Intelligence and the Cognitive Unconscious

P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

CHAPTER 22 Intelligence and the Cognitive Unconscious

Scott Barry Kaufman

The definition of genius is that it acts that have explicit instructions and require unconsciously; and those who have effortful concentration is administered to a produced immortal works, have done so diverse group of people, all of the tests tend without knowing how or why. The greatest to be positively correlated with one another, power operates unseen, and executes its a finding often referred to as a “positive man- appointed task with as little ostentation as ifold.” Spearman labeled the factor on which difficulty. all individual tests loaded g, for general intelligence. – William Hazlitt1 Over the past 100 years, the existence of g as a statistical phenomenon is one of the Intelligence tests were originally created most replicable findings in all of psychology with the practical goal of identifying stu- (Carroll, 1993; Chabris, 2007; Jensen, 1998). dents in need of alternative education Nonetheless, there is still work to be done to 1916 (Binet & Simon, ). Because intelligence determine what explains the positive man- tests were originally devised to predict ifold (see Maas et al., 2006), the cognitive school grades, the items were intention- mechanisms that support g (see Chapter 20, ally designed to measure a general ability Working Memory and Intelligence, this vol- to profit from explicit instruction, concen- ume; Kaufman, DeYoung, Gray, Brown, & trate on a task, and engage in intellectual Mackintosh, 2009; Sternberg & Pretz, 2005), material. Indeed, research shows that such and whether there are other forms of cog- a general ability does seem to exist. Over nition that display meaningful individual 1904 a century ago, Spearman ( ) discovered differences and predict intelligent behavior that when a wide range of cognitive tests above and beyond g and the cognitive mechanisms that support g. 1 William Hazlitt (1846), “Essay IV. Whether Genius This chapter presents evidence that Is Conscious of Its Powers?” in Table Talk: Opinions on Books, Men, and Things, Second Series, Part I mechanisms relating to the cognitive uncon- (pp. 37–49). New York, NY: Wiley & Putnam. scious – “mental structures, processes, and

442 P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 443

states2 that can influence experience, Integrating Two Research Traditions thought, and actions outside phenomenal awareness and voluntary control” (Dorf- The 20th century witnessed at least two man, Shames, & Kihlstrom, 1996,p.259) major paradigm shifts within psychological also make an important contribution to science. One major shift was from behav- intelligent behavior. Although intelligence iorism to the “cognitive revolution,” which testers have done a remarkable job devel- brought along with it a shift in focus from oping tests that measure individual dif- learning and conditioning toward investigat- ferences in explicit, controlled cognitive ing the mental processes involved in con- processes, the investigation of individual dif- scious thought, including memory, think- ferences in implicit, nonconscious processes ing, and problem solving (Miller, 2003). This has not received nearly as much attention shift has had an enduring effect on concep- (Kaufman, 2009a, b). tualizations of human intelligence as well Furthermore, researchers have created as research methodology. Indeed, one of clever experiments to probe the nature the earliest investigators of the develop- of the cognitive unconscious by looking ment of intelligence in children was Jean at implicit memory, implicit perception, Piaget (1952), whose focus was on conscious and other forms of implicit cognition and higher order reasoning and how children thought3 (for reviews, see Kihlstrom, 1987, at different ages think. This emphasis on and Litman & Reber, 2005), but they age differences in thought as well as the have focused primarily on group-level data, notion that intelligence involves conscious, ignoring individual differences (see Cron- deliberate reasoning also underlies the logic bach, 1957). Additionally, some researchers behind the first widely administered intel- have downplayed the existence of continu- ligence test, the Binet-Simon Scale (Binet ous individual differences in the cognitive & Simon, 1916). Furthermore, the discovery unconscious that are meaningfully related that performances on diverse tests of explicit to important life outcomes (Reber, 1993; cognitive ability tend to correlate with one Stanovich, 2009). another – Spearman’s (1904) so-called pos- There have been some recent studies, itive manifold – further supported the idea however, that look at individual differences that intelligence tests are tapping into a “gen- in the cognitive unconscious. This chap- eral cognitive ability.” ter focuses on individual differences and Around the same time the shift from reviews recent empirical work on relations behaviorism to the cognitive revolution was among the cognitive processes underlying taking place, another dramatic shift in psy- psychometric intelligence and the cognitive chology was occurring. The conceptual- processes underlying the cognitive uncon- ization of the unconscious that was pre- scious, attempting to bridge two major dominant with psychodynamic theories of research programs that, until recently, have personality was slowly being transformed traveled on separate but parallel paths. into an unconscious recognized to serve many adaptive functions among both modern-day humans and our evolutionary 2 I include “implicit thought” in this definition as well, ancestry (Epstein, 1991; Hassin, Uleman, & although Kihlstrom tends to refer to “implicit cog- Bargh, 2005; Wilson, 2004). Over 30 years nition” differently from the “cognitive unconscious” (Dorfman, Shames, & Kihlstrom, 1996). of research in cognitive science reveals that 3 I assume in this chapter that intelligent “thought” a considerable amount of information pro- can operate either with or without awareness of cessing takes place on a daily basis automat- that thought. As Dorfman, Shames, and Kihlstrom (1996) astutely note, the idea of “implicit thought” ically – without our intent, awareness, and is a difficult concept because the notion of think- deliberate encoding – and plays an impor- ing has traditionally been equated with notions of 1890 tant role in structuring our skills, percep- consciousness. For instance, William James ( ) 1991 thought the notion of “unconscious thought” was a tions, and behavior (Epstein, ; Hassin contradiction in terms! et al., 2005; Kihlstrom, 1987; Lewicki & Hill, P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

