making re- the formation of these higher level represen- search in mathematical psychology has re- tations are of core interest to systems and sulted in detailed mathematical models of cognitive neuroscience. In ongoing projects, the assumed underlying cognitive processes. we build on our previous work and investi- Sequential sampling models are particularly gate how sensory systems infer the identity successful in explaining response time data of real-world objects from their noisy neural and accuracy in two-choice reaction time representations, and how the outcome of tasks, such as the ones described above. A such a process is signaled to the motor sys- prominent version of sequential sampling tem. Furthermore, we want to elucidate how models are diffusion models, which assume the brain uses contextual information (in that decisions are formed by continuously particular prior probabilities about possible accumulating sensory information until one alternatives) that is integrated with stimulus of the two response criteria is reached (e. g., Smith & Ratcliff, 2004; cf. Figure 2). Given the well-described nature of percep- Response “right” a tual decision-making tasks, such as face- house discrimination and the relatively well- μ=kc known functional neuroanatomy of the sub- systems involved, perceptual decision making has recently been used as paradigm to study z Time (ms) simple decision processes in the brain of human and nonhuman primates. Single-unit recordings in monkeys, and more recently neuroimaging studies in humans, 0 have begun to investigate the neural mecha- Response “left” nisms involved in these processes (e. g., Heekeren, Marrett, Bandettini, & Ungerleider, Figure 2. Stochastic diffusion models assume that a decision is the result of continuously accumu- 2004, 2006; Mazurek et al., 2003; Romo & lating noisy stimulus information from a starting Salinas, 2003) and to model not only psy- point z until one of two response criteria, a and chophysical data but also neurophysiological 0, is reached. Once the boundary is reached, the decision process is concluded and a response is data as a diffusion-to-barrier process (e. g., elicited. Moment-by-moment fluctuations in the Gold & Shadlen, 2002; Philiastides et al., sample path reflect noise in the decision process. 2006). Thus, perceptual decision-making The drift rate (μ) is related to the efficacy of information processing. μ is dependent on the tasks provide an ideal paradigm to pursue our ncrease in the decision variable (accumulation goal to explicitly link brain function and be- rate, k) quantifying how much evidence is accu- havior using formal models. mulated per time interval as well as on the strength of the sensory signal (c) (e. g., Smith & Studies in monkeys and humans performing Ratcliff, 2004). sensory discrimination tasks suggest that

184 | Independent Research Group 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 185

inherent information, and thereby conveys sion, only a brief burst of oscillatory activity additional information to improve decision occurs. These sustained oscillatory responses making. are accompanied by rather long reaction times (~500 ms) suggesting that time-endur- Perceptual Decision Making in the ing internal computational processes are Somatosensory Domain necessary to distinguish between the two The project “Tactile Decision Making” investi- patterns to make a decision. A central ques- gates how sensory information is processed tion inherent to the project presented here to make categorical judgments and aims at will be whether these sustained oscillations revealing the neural processes behind the are instrumental to the decision process and accumulation of sensory evidence. reflect the accumulation of evidence. To date, it is unclear whether the accumula- In a recent experiment, three pattern pairs tion of evidence by a diffusion process as were presented varying in difficulty level op- found in the visual domain (see above) gen- erationalized by the spatial separability of eralizes to other sensory modalities and stim- the patterns of a pair. Subjects had to dis- ulus materials with only brief exposure. Fur- criminate between two tactile stimuli of a thermore, the precise neuronal mechanisms pair. Response times as well as error rates in- behind such accumulation of evidence re- creased with increasing difficulty, indicating main unclear. In a first study, we therefore that temporal integration (accumulation) investigated whether there is evidence for of evidence seems to occur (see Figure 3). temporal integration in the somatosensory In ongoing studies, we are investigating system using a tactile decision task. Previous whether, similar to neurophysiological studies research from Bauer, Oostenveld, Peeters, and in monkeys, human EEG and MEG data pro- Fries (2006) has shown that in a two-alter- vide neurophysiological evidence for a “diffu- native decision task, very brief exposure sion to boundary process” determining the (35 ms) to tactile pattern stimuli leads to elicitation of a response. sustained high-frequency oscillations in pri- Future directions of this project will include mary somatosensory cortex when subjects simultaneous EEG and fMRI experiments to have to report the identity of these patterns. study the interactions between higher level On the contrary, when stimuli are presented, cortical structures, such as the DorsoLateral but subjects are not required to make a deci- PreFrontal Cortex (DLPFC; which is known to

Figure 3. Three different Braille pairs (varying in Easy Medium Hard 0.3 spatial similarity) were presented to subjects' 0.2 right index fingers (left). Discrimination of these pairs resulted not only 0.1 in substantially different Error rate (in %) performance levels, but 12 12 12 Easy Medium Hard also longer reaction times (right).

1,100

900 RT (ms)

700 Easy Medium Hard

Correct trials Incorrect trials

Independent Research Group | 185 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 186

Key Reference be involved in somatosensory decision mak- higher level response selection and abstract Scheibe, C., Ullsperger, ing, cf. Preuschhof, Heekeren, Taskin, motor programming at a more central level M., Sommer, W., & Schubert, & Villringer, 2006, but difficult to while the Lateralized Readiness Potential Heekeren, H. R. (2007). map with EEG/MEG) and sensory cortices (LRP) is an index for lower level motor Spatiotemporal dynamics munderlying the integra- (which are more easily localizable with preparatory processes. To investigate the in- tion of prior probability EEG/MEG, cf. e. g., Bauer et al., 2006). Fur- fluence of prior probability, we used a num- information in decision thermore, it will be of interest to show how ber comparison task as a modified precue making. Conference of Cognitive Neuroscience, information from different sensory modalities paradigm. Our data revealed that the CNV New York. Abstract. is combined to make decisions about external during the foreperiod is parametrically modu- objects. Parts of this project are realized in lated by prior probability, indicating an inte- collaboration with Driver (Institute of Cogni- gration of prior probability at an early stage tive Neuroscience/Wellcome Trust Centre for of response selection and premotor prepara- Neuroimaging, UCL, London) as well as von tion (cf. Figure 4) (Scheibe, Schubert, Oertzen (Center for Lifespan Psychology at Sommer, & Heekeren, 2006). In contrast, the MPI for Human Development). LRP was modulated in a binary fashion, that is, it only distinguished between certain The Influence of Prior Probability on Decision (100 % prior probability) and uncertain infor- Making mation (<100 % prior probability). The prior probability of the occurrence of an Based on these results, we recently con- event is a critical variable in decision-making ducted an fMRI experiment with simultane- processes. The response time in two-alterna- ous EEG recordings (in cooperation with tive forced choice tasks is speeded by prior Ullsperger and von Cramon, MPI for Human probability. Single cell recordings in monkeys Cognitive and Brain Science, Leipzig) to ex- revealed a modulation of the firing rate of amine the neural correlates of the integra- motor-related neurons by prior probability. tion of prior probability in the decision Furthermore, modeling studies suggest that process. Despite the EEG-adverse environ- the motor system is an integral component of ment in the scanner, we replicated the decision-making processes. However, it is un- increasing CNV amplitude with increasing clear at which stage of the decision-making prior probability. The fMRI data revealed that process in the human brain prior probability areas in the prefrontal cortex, particularly the is integrated. In a recent experiment, we used dorsolateral and frontomedial regions, seem Event-Related Potential (ERP) components as to play a crucial role in preparatory processes proxies for different processing stages of the based on prior probability (Scheibe et al., decision-making process. In this context, the 2007). In a further EEG-informed analysis, we Contingent Negative Variation (CNV) reflects will relate the CNV amplitude and Blood Oxy-

Figure 4. EEG component at the electrode position –8 –5,000 { Cz for the different ** –4,500 conditions. The CNV –6 between S1 and S2 –4 –4,000 (foreperiod) increases –3,500 { ** parametrically with in- –2 –3,000 creasing prior probability. 0

1,000 2,000 3,000 time (ms) Area ( μ V*ms)+SE –2,500 2 –2,000 PP 0.5 PP 0.75 PP 1.0 Mean voltage ( μ V) 4 valid invalid 6 8 Prior probability S1 S2 PP 0.5 PP 0.75/invalid PP 0.75/valid PP 1.0

186 | Independent Research Group 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 187

genation Level Dependent (BOLD) correlates mation that needs to be represented and ma- at a single-trial level. Thereby, we want to nipulated to make good decisions. Knowledge identify brain regions that are associated about how the brain represents information, with the development of the CNV and the in- and about the type of connections between tegration of prior probability into the deci- different brain regions, constrains theories sion process. about which information is used, and how it is manipulated. Reward-Based Decision Making Guided by this general approach, a typical Many of our decisions are influenced by the project starts with one or several behavioral potential outcomes associated with different theories of reward-based decision making, options. For instance, consumers test differ- which are compared in their ability to predict ent brands of a particular product to identify behavioral data. In a second step, the theo- their favored brand, or investors first learn ries are tested on the neuronal level. Testing about the past performance of investments theories on a neuronal level amounts, for in- and then decide about the allocation of their stance, to investigating if a representation of funds. The project “reward-based decision information that is essential to that theory making” examines how people use informa- can be found on the neuronal level, or if a tion to achieve rewarding outcomes. representation of a behaviorally unobservable To examine reward-based decision making, variable predicted by a theory can be identi- we abstract basic features from real-life de- fied. Due to our multilevel approach to cisions, such as the type of information and reward-based decision making, the projects feedback available, and implement them in are usually interdisciplinary, and involve simpler tasks, which are amenable for the ap- cooperation with other research institutions. plication in an fMRI environment and to pre- The remainder will introduce some ongoing cise modeling. Conducting fMRI experiments projects. allows us to test models and theories with variables, such as the prediction error in tem- Neural Representation of Decision Thresholds poral difference learning models that are One question that is particularly interesting usually not accessible with exclusively be- for the fields of both perceptual decision havioral experiments. making and reward-based decision making is Reward-based decision making has been how decision makers adjust their decision tackled by different disciplines, therefore, our criterion to obtain optimal rewards. One kind theoretical approach combines elements of model addressing this question are se- from these: Economics and Machine Learning quential sampling models because they suc- describe procedures which aim to maximize cessfully explain response times and percent- the decision maker’s outcome (Markowitz, age of correct decisions in tasks as different 1952; Sutton & Barto, 1998), psychological as perceptual decision making and short- theories describe how people learn from term memory tasks (Ratcliff & Smith, 2004). feedback (Erev & Barron, 2005), and neuro- The common idea of sequential sampling scientific research describes which kind of models is that the decision maker accumu- information is represented, and how it is ma- lates information in favor of the different de- nipulated to reach a decision (e. g., Frank & cision options, and that the decision is made Claus, 2006). While it is a challenging en- in favor of the option for which the accumu- deavor to examine reward-based decision lated evidence first exceeds a decision making across these different levels, we be- threshold (see above). Recent research at- lieve that a solid understanding of reward- tempted to test sequential sampling models based decision making needs to describe and by identifying neurons that accumulate evi- understand decision making on different the- dence over time, and by correlating EEG pa- oretical levels (Marr, 1982). Understanding rameters with the accumulation gradient in how goals can be achieved in an optimal way sequential sampling models. From a reward- allows deriving hypotheses about the infor- based learning perspective, the decision

Independent Research Group | 187 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 188

Figure 5. Decision thresholds: Speed-accuracy trade-off can be 1.8 manipulated by the Re- sponse Stimulus Interval 1.6 (RSI) and by punishment for incorrect decisions. The figure depicts opti- 1.4 mal thresholds (Y-axis) 2.0 for a sequential sampling 1.2 decision model. The payoff for a correct decision 1.5 is fixed to 1, the param- 1.0 eter for the drift rate is 1.0 .5, the residual decision 3 0.8 time is set to empirically realistic 0.4. 0.5 2 0.6

1 0.0 0.4 0.0 –0.5 –1.0 –1.5 –2.0 0 RSIs –2.5 –3.0 Punishments

threshold is an especially interesting parame- and Cognition (ABC) at the MPI for Human ter because it is under the influence of the Development. decision maker. Therefore, in this project, we investigate the neural correlates of the deci- Is Ambiguity Aversion Caused by Pessimism or sion threshold. Specifically, we plan to con- Dislike? duct experiments on perceptual (visual) deci- An observation of considerable interest to sion making, in which we systematically economists and psychologists alike is that modify payoffs, so that participants have a people prefer risky situations over ambiguous strong incentive to adjust their decision situations with the same expected outcome. threshold. For instance, increasing the pun- For example, most subjects would prefer a ishment for a wrong decision for one alterna- gamble with a 50 % chance to win 10 Euros tive in a two-alternative forced choice task and to win nothing otherwise, over a gamble should increase the decision threshold for with unknown probabilities to win either 10 that alternative (Figure 5). In a first behav- Euros or nothing. The aim of this experiment ioral study, we investigate how people re- is to test the hypothesis that ambiguity aver- spond to changing payoff structures. The first sion can be explained by a pessimistic atti- goal of the behavioral study is to identify tude toward ambiguous situations, that is, payoff matrices leading to different thresh- that people assign the worse outcome a olds. The second goal is to evaluate which higher likelihood than justified by the avail- sequential sampling model explains the data able information (Gilboa & Schmeidler, 1989). best. In an fMRI study, we will then record To test this hypothesis, we use a new variant brain activation during perceptual decision of the monetary incentive delay task, which making under different payoff conditions has previously been used to elicit representa- identified in the behavioral study. Here, the tions of event probabilities (Knutson et al., objective is to identify how the decision 2005). During the task, participants first re- threshold parameter of the best sequential ceive information about the difficulty of a sampling model is represented in connectivity shortly following face-house discrimination between different brain regions. This project task (see Figure 6). Crucially, the participants’ is realized in collaboration with Galesic and payment depends on the performance in this Olsson from the Center for Adaptive Behavior task, where the correct response needs to be