444 SCOTT BARRY KAUFMAN

1987; Reber, 1993; Stadler & Frensch, 1997) Psychology that the key to consciousness is as well as facilitating problem solving and self-reference: creativity (Dijksterhuis & Nordgren, 2006; Dorfman, Shames, & Kihlstrom, 1996;Lit- In order for ongoing experience, thought, and action to become conscious, a link man & Reber, 2005). 1987 must be made between its mental repre- Kihlstrom ( ) distinguishes between sentation and some mental representation three types of nonconscious mental struc- of the self as agent or experiencer – as well, tures that together constitute the domain perhaps, as some representation of the envi- of the “cognitive unconscious.” Unconscious ronment in which these events take place. representations fit within the domain of pro- These episodic representations of the self cedural knowledge and are inaccessible to and context reside in working memory, but introspection under any circumstances. “By apparently the links in question are nei- virtue of routinization (or perhaps because ther automatic nor permanent, and must they are innate), such procedures operate be actively forged . . . without such linkages on declarative knowledge without either certain aspects of mental life are dissoci- ated from awareness, and are not accom- conscious intent or conscious awareness, panied by the experience of consciousness. in order to construct the person’s ongoing (Kihlstrom, 1987,p.1451) experience, thought, and action” (p. 1450; also see Anderson, 1982). Subliminal percep- A great deal of research has demonstrated tion, implicit memory, and implicit learn- the sophisticated and intelligent nature of ing fit the category of preconscious declar- the cognitive unconscious (Epstein, 2001; ative knowledge structures. In contrast to Lewicki, Hill, & Czyzewska, 1992; Loftus & unconscious representations, preconscious Klinger, 1992). For instance, after reviewing structures can be available to phenomenal the literature on the nonconscious acqui- awareness and can be introspected upon, but sition of information, Lewicki, Hill, and they can also influence ongoing experience, Czyzewska (1992) asked, “Is the noncon- thought, and action without ever entering scious information-processing system ‘intel- into working memory. Finally, Kihlstrom ligent’?” – to which they concluded: describes subconscious declarative knowl- The answer to the question about intelli- edge mental representations such as those gence would be affirmative if intelligence is activated during hypnosis, which can be understood as “equipped to efficiently pro- quite available to introspection but inacces- cess complex information.” In this sense, 4 sible to phenomenal awareness. our nonconscious information-processing Note that even though some noncon- system appears to be incomparably more scious representations have such high lev- able to process formally complex knowl- els of activation that they enter working edge structures, faster and “smarter” over- memory, they still might not meet the all than our ability to think and identify criteria of conscious awareness. As noted meanings of stimuli in a consciously con- by Kihlstrom, William James (1890) sug- trolled manner. (p. 801) gested over a century ago in his Principles of The idea that the unconscious can be smart is also illustrated by the title of a recent popular summary of the fast-and- 4 Note that only Kihlstrom’s (1987) notion of “uncon- frugal heuristics literature: Gut Feelings: The scious” mental structures meets all four of Bargh’s Intelligence of the Unconscious (Gigerenzer, (2004) horsemen of automaticity: lack of aware- 2007).5 Today there is a strong consensus ness, lack of intention, high efficiency, and inability to control. Kihlstrom’s notion of the preconscious lacks intention, but only under some circumstances 5 But note that Gigerenzer (2007; Gigerenzer & is efficient, lacks awareness, and can’t be controlled. Brighton, 2009), in contrast to those who view the Kihlstrom’s notion of the subconscious can be inten- cognitive unconscious as able to process complex tional and efficient, and even can be controlled, but information, views the cognitive unconscious as the key to defining the subconscious according to operating by the principle “less is more,” selecting Kihlstrom is the lack of phenomenal awareness. the right rule of thumb for the right situation. P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 445

among contemporary researchers in cog- Data indicate that as compared with con- nitive science, philosophy, cognitive psy- sciously controlled cognition, the noncon- chology, social psychology, reasoning, and scious information-acquisition processes morality that humans possess two quite dis- are not only much faster but are also tinct modes of thought – one controlled structurally more sophisticated, in that and the other more automatic (Epstein, they are capable of efficient processing of multidimensional and interactive relations 2003; Evans & Frankish, 2009; Stanovich & 2002 between variables. Those mechanisms of West, ). Indeed, dual-process theories non-conscious acquisition of information of cognition are becoming increasingly nec- provide a major channel for the develop- essary for explaining a wide variety of cogni- ment of procedural knowledge that is indis- tive, personality, social developmental, and pensable for such important aspects of cog- cross-cultural phenomena (Evans & Frank- nitive functioning as encoding and inter- ish, 2009). For instance, Klaczynski (2009) pretation of stimuli and the triggering of makes a case for adopting and develop- emotional reactions. (p. 796) ing a comprehensive dual-process theory of development, reviewing studies from such The advantages of type 1 processes can also diverse research topics as memory, judg- become disadvantages under certain circum- ments and decisions, reasoning, motivated stances. When thinking is dominated by reasoning, stereotypes, and magical reason- type 1 processes, task representations are ing to support his argument. highly contextualized. This contextualization can lead to the thoughtless application of judgment and decision heuristics. Dual-Process Theories of Cognition According to Stanovich and West (2000), Type 1 processes6 are thought to com- this mode of thought is in fact the “default” prise a set of autonomous subsystems mode in humans. They refer to this ten- (Stanovich, 2004) that include both innate dency toward automatic contextualization input modules (Fodor, 1983) and domain- of problems as the “fundamental computa- specific knowledge acquired by domain- tional bias” in human cognition (Stanovich general learning mechanisms that operate &West,2000). A similar idea can be found in automatically and efficiently (Reber, 1993). Chaiken’s (1987) heuristic systematic model Type 1 processes process information fast of persuasion, according to which people (relative to type 2 processes); are heavily are guided in part by a “principle of least influenced by context, biology, and past effort.” Because people have limited cog- experience; and aid humans in mapping and nitive resources, and because heuristic pro- assimilating newly acquired stimuli into pre- cessing is easy and adequate for most tasks, existing knowledge structures. heuristic processing from type 1 is generally An advantage of type 1 processes over used unless there is a special need to engage type 2 processes is that the former require lit- in systematic processing (see also Simon, tle conscious cognitive effort and free atten- 1979). In line with this idea, Klaczynski and tional resources for computationally com- Cottrell (2004) have argued that “metacog- plex reasoning. According to Lewicki, Hill, nitive intercession” often occurs, whereby and Czyzewska (1992), responses derived from intuition are available in working memory, where reflection is possible. However, according to Klaczyn- ski, most people do not take advantage 6 Many dual-process theorists refer to two “systems” of the opportunity to reflect on the con- (see Kahneman & Frederick, 2002). In recent years, however, critics of dual-system theorists have called tents of working memory, taking the con- for the use of a different name, arguing that “sys- tents from the experiential system as self- tem” carries with it a lot of conceptual baggage 2008 2009 evidently valid. Finally, the view of type (see Evans, ; Keren & Schul, ). In line with 1 Evans’s (2008) suggestion, I refer here to “types” of processes as the default mode of human thought processes instead of “systems.” cognition is also present in Haidt’s (2001) P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