188 | Independent Research Group 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 189

Gamble presentation (4,000 ms)

20 € 10 € Anticipation phase (2,000–3,000 ms)

Reaction time task (max. 500 ms)

Outcome feedback (max. 3,000 ms)

Feedback

Figure 6. Modified delayed incentive task. Participants are first presented with a gamble, which also predicts the probability of a house- target or a face-target in the reaction time task. If participants are ambiguity averse due to pessimism, this increases their expected probability to react fast enough after presentation of an ambiguous lottery, compared to an 50 %-50 % lottery. Neural corralates of this expected probability, measured in the anticipation phase, have been identified in earlier studies.

made in limited time. After a brief delay cause pessimistic participants will expect the (2,000-3,000 ms), participants perform the low-outcome object with a high certainty. reaction time task, and finally receive feed- This likelihood of success is a direct function back about their success. Previous studies of the probability with which a participant showed that activation in the medial pre- expects a specific face, importantly, the frontal cortex correlates with the success paticipant’s expectation is difficult to assess probability in the reaction time task. We in a purely behavioral experiment. modified the task so that participants first This project is a collaboration with Rieskamp receive probabilistic or ambiguous informa- from the ABC group at MPI for Human Devel- tion about which of two different stimuli opment, Gerhardt from the Humboldt Univer- (i. e., face or house) they will see in the face- sity Berlin, and Mechtenberg from the Tech- house discrimination task. The two pictures nical University Berlin. represent the two possible outcomes of a gamble (high and low), which is only paid out Integration of Possible Gains and Losses in a to the participants if they make the correct Go-No-Go Task response fast enough. The ambiguity aversion How do people integrate possible gains and from pessimism hypothesis predicts on the losses when they decide for or against a par- behavioral level that reaction times for the ticular action? So far, this seemingly simple low-outcome picture are slower in the 50 %- but also important question has stimulated 50 % condition compared to the ambiguous little research in the field of neuroscience. condition. On the neuronal level, the medial While extensive research has examined the prefrontal cortex should signal a higher like- representation of experienced rewards and lihood of success in the ambiguous decision punishments, or anticipated reward, less is compared to the 50 %-50 % condition be- known about how potential gains and losses

Independent Research Group | 189 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 190

are represented and integrated before a decision is reached. Simple weighted additive 15 models, which are often used to describe multiattribute choices, predict that benefits and costs of a particular action are weighted 10 to generate a final evaluation. Translated into a hypothesis for an fMRI experiment, the weighted additive model suggests that peo- 5 ple represent gains and losses in two differ-

ent neural populations, whereas a third pop- Observed estimates 0 ulation might integrate this information. To gain insight in the neuronal processes of gain-loss integration, we designed an experi- –5 ment in which participants first learn cues –5 0 5 10 15 that signal monetary gains and losses of dif- Predicted estimates ferent amounts. After the learning phase, a Figure 7. Modeling estimates of expected return. decision condition follows in which people We selected for each participant the model that are confronted simultaneously with two cues best predicts his/her estimates in the estimation representing possible losses and gains, and trials. Most participants were best described by a model calculating the mean of the presented then have to decide if they want to receive returns while putting higher weights on returns the outcome of this gamble or not. The main below 5 %. The figure shows the regression lines goal of the study is to investigate which re- for estimated and modeled estimations of the expected return, indicating a good agreement of gions represent the gains and losses signaled both (n = 16). It therefore seems justified to use by the cues. Whereas the OrbitoFrontal modeled estimates as predictors of expected Cortex (OFC) is a natural candidate for this returns in the decision trials. (Kringelbach, 2005), recent results about the function of the OFC are still so diverse that this hypothesis deserves empirical examina- 1952). The aims of this project are (a) to tion. A second goal consists in identifying the model how people evaluate the expected neuronal processes that generate the final return and the risk of an investment, (b) to evaluation of the gamble, for example, by identify the neural representations of these integrating information from brain regions attributes, and, crucially, (c) to test if the representing reward and punishment, respec- representations are also present when peo- tively. This project is realized in collaboration ple have to choose between two invest- with Basten and Fiebach (University of ments. In the experiment used to investigate Heidelberg). these questions, participants first see a sequence of ten past returns of a risky invest- Neural Correlates of Risk-Value Models for ment and then have to either predict the ex- Financial Investment Decisions pected return, judge the subjective risk, or When people can choose between different choose between the presented investment choice options, the option with the highest and an investment with a known fixed return expected value is often also the riskiest op- of 5 %. On the behavioral level, we test if tion. One domain in which this is particularly risk-value models predict choices better than true is financial decision making, where in- alternative models, like simple decision vestments with higher expected returns are heuristics (Brandstätter, Gigerenzer, & usually also more volatile. Economic theory Hertwig, 2006) or sequential sampling mod- suggests that people should trade off ex- els (Busemeyer & Townsend, 1993). On the pected return and risk in such cases, that is, neuronal level, we will use fMRI to test if they should subtract some measure of risk representations for both expected return and from the expected return to generate an in- risk (see Figure 7) can be found when people tegrated evaluation of an option (Markowitz, choose between two investments, or if dif-

190 | Independent Research Group 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 191

ferent kinds of information are represented. the feedback phases. First results indicate Finally, we aim to validate the behavioral that outcomes from recommended options tests of the models by investigating if the might indeed be evaluated differently than integrated evaluation of the risky option, as outcomes from alternative options (Biele, predicted by the best model, is also repre- Krugel, Rieskamp, & Heekeren, 2007). sented in the brain. Decision Making in Social Contexts The Influence of Advice on Reinforcement Decision making processes during social in- Learning teractions are highly complex, unstructured, The final project in this section makes the and often intuitive. For example, when we link between reward-based decision making have to make a decision about the trustwor- and decision in social contexts. Starting point thiness of a stranger, a variety of informa- for the project is the observation that people tion, such as facial expression, tone of voice, often do not learn based on experienced re- and the pragmatics of language, have to be wards alone but also receive advice from interpreted simultaneously. Although part of other people. For example, when moving for those interpretations are accomplished a new job to a new city, one will probably get through cognitive (i. e., conscious and analyt- some advice about the best way to get to the ical) processes, given the velocity with which office but also experiment with other alter- social decisions have to be made, the faster natives. To examine the influence of advice and mostly unconscious emotional processes on reinforcement learning, a first behavioral (e. g., intuition) are likely to be involved as study by Biele, Rieskamp, and Gonzalez com- well (Adolphs, 2006). Thus, any social deci- pared different computational models de- sion will be the result of social cognitive as scribing the integration of advice and own well as socioemotional processes. The com- experience. The main result was that one mon goal of our subprojects within the topic model, which assumed that people evaluate of decisions in social contexts is to elucidate payoffs from the recommended option more the unique as well as combined contributions positively than the same payoff from alterna- that these cognitive and emotional processes tive options, described people’s behavior best. have on social decision making. Given the The current fMRI project examines in more complexity of social interactions, identifica- detail how the evaluation of payoffs differs. tion of brain correlates for social decisions One hypothesis is that people are not only poses a great challenge. By putting emphasis rewarded by the monetary payoffs but also on the development of psychological para- intrinsically for following advice. This hy- digms that focus on subcomponents of social pothesis can be derived from early theoretical decisions, we are trying to reduce this com- work on the development of imitative capaci- plexity, and to make it more controllable for ties (Miller & Dollard, 1941) and from more research settings while aiming for ecological recent work highlighting the rewarding ca- validity at the same time. Using structural pacity of cooperative behavior (Decety et al., and functional MRI as well as psychophysio- 2004). The fMRI prediction derived from the logical measures, such as skin conductance intrinsic reward hypothesis is that reward and pupillometry, our group is trying to eluci- from the recommended options leads to date how and where in the brain of healthy greater activation in the OFC than reward individuals social decisions are processed. from an alternative option. To test this and Moreover, to complement our understanding alternative hypotheses, we have collected of the “social brain,“ we are studying neu- fMRI and behavioral data from participants ropsychiatric patients with socioaffective who first received a recommendation about impairments, such as Borderline personality which deck to choose from in a multiple- disorder and autism. We expect that, in turn, armed bandit task. We employed an event our studies will inform future research and related design, which allows to separately intervention efforts in these and other condi- examine the decision, the anticipation, and tions with deficient decision making.

Independent Research Group | 191 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 192

Sociotopy of the Posterior Superior Temporal posterior superior temporal sulcus (pSTS). The Sulcus goal of the present fMRI study was to eluci- Social tasks, such as judging another person’s date the specific function of the pSTS region movements (biological motion, BM), mental within the social cognitive network. Based on states (theory-of-mind, ToM), and norm-con- a meta-analysis of previous imaging studies, gruency of behavior (normative judgment, we hypothesized that subregions within the NJ), lead to the recruitment of a network of pSTS would be differentially activated by the brain areas that consistently includes the three different tasks (BM, ToM, NJ). For our

Figure 8. Sample stimulus for the ToM condition Block instruction with timeline of presen- 5 sec. tation. The protagonist was marked by the white Which is the protagonist’s 1st frame shirt. The same two mental state? 1.5 sec. 2nd frame frames were used for the 3.75 sec. BM and NJ conditions, only the block instruction and the response t choice pair were adapted accordingly. BM: In which direction is the protagonist moving? (left or right). NJ: Is the protagonist committing a norm violation? (yes or no). Furious Sad

Figure 9. Sagittal view of the pSTS in the right and R x = 58 x = 58 L fMRI study, we developed a new set of ani- left hemisphere for BM, mated stimuli, which depicts two persons in ToM, and NJ, p < 0.001, social interactions (Figure 8). uncorrected. The area of interest is highlighted in Each stimulus includes the display of a per- green in the uppermost son’s movements and mental states, and the Area of interest panel. depicted behavior could clearly be judged concerning a possible violation of social norms. Importantly, the material was kept constant for all three conditions, and only task instructions were manipulated. The fMRI results reproduced task-specific activation BM patterns for each task, as reported by previous studies (Z = 3.1, uncorrected) (e. g., Allison et al., 2000; Gallagher & Frith, 2003; Goodenough & Prehn, 2004; Heekeren, Wartenburger, Schmidt, Schwintowski, & Villringer, 2003; Heekeren et al., 2005). With ToM respect to the pSTS, the activity evoked by the BM task was strongly right-lateralized, while that evoked by the ToM task was left- lateralized. The NJ task led to bilateral activation. There was a significant overlap of activity of the ToM and NJ tasks (Figure 9). The distinctive pattern of activation during NJ the BM task in relation to the ToM and NJ

192 | Independent Research Group 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 193

tasks suggests the involvement of different troscopy (NIRS) and its use in a study with cognitive processes. In contrast, the consider- subjects with Asperger Syndrom. able overlap between activations evoked by the ToM and NJ tasks indicates that the two Interaction of Emotional and Cognitive functions engage a common cognitive Processing in Decision Making process, such as an automated inference of To address the question, how emotional and thoughts/feelings or intentions, even if not conceptual processes interact in decision explicitly demanded by the task. Currently, making, we developed an analogical reason- we are preparing the adaptation of the ing task. In this task, word pairs were pre- paradigm for use with Near InfraRed Spec- sented simultaneously, and participants had

Figure 10. Neural corre- (a) Covariation of C-scores with BOLD responses in right DLPFC late of individual differences in moral judgment RLAPRLcompetence.

z = +18 x = +48 y = –18

(b) Socio-normative judgments 400 150

300 100 200

100 50

0 **

–100 0 BOLD in right DLPFC/a.u. BOLD in right DLPFC/a.u. Δ Δ –200 –50 0 10 2030 40 50 60 70 Low High C-score C-score

100 50 0

–100 0 BOLD in right DLPFC/a.u. BOLD in right DLPFC/a.u. Δ Δ –200 –50 01020 30 405060 70 Low High C-score C-score

Independent Research Group | 193 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 194

to decide whether they corresponded in their moral judgment competence (assessed with emotional as well as in their conceptual rela- the Moral Judgment Test; Lind, 2005) was re- tions. Response times, error rates, and pupil- flected in BOLD activity in right DLPFC during lary responses were recorded. Response time socio-normative judgments relative to gram- data showed that emotional processing is matical judgments. Participants with lower faster than conceptual processing. Pupillary moral judgment competence recruited the responses as a proxy for the processing load, right DLPFC more than those with higher however, clearly indicated that semantic pro- competence in this domain (Figure 10), indi- cessing is consuming as many cognitive re- cating higher processing demands (Prehn sources as emotional processing. This result et al., 2006). suggests that both cognitive and emotional aspects are processed in parallel (Prehn et al., Neural Correlates of Individual Differences in 2005). In a next step, we will identify the Cognitive Processing of Emotions network of brain regions that mediates emo- People differ in the extent to which they ex- tional and conceptual processing using fMRI perience emotions. These individual differ- and concurrent psychophysiological monitor- ences affect even simple perceptual deci- ing. In another project, we continue our pre- sions, and influence the psychophysiological vious work on the neural correlates of moral and neural correlates of emotion processing. judgment processes (Goodenough & Prehn, In a first fMRI study, we investigated how in- 2004; Heekeren et al., 2003, 2005). Moral dividual differences in cognitive processing of judgment as a higher cognitive function en- emotions influence perceptual decisions on tails a number of processes implemented by a the gender or the emotional expression of network of brain regions that are involved in faces (Figure 11). We found that those indi- cognitive as well as in emotional processing. vidual differences were associated with in- People differ with respect to their compe- creased activity of the left dorsal anterior tence to make moral judgments. We used cingulate cortex and altered connectivity of fMRI and skin conductance recording to in- this region with prefrontal areas (cognitive vestigate how individual differences in this processing) and limbic regions (emotional domain modulate neural correlates of norma- processing). We argue that differences in tive judgments. We found that individual functional integration of this neural network

dACC 60 R L 55 b = 0.643 50 G p = 0.001 45

z = 46 40 bE = 0.183 TAS score TAS p = 0.403 35

25 x = 9 –150 –100 –50 0 50 100 150 Δ BOLD in bilateral dACC/a.u.

Figure 11. Covariation of BOLD responses with TAS score. Left panel: TAS score covaried significantly with BOLD responses in the bilateral dACC during gender decisions. The figure shows activations from higher level analysis thresholded at Z > 3.1, corresponding to p = 0.001, uncorrected, that were used as a mask for subsequent PPI analyses. Right panel: Covariation of TAS score and BOLD responses (arbitrary units, a.u.) during emotion decisions (red dots) and gender decisions (grey dots) (bE and bG represent regression slopes for emotion decisions and gender decisions, respectively; p = p value).

194 | Independent Research Group 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 195

may give rise to individual differences in the group of well-matched Normal Controls (NC) ability to cognitively process emotions. More- (n = 29) in tests of social cognition, memory, over, we present neurophysiological evidence attention, executive functioning, and cortical for a theoretical model which posits that dis- thickness measures. Neuropsychological re- communication of the neocortex with limbic sults showed that differences between the areas impairs cognitive processing of emo- diagnostic groups were restricted to social tions (Mériau, Wartenburger et al., 2006). In perception and cognition (facial emotion another study, we combined psychophysio- recognition and theory of mind [as assessed logical and fMRI measurements to investi- using the Movie for the Assessment of Social gate how individual differences in negative Cognition, MASC; Dziobek, Fleck et al., affect (NA) are reflected in changes in BOLD 2006]). Imaging analyses revealed that the responses and Skin Conductance Level (SCL) AS group had thicker cerebral cortex in the during passive viewing of emotional stimuli. left fusiform gyrus and the right Temporo- Our data show differential insular activity as Parietal Junction (TPJ). Interestingly, there a function of individual differences in NA was a negative association between cortical when passively viewing aversive stimuli. Fur- thickness of the fusiform Region of Interest thermore, this finding was independent of (ROI) and facial emotion recognition in the autonomic arousal. AS, but not in the NC group (due to later in- The insula has been implicated in interocep- clusion in the study, data on facial emotion tive processes and in the integration of sen- recognition was available only for a subset of sory, visceral, and affective information, thus study participants [n: AS = 18; NC = 13]), contributing to subjective emotional experi- (Figure 12). ence. Therefore, the greater recruitment of One of the most consistent findings in autism the insula in response to aversive stimuli in spectrum conditions is increased head and subjects with high NA may indicate increased brain size (e. g., Redcay & Courchesne, 2005). interoceptive sensitivity to internal changes In addition, both the fusiform gyrus and the independent of autonomic arousal, and may TPJ have previously been identified as crucial contribute to greater self-reported NA for face processing and social cognitive func- (Mériau, Kazzer et al., 2006). tions in typically developed individuals, as has also been shown by our previous research Social Cognition and Cortical Thickness in (see Bahnemann, Dziobek, Prehn, Wolf, & Asperger Syndrome Heekeren, 2007; Prehn et al., 2006). Thus, our Asperger Syndrome (AS) represents the cog- data represent convergent neuroanatomical nitively higher functioning end of the autism and neuropsychological evidence for specific spectrum with selective deficits in social per- social perceptive and social cognitive impair- ception and cognition. Thus, brain imaging ments in AS. The results may be important in research in AS can provide important infor- the understanding of yet unknown brain be- mation on the neuronal structures, which havior relationships in autism spectrum con- are crucially involved in social cognitive ditions. processes in healthy individuals. In a previous study, we have found some evidence for Differentiating Cognitive and Emotional amygdala abnormalities and impaired social Empathy in Asperger Syndrome cognitive functions in adults with AS Empathy is a multidimensional construct en- (Dziobek, Rogers, Fleck, Wolf, & Convit, tailing cognitive (understanding others’ men- 2006). In the present study, in collaboration tal states, theory of mind) and affective with researchers from the New York Univer- (emotional reaction to the observed experi- sity School of Medicine, we investigated ences of another) components. Despite a lack whether social cognitive impairments in AS of research, deficient empathy is considered a are selective, and how they are related to central characteristic of AS. In a preliminary neuroanatomical differences. We compared a study, we profiled empathy multidimension- group of individuals with AS (n = 27) with a ally in individuals with AS using question-

Independent Research Group | 195 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 196

5.00

Lateral surface 1.97

NC 3.2 AS –1.97 AS 3.0 NC

Medial surface –5.00 2.8

2.6

2.4

2.2

Basal surface Fusiform gyrus (cortical thickness in mm) 2.0

1.8 12.0 14.0 16.0 18.0 20.0 22.0 24.0 Faces total score

Figure 12. Left: Areas of increased cortical thickness of the AS group were restricted to the fusiform gyrus (p < 0.001). Right: Associations between fusiform gyrus cortical thickness and facial emotion recognition for the AS (r = -.47, p = 0.048) and NC (r = .00, p = 0.99) groups.

naires (Rogers, Dziobek, Hassenstab, Wolf, & Rogers, Fleck, Bahnemann et al., 2006) allows Convit, 2006). In a next step, to further char- simultaneous assessment of cognitive and af- acterize empathic abilities, we used in-house fective empathy, asking study participants to developed empathy paradigms that allow rate their emotional reaction to social picture separate assessment of emotional and cogni- stimuli (Figure 13). tive components, while also aiming at eco- Our data showed that while individuals with logical validity. The MASC (Dziobek, Fleck AS seem to have impairments in inferring et al., 2006) is a video-based measure of cog- others’ mental states (cognitive empathy), nitive empathy that requires a subject to in- they are as empathically concerned for others fer mental states of movie characters. The (emotional empathy) as control subjects. In Multifaceted Empathy Test (MET) (Dziobek, a next step, we will test the hypothesis that

Figure 13. MET example Cognitive empathy Emotional empathy Emotional empathy item depicting the vari- Indirect questioning Direct questioning ous conditions.

How do you think this woman is How calm/aroused does this How concerned are you for this feeling? picture make you feel? woman? (1) Jealous (2) Defeated (3) Terrified (4) Accusing 1 24683579 135792468

196 | Independent Research Group 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 197

healthy controls and individuals with AS re- populations: In collaboration with the cruit dissociable brain networks during cog- Department of Psychiatry and Psychotherapy nitive, but not emotional empathy tasks. We of the Charité—University Medicine, Campus developed fMRI compatible adaptations of Benjamin Franklin in Berlin (Prof. Heuser), we the MASC and MET and will be using them are evaluating patients with narcissistic and with a combination of structural (brain volu- Borderline personality disorder. In collabora- metrics) and fMRI measures. tion with the Department of Social Pedagogy An important goal of our work is to make it of the Free University Berlin (Prof. Körner), applicable for clinical cognitive neuroscience we are currently examining delinquent ado- questions. In pursuit of this goal, we started lescents and school children with behavioral investigating empathic abilities in clinical deviances.

The Independent Research Group in March 2007

From left to right: Sonali Beckmann, Philipp Kazzer, Markus Bauer, Hauke Heekeren, Isabel Dziobek, Christina Scheibe, Katja Mériau, Nikos Green, Jenny Kirchner, Kristin Prehn, Guido Biele, Hermine Wenzlaff, Peter Mohr, Sandra Preißler, Janine Bayer, Ulrich Reinacher, Lea Krugel, Markus Bahnemann, Flora von Randow, Ingo Wolf.

Independent Research Group | 197 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 198

Publications 2005–2006 (last update: May 2007)

Bahnemann, M., Dziobek, I., tifaceted Empathy Test (MET). sponse modality. Proceedings of events. Acta Psychologica, 125, Prehn, K., Wolf, I., & Heekeren, Kongress der Deutschen Gesell- the National Academy of 99–116. H. R. (2007). Sociotopy of the schaft für Psychiatrie, Psycho- Sciences of the United States of superior temporal sulcus: therapie und Nervenheilkunde America, 103 (26), 10023– Prehn, K., Heekeren, H. R., Searching for dissociable repre- (DGPPN), Berlin November 2006. 10028. Raisig, S., Schiemann, T., sentations of judgments on Der Nervenarzt, 77 (Suppl. 3), Villringer, A., & van der Meer, E. movements, mental states and S343:0095. Heekeren, H. R., Wartenburger, (2005). Integration of emotional norm-congruency of behavior. I., Schmidt, H., Prehn, K., and conceptual processing in Human Brain Mapping. Abstract. Dziobek, I., Rogers, K., Fleck, S., Schwintowski, H. P., & Villringer, decision-making: Evidence from Wolf, O. T., & Convit, A. (2006). A. (2005). Influence of bodily pupillometry. Journal of Cogni- Bauer, M., Oostenveld, R., The amygdala theory of autism harm on neural correlates of tive Neuroscience (Suppl.), Peeters, M., & Fries, P. (2006). revisited: Linking structure to semantic and moral decision- 169–169. Tactile spatial attention en- behaviors. Neuropsychologia, 44 making. NeuroImage, 24 (3), hances gamma-band activity in (10), 1891–1899. 887–897. Prehn, K., Wartenburger, I., somatosensory cortex and re- Mériau, K., Scheibe, C., duces low-frequency activity in Dziobek, I., Rogers, K., Fleck, S., Heekeren, H. R., Marrett, S., Goodenough, O .R., Villringer, A., parieto-occipital areas. Journal Hassenstab, J., Gold, S., Wolf, O. Bandettini, P. A., Ungerleider, van der Meer, E., & Heekeren, of Neuroscience, 26 (2), T., & Convit, A. (2005). In search L. G. (2004). A general mecha- H. R. (2006). Influence of indi- 490–501. of “master mindreaders”: Are nism for perceptual decision- vidual differences in moral psychics superior in reading the making in the human brain. judgment competence on neural Biele, G., Krugel, L., Rieskamp, language of the eyes? Brain and Nature, 431, 859–861. correlates of normative judg- J., & Heekeren, H. R., (2007). Cognition, 58 (2), 240–244. ments. Human Brain Mapping. The influence of advice on re- Marschner, A., Mell, T., Warten- Abstract. ward based decision-making Gold, S. M., Dziobek, I., Rogers, burger, I., Villringer, A., and learning. Human Brain K., Bayoumy, A., McHugh, P. F., & Reischies, F. M., & Heekeren, H. Preuschhof, C., Heekeren, H. R., Mapping. Abstract. Convit, A. (2005). Hypertension R. (2005). Reward-based deci- Taskin, B., Schubert, T., & and hypothalamo-pituitary- sion-making and aging. Brain Villringer, A. (2006). Neural cor- Dziobek, I., Gold, S., Wolf, O. T., adrenal axis hyperactivity affect Research Bulletin, 67 (5), relates of vibrotactile working & Convit, A. (2007). Hypercho- frontal lobe integrity. Journal 382–390. memory in the human brain. The lesterolemia in Asperger syn- of Clinical Endocrinolgy and Journal of Neuroscience, 26, drome: Independence from Metabolism, 90 (6), 3262–3267. Mell, T., Heekeren, H. R., 13231–13239. lifestyle, obsessive-compulsive Marschner, A., Wartenburger, I., behavior, and social anxiety. Goodenough, O. R, & Prehn, K. Villringer, A., & Reischies, F. M. Rogers, K., Dziobek, I., Hassen- Psychiatry Research, 149 (1–3), (2004). A neuroscientific ap- (2005). Effect of aging on stim- stab, J., Wolf, O. T., & Convit, A. 321–324. proach to normative judgment ulus-reward association learn- (2006). Who cares? Revisiting in law and justice. Philosophical ing. Neuropsychologia, 43 (4), empathy in Asperger syndrome. Dziobek, I., Wolf, I., Preißler, S., Transactions of the Royal Soci- 554–563. Journal of Autism and Develop- Kirchner, J., Bahnemann, M., ety of London, Series B: Biologi- mental Disorders, Aug. 12 [Epub Convit, A., & Heekeren H. R. cal Sciences, 359, 1709–1726. Mériau, K., Kazzer, P., Warten- ahead of print]. (2007). Neuropsychological and burger, I., Prehn, K., Villringer, neuroanatomical evidence for Hassenstab, J., Dziobek, I., A., van der Meer, E., & Scheibe, C., Ullsperger, M., social perceptive and social cog- Rogers, K., Wolf, O. T., & Convit, Heekeren, H. R. (2006). Insular Sommer, W., & Heekeren, H. R. nitive impairments in adults A. (2007). Knowing what others activity reflects individual dif- (2007). Spatiotemporal dynam- with Asperger syndrome. Human know, feeling what others feel: ferences in negative affect ics munderlying the integration Brain Mapping. Abstract. A controlled study of cognitive independent of autonomic of prior probability information and affective empathy in psy- arousal. Human Brain Mapping. in decision making. Conference Dziobek, I., Fleck, S., Kalbe, E., chotherapists. Journal of Ner- Abstract. of Cognitive Neuroscience, Rogers, K., Hassenstab, J., Brand, vous and Mental Disease, 195, New York. Abstract. M., Kessler, J., Woike, J., Wolf, O. 277–281. Mériau, K., Wartenburger, I., T., & Convit, A. (2006). Introduc- Kazzer, P., Prehn, K., Lammers, Scheibe, C., Schubert, R., ing MASC: A movie for the as- Heekeren, H. R., Wartenburger, C.-H., van der Meer, E., Sommer, W., & Heekeren, H. R. sessment of social cognition. I., Mell, T., Marschner, A., Villringer, A., & Heekeren, H. R. (2006). Integration of prior Journal of Autism and Develop- Villringer, A., & Reischies, F. M. (2006). A neural network re- probability in the process of mental Disorders, 36 (5), (in press). Role of ventral stria- flecting individual differences in decision making. Human Brain 623–636. tum during reward-based deci- cognitive processing of emo- Mapping Conference, Florence. sion-making. NeuroReport. tions during perceptual deci- Abstract. Dziobek, I., Rogers, K., Fleck, S., sion-making. NeuroImage, 33, Bahnemann, M., Heekeren, H. Heekeren, H. R., Marrett, S., 1016–1027. Scheibe, C., Wartenburger, I., R., Wolf, O. T., & Convit, A. Bandettini, P. A., Ungerleider, L. Wüstenberg, T., Villringer, A., & (2006). Differenzierung kogni- G. (2006.) Involvement of hu- Pachur, T., & Biele, G. (2007). Heekeren, H. R. (2005). Neural tiver und affektiver Emapthie man left dorsolateral prefrontal Forecasting from ignorance: The correlates of the interaction be- bei Menschen mit Asperger Syn- cortex in perceptual decision- use and usefulness of recogni- tween transient and sustained drom: erste Ergebnisse des Mul- making is independent of re- tion in lay predictions of sports processes—A mixed blocked/

198 | Independent Research Group 09_NWG_180_199.qxd 19.07.2007 15:22 Uhr Seite 199

event-related fMRI study. Human Brain Mapping, Sep. 2 [Epub ahead of print].

de Wit, T. C., Bauer, M., Oosten- veld, R., Fries, P., & van Lier, R. (2006). Cortical responses to contextual influences in amodal completion. NeuroImage, 32 (4), 1815–1825.