446 SCOTT BARRY KAUFMAN

social intuitionist model of moral reasoning, modes of thought (e.g., abstract vs. contex- in which it is posited that intuitive process- tualized, associative vs. rule-based, shared ing is the default process, with deliberate with other animals vs. unique to humans) reasoning called upon only when intuitions are not as neat and clear-cut when one con- conﬂict with reason (see also Stanovich & siders that type 1 isn’t a unitary system, West, 2000). but includes a set of autonomous systems, In contrast, type 2 processes are typ- some of which are innately speciﬁed and ically characterized by deliberately con- some of which come about through learn- trolled, effortful, and intentional cognition. ing and practice (Stanovich, 2004;butsee Individual differences in this system have Epstein, 2010). Evans (2008) also points out been linked in the past to psychometric that “type 2” is most likely not a unitary intelligence (see Stanovich, 2009). Accord- system, suggesting that not all type 2 pro- ing to Stanovich and West (1997), a hallmark cesses are consciously controlled. Addition- of this type of thought is the ability to decon- ally, Cokely and Kelley (2009)andCokely, textualize task representations.7 Type 2 pro- Parpart, and Schooler (2009) have noted that cesses can deal with abstract content under even controlled processes may rely on auto- conditions of awareness8 and are not domi- matic processes for processing, even at the nated by the goal of attributing intentional- stage of early attentional selection. Other ity nor by the search for conversational rel- criticisms (see Aczel, 2009; Gigerenzer & evance (Margolis, 1987). It has been posited Regier, 1996; Keren & Schul, 2009) have been that type 2 processes are evolutionarily more leveled against dual-system models, a sign recent and uniquely developed in humans that the study of the dual-process nature than type 1 processes (Epstein, 2003; Evans, of the mind is an active area of research 2008; Gabora & Kaufman, 2009). and debate. In line with these criticisms, the Note that while some aspects are com- remainder of this chapter will refer to “dual- mon across most dual-process theories, process” theories instead of “dual-system” there are also distinct differences (Evans, theories and will assume that the various 2008). Most dual-process theorists agree processes are not completely independent on the automatic/controlled distinction but can interact with each other and facil- between the two modes of thought, as itate (or inhibit) each other in important well as the idea that type 2 processes are ways. constrained by a central working memory Indeed, in his review of dual-process system whereas type 1 processes are uncon- accounts of reasoning, judgment, and social strained by a central pool of resources. Dual- cognition, Evans (2008) notes two distinct process theorists differ, however, in terms kinds of dual-process theories. One kind, of other features they attribute to the two which he refers to as “parallel-competitive” modes of thought. For instance, some dual- forms of dual-process theory, states that process theorists emphasize the affective there are two forms of learning that lead nature of type 1 processes (Epstein, 1994; to two forms of knowledge (explicit and Metcalfe & Mischel, 1999; Zajonc, 1980), implicit) and each form competes for the whereas emotions are not a key component control of behavior. Evans refers to another of other models of implicit cognition (e.g., category of dual-process researchers as the Reber, 1993). “default-interventionists,” who assume that Also, as Evans (2008) rightly points out, rapid preconscious processes supply content some of the distinctions between the two for conscious processing and that the explicit system can intervene with the application 7 Although note that this system can also deal with of controlled processes. It should be noted contextualized content (see Cokely & Kelley, 2009; that not all dual-process theories fall neatly Cokely, Parpart, & Schooler, 2009). 8 into one category or the other. For instance, Although note that some researchers have argued 2003 that aspects of System 1 (e.g., implicit learning) can Epstein ( ) assumes that the two systems also deal with abstract material (see Reber, 1989). operate in parallel and are bi-directionally P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 447

interactive. As the implicit system has a strongly recruit basal ganglia functions, and faster reaction time it is more likely to initi- performance is thus improved on these tasks ate an action sequence. Nonetheless, Evans when the medial temporal lobe is removed (2008) does offer a useful classification of (Lieberman, 2007). different dual-process theories. Therefore, intelligence and the cognitive There is evidence for both categories; unconscious mostly work in concert with fMRI evidence suggests that the type of each other during our daily lives, but in processes are independent – under process- some situations they may be competitive – ing conditions that favor automatic process- and depending on the situation, either con- ing, automatic cognitive processes and the trolled or spontaneous cognitions will be the brain regions supporting those processes are more important contributor to intelligent more active than the brain regions support- behavior. ing controlled cognition. Conversely, under Interestingly, while various dual-process conditions that favor controlled processing, theories of cognition have been proposed controlled cognitive processes and the brain over the years, only two are explicitly theo- regions supporting those processes (such as ries of human intelligence. Below I will review the dorsolateral prefrontal cortex) are more both: Anderson’s (M. Anderson, 2005) theory active than the brain regions supporting of the minimal cognitive architecture under- automatic cognitive processes (Lieberman, lying intelligence and development and the 2007). recent dual-process (DP) theory of human There is also support for the default- intelligence (Kaufman, 2009a). interventionists’ view in that humans on average have a tendency to contextual- The Theory of the Minimal Cognitive ize information (i.e., automatic cognition Architecture underlying Intelligence is the default mode in most humans) and Development and that in some instances it is important for controlled cognition to reflect on Based on Fodor’s (1983) distinction between that contextualization and potentially over- central processes of thought and dedi- ride the outputs of automatic cognition cated processing input modules, Anderson’s (Kahneman & Frederick, 2002; Chapter 22, (2005) theory synthesizes the idea of gen- Intelligence and Rationality, this volume). eral and specific abilities and incorporates Nonetheless, in some situations the output the notion of development. Anderson argues of the automatic system is beneficial for that knowledge is acquired through two dif- intelligent behavior, and controlled cogni- ferent “processing routes,” with central pro- tion is not necessary, or can even get in cesses (route 1) being tied to individual dif- the way. ferences and input modules being tied to Interestingly, a number of neuroimag- cognitive development (route 2). Accord- ing studies in humans and lesion stud- ing to Anderson, route 1 involves “thought- ies on rodents have found that the basal ful problem solving” and is constrained by ganglia and medial temporal lobe (mTL) the speed of a basic processing mechanism. function competitively (Packard, Hirsh, & Anderson argues that “it is this constraint White, 1989; Poldrack & Packard, 2003). In that is the basis of general intelligence and an interesting study, Packard, Hirsh, and the reason why manifest specific abilities are White (1989) found that rats with basal gan- correlated” (p. 280). Anderson’s basic pro- glia lesions performed better than normal cessing mechanism comprises both a ver- rats on an mTL-specific task, and rats with bal and a spatial processor that are nor- mTL lesions performed better than nor- mally distributed, uncorrelated with each mal on the basal ganglia–specific task. These other, and each having their own predictive results suggest that the presence of a nor- powers. mally functioning medial temporal lobe may In contrast, the second route for acquiring interfere with performance on tasks that knowledge in Anderson’s model is tied to P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