Wolf, O. T., Dziobek, I., McHugh, P. F., Sweat, V., Javier, E., de Leon, M. J., & Convit, A. (2005). Subjective memory complaints in aging are associated with elevated cortisol levels. Neuro- biology of Aging, 26, 1357– 1363.

Independent Research Group | 199 10_Service_200_207.qxd 18.07.2007 14:02 Uhr Seite 200 10_Service_200_207.qxd 18.07.2007 14:02 Uhr Seite 201

Service Units 10_Service_200_207.qxd 18.07.2007 14:02 Uhr Seite 202

Contents

Library and Research Information ...... 203

Information Processing Center ...... 205

Scientific and Professional Staff (2005–2006)

Library and Research Information: Nicole Engelhardt, Ursula Flitner

Information Processing Center: Wolfgang Assmann

202 | Service Units 10_Service_200_207.qxd 18.07.2007 14:02 Uhr Seite 203

Library and Research Information

Rapid access to printed and digital information is a decisive prerequisite for successful studies and for internationally renowned research. The Library and Research Information Unit of the Max Planck Institute for Human Develop- ment aims to anticipate, determine, and respond to the Institute's needs for information in the areas of education, sociology, psychology, and neighboring disciplines. To support the research, teaching, and publishing activities of the Institute's researchers, the Library seeks to provide an environment and facilities conducive to efficient and independent use and dissemination of information.

The Library's collection currently comprises Among the new electronic resources added to around 200,000 volumes, 520 printed perio- the Library’s services are: dicals, several thousand electronic journals, – CogNet, the brain sciences portal main- and an extensive selection of electronic re- tained by the Massachusetts Institute of sources. It offers easy and fast access to our Technology; own materials and information worldwide. – The Cochrane Library, a collection Comprehensive intranet services include, of databases to inform health-care among other things, online access to major decision making; bibliographic and abstract databases, to – the International Bibliography of Periodical portals and e-books, to document delivery Literature (IBZ), containing almost 2.8 mil- services, and listings of new acquisitions. lion journal articles from about 11,000 In 2005 and 2006, the range of electronic full journals mainly from the Humanities, texts available within the context of the Max Social Sciences, and Arts; Planck Society’s basic provision (Grundver- – the International Bibliography of Book sorgung) was expanded to include a total of Reviews (IBR), containing over 1 million more than 15,000 journals on all fields of entries on book reviews from the year 1985 science, thus offering a sound basis for inter- onwards, published in approximately 6,600 disciplinary research. academic journals mainly on the Humani- The collection of centrally and locally li- ties and Social Sciences; cenced electronic resources, accessible from – SocINDEX with Full Text, a sociological every office desk in the Institute, is growing database containing full text for several steadily. hundred journals, for more than 700 books

Service Units | 203 10_Service_200_207.qxd 18.07.2007 14:02 Uhr Seite 204

and monographs, as well as conference full-text collection in spring 2005: the Insti- papers; tute’s Serials—Studien und Berichte, since – SourceOECD organizes the OECD’s authori- 1965, and Materialien aus der Bildungs- tative research output into a full-text on- forschung, since 1972—with a total of 173 line library that consists of more than 20 volumes are now available online in formats periodicals, 24 statistical databases, refer- allowing smooth access both from inside and ence titles, thematic book collections, and outside high-performance networks. working papers; A moving wall of 5 years was deemed an ap- – Periodicals Index Online, currently indexing propriate provision to secure interest in the more than 15 million articles from the Hu- ongoing publication of the printed volumes. manities reaching back to the year 1665. Usage reports show that this offer is in high In addition, the Institute profits from the Na- demand. tional Licences, a program financed by the In 2005, the Library also addressed the issue German Research Foundation (DFG), offering of catalog enrichment. free access on a national level to a vast num- The idea is to enhance library catalog brows- ber of databases, digital full-text collections, ing by adding content to help users make and e-journals. more informed decisions on the applicability The staff of the Library continue to offer reg- of a resource from the catalog record. ular training in the use of databases. This seems all the more desirable given the Additional funds were raised to satisfy the closed-stack shelving of most of our Library’s information needs of the new Independent collections. Junior Research Group “Neurocognition of Thus, in fall 2005, the Library started to digi- Decision Making.” tize tables of content of newly acquired To enhance the visibility of the Institute’s sci- books to be included in the library catalog. entific output, the Library took up digitaliza- Since 2005, we have been training an ap- tion in 2004 and launched its first electronic prentice in Library and Information Science.

204 | Service Units 10_Service_200_207.qxd 18.07.2007 14:02 Uhr Seite 205

Information Processing Center

The Information Processing Center supports projects and other service units in the Institute through its central facilities. Central servers are installed with Windows 2000/2003 or LINUX for dedicated purposes: internet/intranet servers, software server, etc. Five Windows 2000 cluster systems with large RAID storage installations provide the capacity (more than 10 TByte) for the central data management. Several powerful terminal servers establish a CITRIX server farm. They allow the user to run programs (SPSS, SAS, MAT-Lab, EQS, etc.) on the server CPUs from their own workstations (Windows or Apple PC) or the Internet. “Server- based computing” helps to overcome the constraints of the different workstations concerning CPU power and local storage. There has been a paradigm shift from the classical RAID system to a SAN solution on the central servers in 2005. Thereby, 80 TByte has been created in the last 2 years.

The decentralized personal computing capacity comprises about 300 Intel PCs and 150 Apple computers. The Apple computers are running MacOS X and the Intel PCs are operating on Microsoft Windows 2000 or Windows XP. A central backup service is provided for all data on cluster and SAN disks. A wide array of software is available for the desktop systems. To provide the necessary security, a Cisco-PIX firewall system was installed which allows for the failure of certain modules without causing a complete breakdown. Central virus scanner software—continuously updated via the internet—monitor all Intel and Apple workstations to avoid data loss caused by viruses. The Local Area Network (LAN) integration of all desktop computers provides access to central resources and cluster capacity. At the end of 2004, the previous “fiber-to-desktop” solution of 10 Mbit/s was replaced by the “fiber-to-office” concept. Each office is now connected with 1,000 Mbit/s, and here the The Center’s services include: changeover from fiber to four copper ports – operating, optimizing, and developing the takes place by using a small Gigabit Ethernet devices of the central cluster and network Switch. equipment; To provide notebook users with a wireless – maintenance and updating Windows 2000, connection to the internet, the Institute has XP, and Linux operating systems; installed WLAN. – centralized printing capacity, including The Institute’s connectivity to Wide Area high-speed and color printers; Networks (internet, etc.) is provided via the – LAN integration of desktop computers, and Research Network (WIN) of German Telekom the continuous enhancement of LAN facili- and the German Research Network Associa- ties; tion (Deutsches Forschungsnetz [DFN]). Late – national and international connectivity in the year 2000, the GIGA-WIN (1 Gbit/s) (Wide Area Network); was installed, allowing the Institute to in- – internet services: E-mail, WWW, NEWS, crease the use of high-speed connectivity. FTP, and Telnet;

Service Units | 205 10_Service_200_207.qxd 18.07.2007 14:02 Uhr Seite 206

– maintenance and webmaster tasks of the Institute’s internet and intranet servers; – security measures; – management of the central telephone system, including the voice-mail server; – user support and trouble-shooting for Intel PCs and Apple computers; – coordination and technical support for desktop computers and software; – software acquisition.

The Center provides: – general design and coordination of the In- stitute’s information technology equipment; – the documentation of data concerning the existing computer and network equipment; – an overview of market developments; – advice for the Institute’s boards and de- partments; – the promotion of new concepts for state- of-the-art computer equipment.

206 | Service Units 10_Service_200_207.qxd 18.07.2007 14:02 Uhr Seite 207

Service Units | 207 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 208

Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 209 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 210

Contents

1. Research Colloquia 2005–2006 ...... 211

2. Visiting Scientists 2005–2006 ...... 213

3. Other Professional Activities 2005–2006 ...... 215

4. Academic Degrees 2005–2006 ...... 218

5. Scientific and Professional Staff 2005–2006 ...... 220

Last update: May 2007

210 | Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 211

1. Research Colloquia 2005–2006

Kaarin J. Anstey Neal Roese 2006 Australian National University: University of Illinois: The Relationship Between Weekly Alcohol Con- The Functional Basis of Counterfactual Thinking sumption, Cognitive Function and Brain Structure May 9 in Young, Middle, and Older Adulthood Michael Schäfer September 5 National Institute for Neurological Disorders and Jean Decety Stroke, Bethesda, MA: University of Chicago: Social Neuroscience: From Body Perception Over A Developmental and Cognitive Neuroscience View Mirror Neurons to Culture of Human Empathy January 18 June 13 Glenn Schellenberg Hartmut Ditton University of Toronto at Mississauga: Ludwig Maximilians University Munich: Does Music Make You Smarter? Übergänge auf weiterführende Schulen nach der April 8 Grundschule: primäre und sekundäre Effekte Wolfgang Schneider September 27 University of Würzburg: Jon Driver Die Entwicklung von Metakognition bei Kindern Institute of Cognitive Neuroscience at University und Jugendlichen College London: November 28 Crossmodal Links in Human Spatial Attention Lee S. Shulman October 26 Carnegie Foundation for the Advancement of Jutta Heckhausen Teaching, Stanford: University of California, Irvine: "Signature Pedagogies" in the Professions: Their Developmental Regulation Across the Life Span Characteristics and Lessons Learned for the Power October 24 of Professional Eduation Alumit Ishai September 7 University of Zurich: Martin Sliwinski fMRI Studies of Face Perception: Effects of Syracuse University, New York: Memory, Emotion & Sexual Preference Intraindividual Variability and Change in Affect, March 9 Stress and Health Richard M. Lerner May 23 Tufts University, Medford, MA: Piotr Winkielman The Handbook of Child Psychology, 6th ed. (2006): University of California, San Diego: An Editor's Perspective Preferences With and Without Inferences June 16 May 8 Ulrich Mayr Marjorie H. Woollacott University of Bonn: University of Oregon: Navigating Through Task Space: The Role of Cues Attentional Factors Contributing to Balance and Inhibition Constraints During Gait in Older Adults: Can These June 20 be Reduced With Training? Günther Palm November 21 University of Ulm: Zhou Xiaolin Neuroinformatik: Experiment, Theorie und Praxis Beijing University China: February 24 The Automaticity and Flexibility of the Mental Tenko Raykov Number Line Michigan State University: February 24 Change Measurement and Program Evaluation Using Latent Variable Modeling October 31

Uschi Backes-Gellner Hans-Peter Blossfeld 2005 University of Zurich: University of Bamberg: Incentives and Research Productivity in Universi- Educational Systems as Marriage Markets ties: A Business Economist's Analysis and Empirical January 5 Evidence Marlis Buchmann May 3 University of Zurich/ETH: Lars Bäckman New Challenges in Life Course Research: Karolinska Institute, Stockholm: Persistent Gaps and Promising Avenues Dopamine and Cognitive Aging February 1 May 31 Hannah Brückner Fredda Blanchard-Fields Yale University: Georgia Institute of Technology, Atlanta: Capitalizing on the Life Course Aging, Affective Processing, and Social Cognition March 7 August 9