448 SCOTT BARRY KAUFMAN

dedicated information-processing modules, Dual-Process (DP) Theory of such as perception of three-dimensional Human Intelligence space, syntactic parsing, phonological encoding, and theory of mind. According The dual-process theory of human intelli- to Anderson, this route is tied to cognitive gence aims to integrate modern dual-process development as these modules undergo theories of cognition (e.g., Evans & Frankish, developmental changes in cognitive com- 2009) with research on intelligence (Kauf- petence across the life span. Anderson man, 2009a). The theory is an organizing acknowledges that modular processes can framework for various constructs relating be acquired through extensive practice, to human cognition that are at least par- but both are similar in that they operate tially separable and display individual differ- automatically and independently of the first ences that are meaningfully related to a wide route and are therefore unconstrained by range of socially valued intelligent behav- the speed of the basic processing mecha- iors. A main goal of the theory is to expand nism. both the range of methodologies and the Anderson makes the case that the modu- dependent measures traditionally studied by lar component of his cognitive theory allows intelligence researchers in order to more for an integration of Gardner’s “multiple clearly define the cognitive mechanisms intelligences” and “general intelligence,” as underlying each construct and to develop the theory includes domain-specific mod- interventions to increase these abilities in ular functions as well as a basic process- everyone. ing mechanism. Anderson also argues that According to the theory, performance his theory explains how low-IQ individu- across a wide range of intelligent behav- als can be capable of remarkable cognitive iors can be predicted through a hierarchical feats (e.g., “savant” abilities), including var- structure of controlled and spontaneous cog- ious practical skills, such as the ability to nitive processes. Controlled cognitions are acquire language or see in three dimen- goal directed and consume limited central sions that are considerably more compu- executive resources, whereas spontaneous tationally complex than the abilities that cognitions aren’t constrained by the same are tapped by IQ tests. Anderson argues limited pool of attentional resources. An that his theory also can explain how devel- assumption of the theory is that both con- opmental disabilities such as dyslexia and trolled and spontaneous cognitive processes autism can exist in the presence of typi- to some degree jointly determine all intelli- cal or even above-average IQ (Anderson, gent behaviors, although in varying degrees. 2008). For instance, prediction of performance on Note that in Anderson’s model there is an IQ test will maximize the measurement little room for individual differences in route of controlled cognitive processes whereas 2. Furthermore, Anderson does not propose performance on a test that requires the inci- any domain general learning mechanisms dental learning of a complex pattern or per- that are part of route 2, focusing instead formance in a domain in which someone has on the Fodorian definition of modules. By acquired a large body of expertise will max- limiting the cognitive mechanisms associ- imize the measurement of spontaneous cog- ated with each “route,” the total amount nitive processes. of other research that could be brought Echoes of this idea can be found in to bear on the cognitive processes under- Hammond, Hamm, Grassia, and Pearson lying the two information-processing routes (1987) when they argue that different becomes unnecessarily restricted. Nonethe- decision-making situations will draw on dif- less, Anderson’s model makes an important ferent strategies in a continuum between contribution to investigation of intelligence pure intuition and pure rational analysis. by expanding modes of thought and incor- According to the dual-process theory, nei- porating development. ther component is more important than the P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 449