Appendix | 211 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 212

2005 Lynn Prince Cooke Vered Kraus Nuffield College, Oxford University: University of Haifa: What Does "Doing" Gender Do for Marriage? Secondary Breadwinner: Israeli Women in the Household Bargaining and Risk of Divorce in Labor Force, January 21 Germany and the United States John McArdle May 30 University of Southern California: Thomas DiPrete Recent Developments in Latent Growth Curve Columbia University: Modeling Life Course Destandardization, Social Polarization, September 16 and Social Interaction: A Reflection on Recent Earl K. Miller American Trends and a Proposal for Research Massachusetts Institute of Technology: April 22 The Prefrontal Cortex: Categories, Concepts, and Ido Erev Cognitive Control Technion—Israel Institute of Technology, Haifa: May 24 On Adaption, Maximization, and Reinforcement Peter Molenaar Learning Among Cognitive Strategies University of Amsterdam: November 28 Idiographic Psychology: The Necessary and Princi- Ute Frevert pled Psychometric Answer to the Uniqueness of Yale University: Human Persons and its Theoretical and Practical Europeanizing German History Implications April 28 March 15 Yoav Ganzach Roger Ratcliff Tel Aviv University: Ohio State University: On the Perception of Financial Risk (and Return) Applications of Diffusion Models in Cognitive Aging May 2 and in Analyses of Single-Neuron Recording Data Adam Gamoran April 19 University of Wisconsin-Madison: Henry L. Roediger, III Racial Gaps and School Effects during School Washington University, St. Louis Re-Segregation Illusory Memories: Remembering Events That Never April 1 Happened Theodore P. Gerber May 9 University of Wisconsin-Madison: André A. Rupp Economic Crisis or Second Demographic Transition? University of Iowa: Trends and Correlates of Union Formation in Hunting the Ghost in the Machine: General and Russia, 1985–2001 Cognitive Psychometric Methodologies for Model- April 26 ing and Understanding Response Processes for David Grusky Language Tests Stanford University: February 24 Measuring Inequality: The Heavy Weight of Conven- Robert J. Sampson tion, the Case for Overcoming it, and Why a New Harvard University: Social Indicators Framework Must be Developed Delinquent Boys Grown Up: Trajectories of Crime April 18 from Childhood to Age 70 Agneta Herlitz January 7 Karolinska Institute, Stockholm: David P. Schmitt Sex Differences in Cognition Bradley University, Illinois: April 26 The Many Cultures of Short-Term Mating: How Christopher Hertzog Gender, Personality, and Local Ecology Relate to Georgia Institute of Technology, Atlanta: High-Risk Sexual Behavior in Humans Individual Differences in Strategy Use for December 8 Associative Learning Brent Small May 3 University of South Florida: Hachiro Iwai Growth Mixture Models of Cognitive Impairment Kyoto University: and Cancer-Related Fatigue The Japanese Life Course After the Postwar June 14 Economic Growth Stephen M. Stigler January 28 University of Chicago: Douglas M. Jones The History of Maximum Likelihood From Lagrange University of Utah, Salt Lake City: to Fisher Universals of Language and Kinship: July 19 A Perspective From Evolutionary Psychology Nikolaus Troje December 12 Queen's University: Michael Kahana Local and Global Information in Animate Motion University of Pennsylvania: Patterns, June 21 Human Memory and its Neural Mechanisms Duncan Watts June 28 Columbia University: Networks, Collective Dynamics, and the Small- World Problem, January 11

212 | Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 213

2. Visiting Scientists 2005–2006

Kaarin Anstey Thomas Joseph 2006 Australian National University, Canberra Indian Institute of Management, Bangalore September 2006 May 2006 Nathan Berg Linnea Karlsson University of Texas, Dallas Umea University June–August 2006 September–October 2006 Martin Bischoff Daniel Koretz University of Erlangen-Nuremberg Harvard Graduate School of Education April 2006 December 2006 Fredda Blanchard-Fields Joachim Krüger Georgia Institute of Technology, Atlanta Brown University, Providence June–July 2006 May 2006 Steven Boker Erno Lehtinen University of Notre Dame, Indiana University of Turku, Finnland June–July 2006 May 2006 Eduard Brandstätter Dan McGrath Johannes Kepler University Linz Education Statistics Services Institute, American August 2006 Institutes for Researach, Washington, DC Ruth Butler January–February 2006 Hebrew University of Jerusalem, School of Mary McLaughlin Education, Jerusalem Education Statistics Services Institute, American February 2006 Institutes for Researach, Washington, DC Sy-Miin Chow January–February 2006 University of Notre Dame, Indiana John McNamara June–July 2006 University of Bristol Hartmut Ditton October 2006 Ludwig Maximilians University Munich Kevin Miller September 2006 Department of Educational Studies and Psychology, Daniel Goldstein Center for Human Growth and Development, London Business School University of Michigan January 2006 April–May 2006 Ed Elbers Bruce Moulton University of Utrecht University of Technology, Sydney February 2006 September–October 2006 Daniel Frey Shabnam Mousavi Massachusetts Institute of Technology Pennsylvania State University September–October 2006 August 2006 Brit Grosskopf Nilan Ram Texas A&M University Pennsylvania State University June 2006 June–July 2006 Kenneth Hammond Rajiv Sarin University of Colorado at Boulder Texas A&M University September 2006 June 2006 Werner Helsper Lee S. Shulman Martin Luther University Halle-Wittenberg Carnegie Foundation for the Advancement of April 2006 Teaching, Stanford Christopher Hertzog September 2006 Georgia Institute of Technology, Atlanta Nandan Sudarsanam June–July 2006 Massachusetts Institute of Technology Floris Heukelom September–October 2006 University of Amsterdam Peter M. Todd January–June 2006 University of Indiana, Bloomington Richard Hwang June–August 2006 Academia Sinica, Taiwan November–December 2006

Appendix | 213 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 214

2005 Morris Altman Kathleen B. McDermott University of Saskatchewan Washington University, St. Louis September 2005 May 2005 Lars Bäckman Leo Montada Karolinska Institute, Stockholm University of Trier July 2004–July 2005 September 2005 Fredda Blanchard-Fields Thad Polk Georgia Institute of Technology, Atlanta University of Michigan, Ann Arbor January–July 2005 May–June 2005 Gary Bornstein Richard M. Ryan Hebrew University of Jerusalem University of Rochester September 2005 June–July 2005 Arndt Bröder Naftali Raz University of Bonn Wayne State University, Detroit March–September 2005 June–August 2005 Gayle Christensen Henry L. Roediger, III Urban Institute, Washington, DC Washington University, St. Louis January–August 2005 May 2005 Ido Erev David P. Schmitt Technion—Israel Institute of Technology Bradley University November 2005 December 2005 Claudia Gonzalez-Vallejo Stephen M. Stigler Ohio University, Athens University of Chicago January–August 2005 July 2005 Agneta Herlitz Helen M. G. Watt Karolinska Institute, Stockholm School of Education, University of Michigan, July 2004–July 2005 Ann Arbor Christopher Hertzog September–October 2005 Georgia Institute of Technology, Atlanta Christoph Weidemann January–July 2005 University of Indiana, Bloomington Douglas M. Jones July 2005 University of Utah, Salt Lake City December 2005 Yongfang Liu East China Normal University, Shanghai July–September 2005

214 | Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 215

3. Other Professional Activities 2005–2006

Paul B. Baltes – Academia Europaea (Election Committee, Section on Psychology and Behavioral Sciences; Member of Trust). – Berlin-Brandenburgische Akademie der Wissenschaften (Vice-Chair, Working Group on Science History and Psychology History). – US National Academy of Sciences and National Research Council (Member of Committee on Future Research in Cognitive Aging, 2002–2003, and of Committee on Assessing Behavioral and Social Science Research on Aging, 2004–2006). – International University of Bremen (Member of Board of Governors). – Jacobs Foundation (Member of Board of Trustees and Jacobs Family Council). – Deutsche Akademie der Naturforscher Leopoldina (Vice-President). – IPSEN Foundation, IPSEN Longevity Award (Member of Jury Selection). – University of Konstanz, Center for Junior Research Fellows (Member of Kuratorium).

Jürgen Baumert – Max Planck Society (Vice-President, Human Sciences Section). – Jacobs Foundation, Zurich, Switzerland (Member of Board of Trustees). – Institute for Educational Progress, Humboldt University Berlin (Member of Board of Trustees). – Research Program “Professional Minds of Teachers: On the Development of Standards for Vocational Teachers”, University of Fribourg, Switzerland (Member of Advisory Board). – Quality Agency of the State Institute of School Education and Educational Research (ISB), Munich (Member of Board of Scientific Advisers). – University of Twente, Institute for Behavioral Research (IBR) (Member of Board of Trustees). – Action Program “New Paths in Teacher Training”, Founders’ Association of German Science, Bonn (Member of Board of Scientific Advisers). – Leibniz Association (Member of Senate and Evaluation Committee). – Internationale Grundschul-Lese-Untersuchung (IGLU)/Progress in International Reading Literacy Study (PIRLS) (Member of Advisory Board). – Center for Research on Learning and Instruction, University of Erfurt (Member of Advisory Board). – Psychologie in Erziehung und Unterricht (Member of Advisory Board). – Schweizerische Zeitschrift für Bildungswissenschaften (Member of Advisory Board). – Zeitschrift für Erziehungswissenschaft (Coeditor). – Zeitschrift für Pädagogische Psychologie (Member of Advisory Board). – Center for School Research and Questions of Teacher Education, Martin Luther University Halle-Wittenberg (Member of Advisory Board). – Waxmann Verlag, Reihe Pädagogische Psychologie und Entwicklungs- psychologie (Member of Advisory Board). – Zeitschrift für Unterrichtswissenschaft (Coeditor). – Jacobs Center of Lifelong Learning and Institutional Development (JCLL), International University Bremen (Member of Advisory Board). – European Science Foundation, Strasbourg (Member of Pedagogical and Educational Research Expert Panel). – University of Kassel (Member of University Council). – "Bürgernetzwerk Bildung" (Member of Advisory Board). – State Institute for School Quality in Berlin und Brandenburg (ISQ) (Member of Advisory Board). – Expert Committee to Assess Teacher Training in North Rhine-Westphalia (Chair). – International Academy of Education (Fellow). – Carnegie Foundation for the Advancement of Teaching (Member of Board of Trustees). – "Haus der kleinen Forscher" (McKinsey) (Member of Advisory Board). – "School Reform and Beyond" (David L. Featherman) (Member of Advisory Board). – Dahlem Conferences (Member of Advisory Board).

Nathan Berg – Journal of Socio-Economics (Editorial Board). – Global Business Economics Review (Member of Advisory Board). – Society for the Advancement of Behavioral Economics (Member of Board). – European Science Foundation (Member of Program Review Committee).

Appendix | 215 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 216

Henry Brighton – The First Conference on Artificial General Intelligence (AGI-08) (Member of Program Committee).

Wolfgang Edelstein – Institut für angewandte Familien-, Kindheits- und Jugendforschung e.V., Potsdam (Member of Board of Scientific Advisers). – BLK-Program “Demokratie lernen und leben” (Learning and Living Democracy) (Member of Steering Committee). – Irmgard-Coninx-Stiftung (Member of Advisory Council). – Kulturstiftung der Länder, Program “Kinder zum Olymp” (Member of Board of Trustees). – Yehudi Menuhin Stiftung Deutschland (Member of Board of Trustees). – Deutsche Gesellschaft für Demokratiepädagogik (German Association for Democratic Education) (President). – Social Justice Research, New York (Member of Editorial Board). – New Directions for Youth Development, New York (Member of Editorial Board). – Entwicklungswissenschaft. Biopsychosoziale Forschung und Anwendung“, Verlag Vandenhoeck & Ruprecht, Göttingen (Member of Board of Advisers).

Ursula Flitner – German Special Library Association (ASpB), Section 5 of the German Library Association (DBV) (Deputy Chair). – German Library Association (DBV) (Board Member of Berlin Division). – Scopus Content Selection & Advisory Board.

Gerd Gigerenzer – Berlin-Brandenburgische Akademie der Wissenschaften (Member). – Deutsche Akademie der Naturforscher Leopoldina (Fellow). – Theory and Psychology (Coeditor). – Journal of Behavioral Decision Making (Editorial Board). – Evolution and Human Behavior (Editorial Board). – Summer Institute on Bounded Rationality in Psychology and Economics (Codirector). – British Medical Journal (Member of Advisory Committee). – Organizational Behavior and Human Decision Process (Editorial Board). – Bundesinstitut für Risikobewertung, Germany (Advisory Board). – European Society for Philosophy and Psychology ESPP (Advisory Board). – Psychological Inquiry (Editorial Board). – Frankfurt Institute for Advanced Studies Forum (Advisory Board).

Hauke Heekeren – Acta Neuropsychiatrica (Editorial Board).

John M. C. Hutchinson – Animal Behavior (Consulting Editor). – Folia Malacologica (Advisory Board).

Konstantinos V.Katsikopoulos – IEEE Transactions on Systems, Man and Cybernetics (Associate Editor).

Monika Keller – Erwägen, Wissen, Ethik (Editorial Board). – European Developmental Science (Editorial Board). – Jean Piaget Society (ex-officio Member of Board of Directors).

Shu-Chen Li – Neuroscience and Biobehavioral Reviews (Board of Consulting Editors). – Psychology and Aging (Board of Consulting Editors). – Brain Research Center, School of Computer Science, National Chiao-Tung University, Taiwan (International Collaborative Faculty). – International Max Planck Research Network on Aging (Deputy Director). – Max Planck Society (Institute Scientific Staff Representative in the Human Sciences Section).

216 | Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 217

Ulman Lindenberger – European Journal of Developmental Science (Member of Editorial Board). – Centre for Population Studies (CPS), Umea University, Sweden (Member of the Board of Ageing and Living Conditions Programme, ALC) – Margret M. and Paul B. Baltes Foundation for the Advancement of Research in Developmental Psychology and Gerontology (Member of the Board) – Nordic Centre in Cognitive Control (Member of the Scientific Advisory Board) – Gerontological Society of America (Fellow) – Wilhelm-Wundt-Gesellschaft (Member). – Alexander von Humboldt-Stiftung (Member and Selection Committee for Humboldt Research Awards). – Deutsche Akademie der Naturforscher Leopoldina (Fellow). – AG LeoTech Alter: Opportunities and Problems of an Aging Society: The World of Work and Lifelong Learning, launched together by the German Academy of Sciences Leopoldina and acatech-Council for Engineering Sciences at the Union of the German Academies of Science and Humanities (Member of Working Group). – Psychology and Aging (Member of Editorial Board). – Aging, Neuropsychology, and Cognition (Member of Editorial Board). – International Journal of Behavioral Development (Member of Editorial Board).