other, but what is important is the ability and plan for the future.9 Constructs that to flexibly switch between modes of cog- are part of the controlled cognition hier- nition depending on the task requirements archy include central executive functions (for applications of this idea to creativity, (updating, cognitive inhibition, and mental see Chapter 17, The Evolution of Intelli- flexibility), reflective engagement, explicit gence, this volume; Gabora & Kaufman, cognitive ability (the skill sets that lie at 2009; Howard-Jones & Murray, 2003; the heart of highly g-loaded tasks), intellec- Martindale, 1995; Vartanian, 2009). Accord- tual engagement, and elementary cognitive ing to the theory, what has traditionally tasks that support explicit cognitive ability.10 been labeled general intelligence (g)ispri- What links all of the processes together is marily tapping into explicit cognitive abil- that they all draw on a limited capacity pool ity, and the theory predicts that individual of attentional resources. differences in spontaneous cognition will pre- The second main component (alongside dict variance in a wide variety of intelligent controlled cognition) of the dual-process the- behaviors above and beyond the variability ory, and the component that contains pro- in g, which itself is thought to be only a part cesses relating to the cognitive unconscious, of controlled cognition. is spontaneous cognition. At the broadest Both forms of cognition involve the abil- level, individual differences in spontaneous ity and the tendency to engage in each mode cognition reflect the ability to acquire infor- of thought. The two are related because mation automatically and the tendency to people tend to engage in things they are engage in spontaneous forms of cognition. good at and avoid engaging in things they For instance, whereas most people have the aren’t good at. A key assumption of the ability to spontaneously experience emo- dual-process theory is that abilities are not tions and daydream, there may be indi- static entities but are constantly changing vidual differences in the extent to which through the life span as the person contin- people are willing to engage in their emo- ually engages with the world. The more a tions and to daydream (see Pacini & Epstein, person engages in a mode of thought, the 1999; Zhiyan & Singer, 1997).11 Constructs more that individual will develop skills in that are part of the spontaneous cogni- that modality, which in turn increases the tion hierarchy include spontaneous informa- desire for engaging with that skill. Indeed, tion acquisition abilities (implicit learning, research on expertise skill acquisition shows reduced latent inhibition, etc.), spontaneous that engagement in a domain through many forms of engagement (affective engagement, hours of deliberate practice contributes to the generation of mental structures that can 9 Note that other definitions of “controlled cognition” surpass information-processing limitations have been put forward (see Schneider & Shiffrin, 1977). when performing within that domain (Eric- 10 1994 It should be noted, however, that elementary cogni- sson & Charness, ; Ericsson & Kintsch, tive tasks (ECTs) are not process pure, and motiva- 1995; Ericsson & Lehmann, 1996, but see tion, strategy use, and the allocation of attentional Kaufman, 2007). resources play an important role in performance (see Chapter 38, Intelligence and Motivation, this Controlled cognition is at the top of the volume; Cokely, Kelley, & Gilchrist 2006; Fox, hierarchy (alongside spontaneous cognition) Roring, & Mitchum, 2009). 11 because the capacity for goal-directed action Note that the distinction between controlled and spontaneous cognition is not always the same is an important component of human intel- as the distinction between conscious and uncon- ligence. Controlled cognition consists of a scious modes of thought. Spontaneous cognitions class of cognitive processes that involve can be either conscious, such as when individuals are consciously aware of their daydreaming, fan- the ability and tendency across situations tasy, or mind wandering, or nonconscious such as to think about thinking (i.e., “metacogni- when individuals are dreaming, daydreaming with- tion” – see Dennett, 1992; Hertzog & Robin- out conscious awareness, or implicitly learning the 2005 underlying rule structure of the environment with- son, ), reflect on prior behavior, and out awareness of how that tacit knowledge is affect- use that information to modify behavior ing their behavior. P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

450 SCOTT BARRY KAUFMAN

aesthetic engagement, and fantasy engage- domains of human cognitive functioning: ment), and various implicit domains of mind social cognition and creative cognition. The that are universal human domains pertaining chapter will then conclude with a call for to knowledge of people, language, numbers, more research. The review of studies in this animals, music, visual images, aesthetics, or chapter is by no means exhaustive but is the inanimate physical world (see Carey & meant to highlight some of the latest think- Spelke, 1994; Feist, 2001; Hirschfeld & Gel- ing and research on the relation between man, 1994).12 individual differences in psychometric intel- Other technical details about the the- ligence and individual differences in the cog- ory, including the hierarchical nature of the nitive unconscious. model can be found in Kaufman (2009a). Thus far, there is support for the theory from different branches of psychology and neu- Intelligence and Preconscious ropsychology. The theory has not received Processing many criticisms, but it is still new; thus, Intelligence and Implicit Learning the extent to which the dual-process theory of human intelligence advances the ﬁeld According to Reber (1993), implicit learn- by making new, testable predictions and the ing is “a fundamental root process . . . that extent to which the theory more clearly lies at the very heart of the adaptive behav- deﬁnes various constructs relating to intel- ioral repertoire of every complex organism” ligence is still to be determined. and can be characterized as “the acquisition The rest of this chapter reviews recent of knowledge that takes place largely inde- empirical work on linkages between the cog- pendent of conscious attempts to learn and nitive processes underlying psychometric largely in the absence of explicit knowledge intelligence and various aspects of the cog- about what was acquired” (p. 5;forasim- nitive unconscious. First, relations between ilar view see Epstein & Meier, 1989). We individual differences in controlled cognitive frequently encounter many complex contin- processing and individual differences in two gencies and patterns, and the ability to pre- forms of preconscious processing, implicit consciously learn patterns and then use that learning, and latent inhibition will be dis- knowledge to recognize and detect patterns cussed. Because intuitions and insights gen- in the future is an important component of erally follow preconscious processing, the intelligence (see Hawkins, 2005). next section of this chapter reviews evi- What is the link between psychometric dence on the relation between intelligence intelligence and implicit learning? Accord- and individual differences in both intuition ing to Reber (1993) and Epstein and Meier and insights. The following section will then (1989), individual differences in implicit look at the implications of intelligence and learning should be unrelated to individual the cognitive unconscious for two major differences in measures of explicit cognition. Applying principles of evolutionary biology, they argue that the capacity for explicit 12 Implicit domains of mind are similar to group factors in hierarchical models of intelligence. Indeed, cognition arrived later on the evolutionary research shows that group factors, such as mathe- scene than did implicit cognition. Nonethe- matical, spatial, and verbal reasoning abilities pro- less, the older implicit learning mecha- vide incremental validity for predicting associated vocations above and beyond general intelligence nisms were unaffected by the emergence of (Achter, Lubinski, Benbow, & Eftekhari-Sanjani, explicit thought and continue to function 1999; Humphreys, Lubinski, & Yao, 1993). These autonomously. domains of mind are also related to Howard Gard- ner’s “multiple intelligences” (Gardner, 1993, 1999), Thus far, the majority of the evidence although the dual-process theory acknowledges that supports the notion that implicit learning there are also more general forms of cognition that ability is independent of IQ. Some implicit contribute to intelligent behavior, a criticism that is often leveled against theories of multiple intelli- learning tasks have never demonstrated gences (see Lohman, 2001). a relation with explicit cognitive ability P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 451