Lael J. Schooler – 19th International Congress of Psychology. Organizing a series of controvertial debates for the ICP (Member of Scientific Committee). – Frankfurt Institute for Advanced Studies Forum (Member of Program Advisory Committee). – Berlin Graduate School of Mind and Brain (Member of Curriculum Committee). – International Conference on Cognitive Modeling (Member of Program Committee).

Jacqui Smith – Academia Europaea (Member). – Journal of Gerontology: Psychological Sciences (Member of Editorial Board). – English Longitudinal Study (ELSA) (Member of Advisory Board). – Survey of Older Adults in Germany (Member of Advisory Board). – China Healthy Longevity Study (Research Consultant). – Norwegian Longitudinal and Cohort Study of Aging (Research Consultant). – Max Planck International Research Network on Aging (Deputy Director).

Elsbeth Stern – Mind, Brain, and Education (Associate Editor). – Science (Board of Reviewing Editors). – Academy of Finland (Board of Reviewers for Educational Research). – Zeitschrift für Pädagogische Psychologie (Member of Editorial Board). – Deutsches Instritut für Internationale Pädagogische Forschung (Member of Evaluation Group). – Microsoft-Stiftung “Wissenswert” (Member of Scientific Board). – Max Planck Society, Working Committee for the Advancement of Women in Science (Vice-Chair).

Appendix | 217 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 218

4. Academic Degrees 2005–2006

Habilitations

Artelt, C. (2005). Lesekompetenz und Selbstregulier- Stanat, P. (2005). Disparitäten im schulischen Erfolg: tes Lernen. University of Postdam. Analysen zur Rolle von Migrationshintergrund und Geschlecht. Free University Berlin. Kunzmann, U. (2006). Die differentielle Entwicklung lebenspragmatischer Kompetenz im Erwachsenenalter Trautwein, U. (2005). Hausaufgabenvergabe und und Alter. Free University Berlin. Hausaufgabenerledigung: Entwicklung und Überprü- fung eines pädagogisch-psychologischen Mehr- Li, S.-C. (2006). The ontogeny of the shared and per- ebenenmodells. Free University Berlin. sonalized brain: Contextualized lifelong development of adaptive neurocognitive representations. Free Uni- Watermann, R. (2005). Soziale Disparitäten der Bil- versity Berlin. dungsbeteiligung und des Kompetenzerwerbs. Free University Berlin.

Doctoral Dissertations

Biele, G. (2006). No man is an island: Cooperation in Höhne, A. (2005). Soziale Unterstützung in Partner- groups and social learning. Free University Berlin. schaften: Individuelle Beschäftigungschancen und -risiken im ostdeutschen Transformationsprozess Brehmer, Y. (2006). Episodic memory plasticity across unter Berücksichtigung des Partnerschaftskontexts. the lifespan. Humboldt University Berlin. Free University Berlin.

Brunner, M. (2006). Mathematische Schülerleistung: Huxhold, O. (2007). Processing fluctuations in pos- Struktur, Schulformunterschiede und Validität. tural control: Relations to adult age and fluctuations Humboldt University Berlin. in cognition. Free University Berlin.

Dieckmann, A. (2005). The ecological rationality of Kopp, F. (2006). Rehearsal-Prozesse im Arbeits- heuristics and their building blocks: The making of gedächtnis und EEG-Kohärenz. University of Leipzig. adaptive decisions. Free University Berlin. Limbird, C. (2006). Phonological processing, verbal Dziobek, I. (2006). Biopsychological markers and neu- abilities, and second language literacy development ral correlates of Asperger syndrome: Social cognition, among bilingual turkish children in Germany. Free lipid profiles, and magnetic resonance imaging find- University Berlin. ings. University of Bielefeld. Maaz, K. (2006). Primäre und sekundäre Disparitäten Ebner, N. C. (2005). Striving for gains and preventing des Hochschulzugangs. Free University Berlin. losses: Multi-method evidence on the differences in personal goal orientation in early and late adulthood. Mata, R. (2006). The aging decision maker: Cognitive Free University Berlin. aging and the use of decision strategies. Free Univer- sity Berlin. Felbrich, A. (2005). Kontrastierungen als effektive Lerngelegenheiten zur Vermittlung von Wissen über McElvany, N. (2006). Förderung von Lesekompetenz Repräsentationsformen am Beispiel des Graphen einer im Kontext der Familie. Free University Berlin. linearen Funktion. Technical University Berlin. Müller, A. G. (2007). Aspekte schulbezogener Sprache Gaissmaier, W. (2007). The mnemonic decision als Barriere für schulischen Erfolg von Schülerinnen maker: How search in memory shapes decision mak- und Schülern deutscher und nichtdeutscher Herkunfts- ing. Free University Berlin. sprache. Free University Berlin.

Gummerum, M. (2005). The development of prosocial Nagy, G. (2006). Berufliche Interessen, kognitive und behavior: Integrating psychological, economic, and fachgebundene Kompetenzen: Ihre Bedeutung für die evolutionary perspectives. Free University Berlin. Studienfachwahl und die Bewährung im Studium. Free University Berlin. Gundert, S. (2006). Befristete Beschäftigung bei Berufsanfängern und älteren Arbeitnehmern. Free Pachur, T. (2006). Ecological rationality: Do samples University Berlin. in memory reflect the world? Free University Berlin.

Grühn, D. (2006). Aging and emotional memory: Evi- Röcke, C. (2006). Intraindividual variability in positive dence from a heterogeneity-homogeneity list para- and negative affect: Age-related and individual dif- digm using emotionally-toned words. Free University ferences in magnitude and coupling with cognitive Berlin. performance. Free University Berlin.

218 | Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 219

Saalbach, H. (2006). The cognitive influence of a Schneider, M. (2006). Konzeptuelles und prozedurales classifier system. Technical University Berlin. Wissen als latente Variablen: Ihre Interaktion beim Lernen mit Dezimalbrüchen. Technical University Berlin. Schäfer, S. (2005). Concurrent cognitive and sensorimotor performance: A comparison of children and Stange, A. (2005). The social dimension of wisdom: young adults. Free University Berlin. Conditions for perceiving advice-giving persons as wise. Free University Berlin. Scheibe, S. (2005). Longing ("Sehnsucht") as a new lifespan concept: A developmental conceptualization Wilke, A. (2006). Evolved responses to an uncertain and its measurement in adulthood. Free University world. Free University Berlin. Berlin.

Master’s and Diploma Theses

Blumenthal, A. (2006). The development of perceived Neumann, M. (2006). Der Einfluss schulischer Kon- control in old age: A longitudinal and multidimen- textbedingungen auf die Leistungsentwicklung in sional approach. Free University Berlin. Französisch als Fremdsprache: Eine mehrebenen- analytische Untersuchung in der deutschsprachigen Brose, A. (2005). Eine experimentelle Induktion von Schweiz. Humboldt University Berlin. Sehnsucht: Untersuchung von Wirksamkeit und Korrelaten. Free University Berlin. Paetsch, J. (2006). Mentale Zahlenrepräsentation als Determinante kindlicher Diagrammkompetenz. Diffendaffer, B. (2005). Die Entwicklung grafischen Technical University Berlin. Verständnisses bei Grundschulkindern: Einfluss der Aufgabenform auf den Lernzuwachs. Technical Pereira Kastens, C. (2005). Lernen zu Hause vs. im University Berlin. Unterricht: Wie situationsspezifisch sind das Erleben, die Motivation und das Verhalten von Schülerinnen Grube, A. (2005). Dynamics of the motivational und Schülern? Technical University Braunschweig. system: The role of age, performance evaluation, self- efficacy, and achievement motivation. Free University Pohlmeyer, A. (2006). Adult age differences in intra- Berlin. individual variability of spatio-temporal gait parameters are shaped by working memory load. Humboldt Herbrich, I. (2006). Positiver und negativer Affekt im University Berlin. hohen Alter – ein längsschnittlicher Ansatz. Technical University Berlin Sameiske, S. (2006). Lehrerbeanspruchung und schulischer Unterricht: Zur Rolle von Unterrichts- Ibrahimbegovic, S. (2006). Lebensziele junger überzeugungen und Merkmalen der Unterrichts- Erwachsener aus personenorientierter Sicht: Eine gestaltung. Free University Berlin. Anwendung der Mischverteilungsverfahren. Technical University Darmstadt. Schalk, L. (2006). Ist die Giraffe ein Mama-Tier? Grammatikalisches Geschlecht und Induktives Kampa, N. N. (2006). Eine Studie zum Zusammen- Schließen bei Kindern. Technical University Berlin. hang zwischen sozioökonomischer Herkunft von Lehrkräften und deren Berufslaufbahn sowie berufs- Wieckhorst, B. (2006). Concurrent cognitive and sen- bezogenen Überzeugungen. Free University Berlin. sorimotor performance in children and young adults: Does working memory benefit from simultaneous Kropf, M. (2005). Hausaufgabenverhalten und walking? Free University Berlin. Hausaufgabenmotivation von Schülerinnen und Schülern: Die Rolle der Eltern. University of Erfurt. Wiest, M. (2006). Sehnsucht – ein typisch deutsches Phänomen? Ein interkultureller Vergleich zwischen Lörz, M. (2005). Primäre und sekundäre Effekte bei Deutschland und den USA. Free University Berlin. tertiären Bildungsentscheidungen nach Geschlecht und sozialer Herkunft. University of Mannheim.

Mayser, S. (2005). (Un)Erreichbar fern: Ein Vergleich persönlicher Ziele und Sehnsüchte. Humboldt Univer- sity Berlin.

Appendix | 219 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 220

5. Scientific and Professional Staff 2005–2006

Artelt, Cordula (Dr. phil. in Psychology, 1999, Univer- Czienskowski, Uwe (Dr. phil. in Psychology, 1995, Free sity of Potsdam; Habilitation, 2005, University of University Berlin; Scientific Programmer): Scientific Potsdam): Research in learning (learning strategies software development (resource distribution in social and metacognition); reading; cognitive development networks; feature-pattern-analysis; computer-based (as of 2005: University of Bamberg). scientific research); modeling and simulation; self- reference effect; meta-analysis; experimental design Assmann, Wolfgang (Head of Information Processing and analysis; philosophy of psychology (mind & con- Center): Service management in research institutions; sciousness). information technology in the social and behavioral sciences. Delius, Julia A. M. (Dr. med. in Medicine, 1993, Uni- versity of Frankfurt a.M.): Interdisciplinary gerontol- Baltes, Paul B. (Dr. phil. in Psychology, 1967, Saarland ogy in the context of the Berlin Aging Study (BASE); University; Fellow of the Max Planck Society; Co- BASE Website design and management; editorial director of the Institute; as of July 2004: Director, projects. MaxnetAging; Professor of Psychology, Free University Berlin; as of November 2004: Distinguished Professor Flitner, Ursula (M.A. in American Studies and German of Psychology, and Permanent Fellow at the Shannon Literature, 1991, Free University Berlin; State Exami- Center for Advanced Study, University of Virginia nation in Library and Information Science, 1995, [part-time]): Lifespan human development: Evolution Senatsverwaltung für Kulturelle Angelegenheiten and ontogenesis; aging of the mind (intelligence, Berlin/Cologne; Head of the Library and Information memory, personality, wisdom); theory of successful Research Unit): Information management; electronic development; science policy: interdisciplinarity, his- resources and networked information systems; human tory, and internationality (deceased 7 November resources development. 2006). Gigerenzer, Gerd (Dr. phil. in Psychology, 1977, Uni- Baumert, Jürgen (State Examination for Teachers, versity of Munich; Habilitation in Psychology, 1982, 1968, Hamburg; Dr. phil., 1968, University of Tübin- University of Munich; Fellow of the Max Planck gen; Habilitation in Educational Sciences, 1982, Free Society; Co-director of the Institute; Professor of Psy- University Berlin; Fellow of the Max Planck Society; chology, Free University Berlin): Models of bounded Co-director of the Institute; Professor of Educational rationality; social intelligence; ecological rationality; Sciences, Free University Berlin and Humboldt Univer- heuristics of scientific discovery; philosophy, history, sity Berlin; Vice President of the Max Planck Society): and methodology of social sciences. Research in teaching and learning; cultural comparisons, large-scale assessment, cognitive and motiva- Grabbe, Yvonne (Diploma in Psychology, 2001, Uni- tional development in adolescence. versity of Münster; Dr. phil. in Psychology, 2003, Uni- versity of Münster): Research in instruction and learn- Baumgarten, Jürgen (Dr. phil. in German Language ing; teacher-based assessment of student achieve- and Literature, 1973, Free University Berlin; Head of ment; effective pre-school education and educational Editorial and Public Relations Department): Prehistory outcomes; job demands and stress of teachers. of the Middle East; neolithization; nomadic cultures. Hardy, Ilonca (PhD in Educational Psychology, 1998, Berg, Nathan (B.A. in American Studies, 1995, Uni- University of Iowa): Learning environments incorporat- versity of Kansas; M.A. in Mathematics, 2001, Univer- ing the social character of cognition (emphasis: col- sity of Kansas; PhD in Economics, 2001, University of laborative learning); the role of language in problem Kansa): Behavioral economics; demographic econom- solving; effects and uses of external representations. ics; less-is-more effects in decision making and estimation; social heuristics. Heekeren, Hauke (PhD, 2000, Humboldt University Berlin; License for the practice of medicine, 2002; Brighton, Henry (PhD Cognitive Science, 2003, The Head of Independent Junior Research Group): Neuro- University of Edinburgh, UK; M.Sc. Cognitive Science, biology of perceptual and reward-based decision 1997, The University of Edinburgh, UK; B.Sc. (Hons) making; influence of cognition and emotion on deci- Artificial Intelligence and Computer Science, 1996, sion making in social contexts; multimodal neuro- The University of Edinburgh, UK): Cognition-inspired imaging. approaches to artificial intelligence and machine learning; inductive inference, information theory, and Hutchinson, John M. C. (DPhil in Biology, 1990, Uni- MDL; ecological rationality and adaptive decision versity of York): Behavioral ecology and optimality making; computational evolutionary linguistics and modeling; rules of thumb in animals and plants (e.g., cultural computation. weather prediction); spatial decisions in biology and the social sciences; daily routines: the dawn chorus of Brunner, Martin (Diploma in Psychology, 2002, Univer- birds; skylark behavior and morphology; sexual selec- sity of Mannheim; Dr. phil. in Educational Science, tion of slug genitalia; theoretical morphology. 2006, Humboldt University Berlin): Research in instruction and learning; structural equation modeling and Katsikopoulos, Konstantinos V. (PhD in Industrial En- item response theory; mathematics achievement; cog- gineering, 1999, University of Massachusetts, Amherst): nitive abilities (as of 2006: University of Luxemburg). Theory: Development, analysis, and testing of simple

220 | Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 221

models of bounded rationality; applications: Decision Maaz, Kai (Diploma in Education [FH], 1998, Catholic analysis, medical decision making, transportation. University of Applied Sciences Berlin; Diploma in Social Sciences, 2002, Humboldt University Berlin; Keller, Monika (Dr. phil. in Psychology, 1974, University Dr. phil. in Educational Science, 2006, Free University of Heidelberg; Habilitation in Psychology, 1996, Free Berlin): Transition from school to university and work; University Berlin, Professor of Psychology, Free Univer- social background and educational and vocational sity Berlin): Understanding of relationships, moral de- chances; social inequality and educational decisions. velopment and moral emotions in cultural context; social rationality: Interconnecting moral psychology with Mayer, Karl Ulrich (Dr. rer. soc., 1973, University of behavioral game theory; social perspective taking, the- Konstanz; Habilitation in Sociology, 1977, University ory of mind, and domains of social and moral reason- of Mannheim; Fellow of the Max Planck Society; Co- ing; social and moral copetence in education. director of the Institute; Professor of Sociology, Free University Berlin): Social stratification and mobility; Krauss, Stefan (First State Examination in Mathe- comparative analysis of social structure; sociology of matics and Physics, 1995, University of Erlangen- the life course; occupational structures and labor Nuremberg; Dr. phil. in Psychology, 2001, Free Univer- market processes (as of July 2005: Yale University). sity Berlin): Research on teaching and learning; didactics of mathematics, especially didactics of statistics; McElvany, Nele (Diploma in Psychology, 2001, Free probabilistic reasoning (Bayesian inferences). University Berlin; Dr. phil. in Psychology, 2006, Free University Berlin): Research in instruction and learn- Kruse, Imke (Dr. phil. in Political Science, 2005, Free ing; reading literacy; self-regulated learning; social- University Berlin): International relations; institutional ization in the family. theory; asylum and migration policies; aging policy for elderly migrants in Europe. Müller, Viktor (Dr. rer. soc., 1996, Eberhard Karls Uni- versity Tübingen): Lifespan psychology and aging Kunter, Mareike (Diploma in Psychology, 1999, Julius mechanisms; psychophysiology of social interactions; Maximilians University Würzburg; Dr. phil. in Psychol- complexity and brain dynamics; cortical synchroniza- ogy, 2004, Free University Berlin): Research in instruction: local and global networks. tion and learning; teacher research; multiple educational objectives; motivational processes in the class- Nagy, Gabriel (Diploma in Psychology, 2002, Free room; assessment of instructional processes. University Berlin; Dr. phil. in Psychology, 2006, Free University Berlin): Research in teaching and learning; Li, Shu-Chen (PhD in Psychology, 1994, University of developmental regulation across the life span; transi- Oklahoma; Habilitation in Psychology, 2006, Free Uni- tion from school to vocational training; quantitative versity Berlin): Behavioral and neurocognitive devel- methods of empirical social research. opment across the lifespan; neuromodulation of cognition; neurocomputational and formal modeling of von Oertzen, Timo (Diploma in Computer Science, lifespan cognition; ontogeny of Brain-Body-World 1999, Saarland University; PhD in Computer Science, interactive dynamics; neuroeconomics and aging; bio- 2003, Saarland University): Mathematical psychology; cultural co-construction of development. algorithms in psychology; statistical modeling; formal modeling. Lindenberger, Ulman (Dr. phil. in Psychology, 1990, Free University Berlin; Habilitation in Psychology, Oesterreich, Detlef (Diploma in Psychology, 1968, 1998, Free University Berlin; Fellow of the Max Planck Free University Berlin; Dr. phil. in Psychology, 1975, Society; Co-director of the Institute; Professor of Psy- Free University Berlin): Theory of authoritarianism, chology, Saarland University; Professor of Psychology, authoritarianism and political consciousness; civic Free University Berlin; Professor of Psychology, Hum- education of adolescents (November 2006 retired). boldt University Berlin): Lifespan psychology: Theories and methods; behavioral plasticity and its neural cor- Olsson, Henrik (PhD in Psychology, 2000, Uppsala relates in childhood and old age; sensorimotor and University, Uppsala, Sweden): Computational model- cognitive development; multivariate measurement of ing; judgment and decision making; categorization, change and variability. estimation, and causal learning; adaptiveness of cognitive mechanisms to environmental structures; work- Lövdén, Martin (B.A. in Psychology, 1998, Lund Uni- ing memory capacity. versity, Sweden; PhD in Psychology, 2002, Stockholm University, Sweden): Lifespan development; cognitive Riediger, Michaela (Diploma in Psychology, 1997, neuroscience of aging; behavioral and neural plastic- Humboldt University Berlin; Dr. phil. 2001, Free Univer- ity; cognitive control; episodic memory; spatial navi- sity Berlin): Lifespan changes in the interplay of motiva- gation; sensorimotor-cognition couplings; methods tion, affect, and social cognition; future-orientation, for studying individual change (as of 2006: Karolinska motivation, and volition; emotion and emotion regula- Institute, Stockholm). tion; social aspects of motivational and affective processes. Lüdtke, Oliver (Diploma in Psychology, 2000, Free University Berlin; Dr. phil. in Psychology, 2004, Free Rieskamp, Jörg (Diploma in Psychology, 1998, Tech- University Berlin): Research in teaching and learning; nical University Berlin; Dr. phil. in Psychology, 2001, quantitative methods in empirical research; personal Free University Berlin): Cognitive modeling of judg- goals. ment and decision making; the role of learning in

Appendix | 221 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 222

decision making; experimental examinations and evo- Stern, Elsbeth (Diploma in Psychology, 1982, Univer- lutionary simulations of simple strategies for social sity of Hamburg; Dr. phil. in Psychology, 1986, Univer- interactions. sity of Hamburg; Habilitation in Psychology, 1994, University of Munich; University Professor, 1994, Schmiedek, Florian (Diploma in Psychology, 2000, University of Leipzig): Cognitive development; University of Mannheim; Dr. phil. in Psychology, 2003, intelligence and knowledge; research in teaching and Free University Berlin): Cognitive lifespan psychology; learning (as of 2006: Swiss Federal Institute of intraindividual variability; model-based analyses of Technology [ETH], Zurich). reaction time distributions; multivariate modeling of cognitive developmental processes; typical intellectual Stevens, Jeffrey (PhD in Ecology, Evolution, and Be- engagement; structural equation modeling, latent havior, 2002, University of Minnesota): Cognitive con- growth modeling, time series analysis; item response straints on cooperative behavior; evolution of decision theory (as of 2006: Humboldt University Berlin). making (particularly temporal discounting); comparative aspects of decision making in humans and non- Schooler, Lael, J. (PhD in Cognitive Psychology, 1993, human animals; the role of ecological and social con- Carnegie Mellon University): Adaptation of human text on choice. memory to the statistical structure of past and present environments; computational models of human mem- Todd, Peter M. (PhD in Psychology, 1992, Stanford ory; memory's role in judgment and prediction tasks. University): Evolution of behavior (including computer simulations of populations of simple organisms Schümer, Gundel (First and Second State Examina- adapting to and creating different environmental tion for Teachers, 1962, Stuttgart; 1970, Tübingen; structures, both physical and social); ecological Dr. phil. in Educational Sciences, 1977, University of rationality (characterizing environment structure- Hamburg): School systems and instruction; theories of decision mechanism interactions); simple heuristics instruction and methods of instructional research; for sequential search (including mate choice, job comparative research on schools and instruction; the search, foraging among patches, looking for cues, and language of the classroom. (June 2006 retired). seeking parking places) and other types of decisions; mechanisms for food choice (and other aspects of Smith, Jacqui (PhD in Psychology, 1984, Macquarie consumer behavior); rhythmic and time-based behav- University, Sydney; Habilitation in Psychology, 1999, ior (including music, sequence learning/production, Free University Berlin; Honorary Professor of Psychol- and evolution of song) (as of 2005: Indiana University, ogy, 2003, Free University Berlin; Deputy Director, School of Informatics). MaxnetAging; Senior Research Scientist): Profiles of psychological functioning in the old and oldest-old; Trautwein, Ulrich (Diploma in Psychology, 1999, Uni- psychological predictors of longevity; potentials and versity of Göttingen; Dr. phil. in Psychology, 2002, risks for development and successful aging; applica- Free University Berlin; Habilitation in Psychology, tion of intelligence and life knowledge during adult- 2005, Free University Berlin): Development of self- hood (as of 2006: University of Michigan, Ann Arbor). related cognitions in educational settings; school development and management; effects of homework Stanat, Petra (Diploma in Psychology, 1992, Free Uni- assignment on academic achievement. versity Berlin; PhD in Psychology, 1998, University of Massachusetts at Amherst; Habilitation in Educa- Watermann, Rainer (Diploma in Education, 1996, tional Science, 2005, Free University Berlin): Condi- University of Munster; Dr. phil. in Educational Science, tions of immigrant students' school success; interna- 2002, Free University Berlin; Habilitation in Educa- tional and comparative educational research; gender tional Science, 2005, Free University Berlin): Inter- differences in school performance; social competence national and comparative educational research; (as of 2005: University of Erlangen-Nuremberg). quantitative methods of social research; political socialization (as of 2005: University of Göttingen).

Emeritus Members of the Max Planck Society

Edelstein, Wolfgang (Dr. phil. in Medieval Studies, Roeder, Peter M. (Dr. phil., 1960, University of 1962, University of Heidelberg; Fellow of the Max Marburg; Habilitation in Educational Science, 1966, Planck Society; until 1997 Co-director of the Insti- University of Marburg; Fellow of the Max Planck tute; Honorary Doctorate in Social Science, University Society; until 1995 Co-director of the Institute; Spe- of Iceland; Honorary Professor of Educational Science, cial Professor of Educational Science, Free University Free University Berlin and University of Potsdam): Berlin): Educational science; school research; history Development and socialization; social-cognitive and of educational science. moral development; developmental and structural aspects of curriculum and instruction; developmental and school-related conditions of successful learning; conditions of successful school transformation.