(e.g., artificial grammar learning; Gebauer learning task from the procedural compo- & Mackintosh, 2007; McGeorge, Crawford, nent using a sample of 455 adolescents; & Kelly, 1997; Reber, Walkenfeld, & Hern- they found that while the declarative learn- stadt, 1991), whereas other tasks have not ing component significantly correlated with shown a significant association in the major- explicit cognition, the procedural compo- ity of the studies (e.g., serial reaction time nent did not. In another line of research, learning; Feldman, Kerr, & Streissguth, 1995; using a population of individuals with autis- Kaufman, DeYoung, Gray, Jimenez,´ Brown, tic spectrum condition (ASC), Brown et al. & Mackintosh, 2010; Pretz, Totz, & Kauf- (2010) found that matching for IQ, there was man, 2010; Unsworth & Engle, 2005 – but see statistical equivalence between participants Salthouse, McGuthry, & Hambrick, 1999). with ASC and typically developing individ- One other implicit learning task, which uals on four implicit learning tasks. Further, involves unintentional exposure to pictures, this finding was not a consequence of com- did show an association once with explicit pensation by explicit learning ability or IQ. cognitive ability (Fletcher, Maybery, & Ben- Taken together, the research supports the nett, 2000). These results may be mixed separation of explicit and implicit cognition as different implicit learning tasks are only and the notion that individual differences in weakly correlated with each other (Gebauer psychometric intelligence are only weakly if & Mackintosh, 2007, 2009; Salthouse et al., at all associated with individual differences 1999). Further, some implicit learning in implicit learning (e.g., McGeorge et al., paradigms may better capture implicit cog- 1997; Reber et al., 1991). nition than others, which may draw more Recent research has found that individ- on explicit cognition (e.g., Seger, 1994). An ual differences in implicit learning make an important future line of research to better independent contribution to complex cogni- understand the relation of implicit learn- tion above and beyond psychometric intel- ing to psychometric intelligence will be to ligence. Gebauer and Mackintosh (2009) construct reliable measures that more accu- administered a large battery of implicit rately assess implicit learning. Then, the fac- learning and intelligence tests to 195 Ger- torial structure of implicit learning tasks can man students. A factor analysis of all the be assessed and the convergent-discriminant tasks revealed two second-order principal validity can be compared to other measures components: the first consisting primarily of psychometric intelligence. of the intelligence measures and the second Another methodology with which to consisting of the measures of implicit learn- investigate the link between implicit and ing. Both factors were only weakly related to explicit cognition is to compare implicit each other. Additionally, the implicit learn- and explicit versions of the same task. ing second-order factor was significantly In one condition, experimenters instructed related to math and English grades, subjects participants to find the pattern, whereas that were foreign languages for the German in another condition participants received students in the sample. Controlling for the no such instruction, thereby making learn- intelligence second-order factor, the associ- ing unintentional. When this methodol- ation between the implicit learning factor ogy is employed, psychometric intelli- and English remained whereas the associa- gence is more highly correlated with the tion with math was no longer significant. task under explicit instructions compared Consistent with this finding, Pretz, Totz, with the condition in which participants and Kaufman (2010) found a relation are not instructed to intentionally search between a probabilistic sequence learning for the pattern (Gebauer & Mackintosh, task and both the American College Test- 2007; Unsworth & Engle, 2005). Using ing (ACT) math and English scores, and a similar methodology, Feldman, Kerr, these effects were in the middle third of and Streissguth (1995) separated an inten- effect sizes reported in psychology (r = .2 tional declarative component of an implicit to .3; Hemphill, 2003). In another recent P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