222 | Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 223

Postdoctoral Research Fellows

Bauer, Markus (Diploma in Psychology, 2002, Hum- tion integration and advice-taking behaviors; group boldt University Berlin, PhD research, F. C. Donders decision processes; simple heuristics for individual Centre for Cognitive Neuroimaging): Neuronal mech- and group decision making (as of 2006: Singapore anisms of perceptual decision processes; functional Management University). relevance of sychronized oscillatory neural activity; methods of human electrophysiology. Mata, Rui (Diploma in Psychology, 2002, FPCE Uni- versity of Lisbon; Dr. phil. 2006, Free University Biele, Guido (Diploma in Psychology, 1999, Free Uni- Berlin): Adaptation of human memory to the statisti- versity Berlin, Dr. phil in Psychology, 2006, Free Uni- cal structure of the environment; social cognition versity Berlin): Reward based decision making; com- (the discipline, not ToM) (LIFE) (as of 2006: University putational modeling; neural correlate of reinforce- of Michigan at Ann Arbor). ment learning models. Pachur, Thorsten (Diploma in Psychology, 2002, Free Chicherio, Christian (PhD in Psychology, 2006, Uni- University Berlin; M.Sc. in Health Psychology, 2002, versity of Geneva): Developmental and differential Sussex; Dr. phil. 2006, Free University Berlin): Role of approaches to the study of cognitive aging; working recognition in decision making; sampling based sim- memory, executive functioning and attentional ple heuristics; ecological rationality; process models resources across the lifespan; individual differences of decision under uncertainty; risk perception and and variability; gene-behavior interactions; age- communication (as of 2006: University of Basel). related changes in brain functioning (plasticity/ reorganization) using SPECT/PET, EEG, fMRI; connec- Persson, Magnus (PhD in Psychology, 2004, Uppsala tivity and dynamics in brain functioning; linking be- University): Decision making under uncertainty; havioral performance to cerebral signal; multivariate bounded rationality; exemplar models; computer go. methodology. Qian, Jing (PhD in Cognitive Psychology, 2006, Uni- Conlin, Juliet (PhD in Psychology, 2006, University of versity of Warwick): Economic psychophysics: How do Durham): Working memory and ‘executive’ processes; sampling, loss aversion, and the distribution of con- decision making across the lifespan; navigational textual information affect our judgement of economic heuristics. quantities; GEMS model: A generalized exemplar model of sampling; statistical structure of the world Dziobek, Isabel (Diploma in Psychology, 2000, and the distributions of social-economic quantities: Ruhr University Bochum, PhD, 2006, University Power-laws and the pareto law; ecological rationality: of Bielefeld): Social cognitive neuroscience; autism An adaptive sampling approach; social rationality and spectrum conditions; neuroanatomy. models of social comparison; the origins of the probability weighting function; price perception and pay Ebner, Natalie C. (Diploma in Psychology, 2001, satisfaction. Free University Berlin; Dr. phil. in Psychology, 2005, Free University Berlin): Multi-method approach to Reimer, Torsten (Dr. phil. in Psychology, 1996, Free processes of developmental regulation; lifespan University Berlin): Cognitive aspects of group decision changes in the interplay of motivation, affect, and making and problem solving (as of 2006: University of (social) cognition; approach and avoidance motivation Maryland). across the lifespan; facial expressions across the lifespan. Schäfer, Sabine (Diploma in Psychology, 2001, Free University Berlin; Dr. phil., 2005, Free University Galesic, Mirta (PhD in Psychology, 2004, University of Berlin): Cognitive-sensorimotor coordination across Zagreb; M.Sc. in Survey Methodology, 2005, Joint the life span; age differences in the regulation of Program in Survey Methodology, University of Mary- sequential action; spatial navigation; behavioral and land & University of Michigan): Risk communication; neural plasticity; ontogenetic changes in behavior sampling approaches to cognition; judgment and regulation. decision making models; cognitive aspects of survey response. Scheibe, Susanne (Diploma in Psychology, 2001, Humboldt University Berlin; Dr. phil., 2005, Free Uni- Garcia-Retamero, Rocio (as of 2006: University of versity Berlin): Strategies of life management and Granada): Causal learning; compound cue processing; successful development across the lifespan; develop- leadership and gender; social learning and imitation; mental psychology of Sehnsucht (life-longings); emo- threat perception. tions and emotion regulation across the lifespan; the role of interindividual differences in psychopathology; Kopp, Franziska (Diploma in Psychology, 1999, Uni- structural equation modeling and questionnaire versity of Leipzig; Dr. rer. nat. in Psychology, 2006, development. University of Leipzig): Cognitive and social development during infancy; synchronized brain activity dur- Schooler, Julia (PhD in Psychology, 1996, University ing cognition in infants and adults. of Pittsburgh): Cues determining portion size; memory for enjoyed food; understanding and remembering Luan, Shenghua (PhD in Psychology, 2004, University health claims on packages. of Florida): Applied signal detection theory; informa-

Appendix | 223 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 224

Takezawa, Masanori (M.A. in Social Psychology, Wegwarth, Odette (Diploma in Psychology, 2003, 1997, Hokkaido University, Japan; PhD in Social Psy- University of Potsdam; Dr. rer. nat. in Psychology, 2007, chology, 2001, Hokkaido University, Japan): Adaptive Humboldt University Berlin): Medical decision-mak- heuristic approach in distributive bargaining under ing; choice modeling; risk communication and under- incomplete information; development and emergence standing of risk; phenomenon of too much choice. of social preferences—altruism, reciprocity, egalitar- ianism and moral punishment; social intelligence in Wilke, Andreas (Diploma in Psychology, 2002, Free cooperative/ competitive situations; human experi- University Berlin; Dr. phil., 2006, Free University ments and computer simulations (as of 2006: Tilburg Berlin): Emotions; evolutionary psychology; judgment University). and decision making (LIFE) (as of 2006: University of California, Los Angeles).

Predoctoral Research Fellows

Alves Conte de Morais, Ana Sofia (Licenciatura in Dubberke, Thamar (Diploma in Psychology, 2006, Psychology, 2004, University of Lisbon): Memory- Philipps University Marburg): Research in teaching based decision making; cognitive aging and the use of and learning. decision strategies; the role of attention in learning. (LIFE) Gaissmaier, Wolfgang (Diploma in Psychology, 2002, Free University Berlin; Dr. phil., 2007, Free University Bahnemann, Markus (Cand. med., University of Berlin): Memory-based decision making; modeling Witten/Herdecke): Decision making in social contexts. and predicting voting behavior; simple heuristics underlying intuitions; medical decision making. Bartling, Karen (Drs. in Psychology, 2006, University of Maastricht): Social development in early childhood Green, Nikos (B.Sc. in Cognitive Science, 2004, Uni- (especially mother-child interaction); neural corre- versity of Osnabrück; M.Sc. in Cognitive Science, lates of early interaction; EEG methods. (LIFE) Track: Brain, Behavior and Cognition, 2006, University of Amsterdam): Cognitive science; cognitive neuro- Barton, Adrien: Philosophical theories of probability; science; computational and cognitive modeling; connections with psychological issues; philosophy of developmental neuroscience; decision-making. quantum mechanics; structure of scientific images. Gresch (Hausen), Cornelia (Diploma in Social Sci- Becker, Michael (Diploma in Psychology, 2004, Free ences, 2004, University of Mannheim): Social origin University Berlin): Cognitive development; changes in and educational career decisions; quantitative meth- educational systems; quantitative methods in social ods of empirical social research; immigrants. (LIFE) sciences. (LIFE) Grühn, Daniel (Diploma in Psychology, 2002, Free Brehmer, Yvonne (Diploma in Psychology, 2003, University Berlin): Emotion and emotion regulation Saarland University; Dr. phil. in Psychology, 2006, across the lifespan; cognitive functioning in the old Humboldt University Berlin): Dynamic and plasticity and the oldest-old; cognitive and emotional mecha- of cognitive development and aging; computational nisms of successful aging (as of 2006: University of modeling of cognitive processes; neural correlates of Geneva). age-related changes (as of 2006: Karolinska Institute Stockholm). Gundert, Stefanie (Diploma in Sociology, 2003, Uni- versity Duisburg-Essen): Life-course and labor-market Brose, Annette (Diploma in Psychology, 2006, Free research; temporary employment and fixed-term con- University Berlin): The interaction of emotion, situa- tracts; women’s employment (as of 2006: Yale Univer- tional influences (events) and cognition within and sity). across individuals across the lifespan; emotion regulation and its developmental trajectories; the study of Hachfeld, Axinja (Diploma in Psychology, 2006, Free intraindividual variability and its relation to develop- Univerity Berlin): Research in instruction and learning; mental change. (LIFE) reading competence.

Burzynska, Agnieszka Zofia (B.Sc. in Biotechnology, Hämmerer, Dorothea (Diploma in Psychology, 2005, 2005, Universities of Gdansk and Perugia, M.Sc. in University of Freiburg): Cognitive Models of Neuro- Neuroscience, 2007, University of Göttingen): White modulation. matter integrity and connectivity in aging brain (diffusion MRI); correlations with cognitive performance von Helversen, Bettina (Diploma in Psychology, 2004, and genetic background. (LIFE) University of Erlangen-Nuremberg): Categorization and estimation strategies; ontogenetic development Diao, Linan (Diploma in Management, 2002, Jilin of heuristic strategies; adaptiveness of basic cognitive University, P.R. China): Agent-based computational capacities to environmental structures; legal decision economics; simulation in financial markets. making. (LIFE)

224 | Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 225

Husemann, Nicole (Diploma in Psychology, 2004, Nagel, Irene (Diploma in Psychology, 2005, University of Bielefeld): Research in teaching and Maastricht University): Age-related changes in human education; personal goals; academic cheating. cognition and behavior; age-related changes in the human brain (structure, functional activation, Huxhold, Oliver (Diploma in Psychology, 2002, Free dopamine system); using fMRI (functional magnetic University Berlin; Dr. phil., 2007, Free University resonance imaging) to examine the interplay of in- Berlin): Lifespan cognitive development and cognitive creasing age, genetic predisposition, brain structure, aging; intra-individual and inter-individual variability and task-specific brain activation patterns, and their in cognitive performance; cognitive processes con- relation to cognitive decline in elderly. (LIFE) tributing to balance control. Neumann, Marko (M.A. in Educational Sciences, Jonkmann, Kathrin (Diploma in Psychology, 2005, 2006, Humboldt University Berlin): National and Humboldt University Berlin): Social dynamics in the international school achievement research; impact of classroom; school achievement research; develop- institutional opportunity structures on the developmental regulation across the life span. (LIFE) ment of scholastic abilities; school effectiveness research; quality assurance and quality improvement Johnson, Tim (BA, 2004, R.D. Clark Honors College, in the educational system. University of Oregon): Experimental economics; cooperation; political science methodology; voting Pieper, Swantje (Diploma in Psychology, 2006, Uni- behavior; institutions. versity of Münster): National and international school achievement research; quality assurence and quality Kleinspehn, Anna (Diploma in Psychology, 2004, Free improvement in the educational system; effects of University Berlin): Development of emotional and homework assignment on acedemic achievement). self-related processes across the lifespan; age differences in cooperative behavior; individual differences Rauers, Antje (Diploma in Psychology, 2004, Free and context covariates underlying interpersonal inter- University Berlin): SOC in social and cognitive lifespan actions; subjective experience of own aging. development; age trajectories of collaborative cognition; intrapersonal and interpersonal knowledge Klusmann, Uta (Diploma in Psychology, 2004, Free organization. University Berlin): Research in teaching and learning; teacher stress; personal goals. Richter, Dirk (M.A. in Quantitave Research, Evaluation and Measurement in Education, 2005, The Ohio State Kotter-Grühn, Dana (Diploma in Psychology, 2004, University; M.A. in Education, 2006, Technical Univer- Technical University Dresden): Developmental psy- sity Dresden): Teacher education; teacher testing; chology of life-longing (Sehnsucht); self-regulation development of novice teachers; statistical modeling. and subjective well-being over the lifespan; self- perceptions of aging; personality and social relations Röcke, Christina (Diploma in Psychology, 2002, Free in middle and old adulthood. (LIFE) University Berlin; Dr. phil. in Psychology, 2006, Free University Berlin): Within-person coupling of well- Krugel, Lea (Cand. med., Charité Universitätsmedizin being and cognitive performance across the lifespan; Berlin): Neurobiology of reward-based decision mak- intraindividual variability as a tool to understand ing, neuroimaging. lifespan development; emotion and emotion regulation across the lifespan; social relationships and Marewski, Julian (Diploma in Psychology, 2005, Free closeness across the lifespan (as of 2006: Brandeis University Berlin): The use and ecological rationality University Boston). of recognition and ease of retrieval in decision making; adaptive memory; process models of decision Scheibehenne, Benjamin (Diploma in Psychology, making under uncertainty; eye-movements and deci- 2004, Humboldt University Berlin): Cognitive model- sion making under uncertainty; ecological rationality ing of preferentical choice; effects of having too in political environments. much choice; food choice.

Mata, Jutta (Diploma in Psychology, 2004, Humboldt Schellenbach, Michael (Diploma in Computer University Berlin): Food choice (decision rules, devel- Science, 2004, Saarland University): Spatial naviga- opmental aspects); preference prediction; develop- tion; pedestrian navigation systems; virtual reality. ment of prosocial behavior (behavioral game experiments). (LIFE) Schneider, Michael (Diploma in Psychology, 2002, Technical University Berlin; Dr. phil., 2006, Technical Mohr, Peter (M.Sc. in Business Administration, 2006, University Berlin): Cognitive learning theories; concep- University of Münster): Economic decision-making; tual and procedural knowledge; diagrams as cognitive risk perception. (LIFE) tools (as of 2006: Swiss Federal Institute [ETH] Zurich).

Müller, Andrea G. (M.A. in Sociology, 2000, Univer- Shing, Yee Lee (M.A. in Educational Psychology, sity of Iowa; M.A. in Educational Science/Sociology, 2003, University of Georgia, Athens): Lifespan and 2002, Martin Luther University of Halle-Wittenberg): developmental psychology theories; the development Conditions of immigrant students' school success; and plasticity of basic cognitive mechanics and intel- bilingual education and second language acquisition; lectual functioning over the lifespan; multivariate everyday school-related communication skills (LIFE). analyses of dynamical developmental and change

Appendix | 225 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 226

processes; neural correlates of cognitive developmen- Wenzlaff, Hermine (Diploma in Biology, 2005, Uni- tal and aging processes. (LIFE) versity of Tübingen): Neurobiology of perceptual decision making; electrophysiological recordings (EEG, Straubinger, Nils (Diploma in Psychology, 2006, Uni- MEG); cognitive modelling; correlation of neurophysi- versity of Münster): Thinking, problem solving, judg- ological components with modelled parameters. ment and decision making; lifespan cognitive development (focus on probabilistic thinking); the role of Werkle-Bergner, Markus (Diploma in Psychology, external representations (e.g. diagrams) in probabilis- 2004, Saarland University): Lifespan development of tic thinking. (LIFE) memory and cognitive control functions; neuronal correlates of lifespan plasticity and change; EEG Tsai, Yi-Miau (M.Sc. in Psychology, 2003, National methods in lifespan research; multivariate statistical Taiwan University, Taiwan): Research in learning and models of variability and change. (LIFE) instruction; achievement motivation and goals; context and cultural effects on motivation. (LIFE)

Verrel, Julius (D.E.A. in Mathematics, 2001, Jussieu, Paris; M.Sc. in Cognitive Neuroscience, 2006, Radboud University, Nijmegen): Motor development and aging; interaction of sensorimotor and cognitive skills. (LIFE)

226 | Appendix 11_Appendix_208_227.qxd 18.07.2007 14:05 Uhr Seite 227

Appendix | 227 Umschlag_2007_Ora_070717.fhmx 18.07.2007 14:06 Uhr Seite 1

C M Y CM MY CY CMY K

Probedruck