452 SCOTT BARRY KAUFMAN

study, Kaufman et al. (2009) investigated conditions to aid in focusing attention, the association of individual differences in whereas only selective attention processes implicit learning with a variety of cogni- are necessary for learning stimuli inciden- tive and personality variables in a sample tally (Cowan, 1988; Frensch & Miner, 1995, of English 16-to17-year-olds. Probabilistic Johnson & Hirst, 1993). In support of this sequence learning was related to intentional view, Unsworth and Engle (2005) found that associative learning more strongly than psy- variations in working memory were asso- chometric intelligence, and it was not associated with an implicit learning task only ciated with working memory. Furthermore, when participants were instructed to explic- structural equation modeling revealed that itly detect the covariation, but no associa- individual differences in implicit learning tion with working memory was found when were independently related to verbal ana- participants were not given that instruction. logical reasoning and processing speed, and Feldman, Kerr, and Streissguth (1995)also implicit learning was significantly corre- found no relation between implicit learning lated with academic performance on two and measures of working memory. foreign language exams (French and Ger- In sum, while the literature is not large, man). Implicit learning also was positively the evidence that does exist suggests that related to self-report measures of personal- implicit learning is often unrelated to psy- ity, including intuition, Openness to Expe- chometric intelligence or working memory rience, and impulsivity. Also, a double but is independently associated with spe- dissociation was found between a latent cific forms of complex cognition, academic Intellect factor and a latent Openness to achievement, and particular aspects of per- Experience factor – with Intellect relating sonality related to Openness to Experience to working memory (.29) but not implicit and impulsivity. Future research on the learning (.00) and Openness to Experience topic is needed to clarify and extend these relating to implicit learning (.31) but not findings. working memory (.13). This lack of association between implicit Intelligence and Latent Inhibition learning and working memory is consistent with other research on attention and execu- It can be important in our everyday lives to tive functioning. Research shows that those be able to automatically distinguish relevant high in working memory are better able to from irrelevant stimuli and to filter out control their attention and stay on task when information irrelevant to the task at hand. there is interference (Kane, Bleckley, Con- For instance, when trying to concentrate on way, & Engle, 2001) and this ability is asso- writing poetry, it’s important to filter out ciated with psychometric intelligence (see the rattle of the radiator. Such a mechanism Chapter 20, Working Memory and Intel- has been investigated and is called latent ligence, this volume). There is an emerg- inhibition (Lubow, 1989). Latent inhibition ing consensus that implicit learning requires is often characterized as a preconscious selective attention to the relevant stimuli gating mechanism that screens from current but then learning about the selected stim- focus those stimuli that have previously uli operates automatically, independent of been regarded as irrelevant (Lubow, 1989). an intention to learn and without drawing Those with increased latent inhibition on further central executive processing (e.g., show higher levels of this form of inhibition Baker, Olson, & Behrmann, 2004; Frensch & (Peterson, Smith, & Carson, 2002). Variation Miner, 1995; Jiang & Chun, 2001;Jimenez´ in latent inhibition has been documented & Mendez, 1999; Turke-Browne, Junge, & across a variety of mammalian species and, Scholl, 2005). at least in other animals, has known biolog- Indeed, researchers have proposed that ical substrates (Lubow & Gewirtz, 1995). central executive functions should be Prior research has shown a relation between engaged only under intentional learning decreased latent inhibition and acute-phase P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 453

schizophrenia (Baruch, Hemsley, & Gray, Intelligence, Intuition, and Insight 1988a, 1988b; Lubow, Ingberg-Sachs, Zalstein-Orda, & Gewirtz, 1992). People Various researchers have come to the con- with schizophrenia also tend to have clusion that in many naturalistic situations, reduced ability for central executive such as decision making in groups, very lit- functioning (Barch, 2005). tle controlled cognition is required (Klein, Recent research suggests that reduced 1999; also see Gladwell, 2007, for a summary latent inhibition can also have its advan- of relevant research). Instead, they note that tages. In students with a high IQ (and pre- expertise seems to be related to recognition sumably a high level of central executive of a situation that had been encountered functioning), decreased latent inhibition is previously and the retrieval of schemas that associated with higher scores on a self-report match the situation.13 They argue that while measure of creative achievement (Carson, controlled cognition is sometimes impor- Peterson, & Higgins, 2003). Interestingly, tant, the key to intelligent behavior is the the researchers did not find a correlation automatic retrieval process. between fluid intelligence and latent inhi- Similarly, Reyna (2004) argued that bition. Kaufman (2009a) also did not find experts acquire knowledge that allows them an association between variations in g and to make fast, intuitive, and effective deci- variations in latent inhibition. Additionally, sions whereas novices need to rely on Kaufman (2009a, b) examined the relation- deliberate, effortful reasoning. Reyna noted, ship between latent inhibition and individ- however, that automatic processes can lead ual differences in the tendency to rely on to bias and error when experts are presented intuition to make decisions. Indeed, latent with novel problems (also see Chabris & inhibition is conceptually related to intu- Simons, 2010, for a summary of research ition: Jung’s original conception of intuition showing the potential perils of relying on is “perception via the unconscious” (Jung, intuition when making expert as well as 1921/1971,p.538). Kaufman hypothesized novel decisions). Wilson and Schooler (1991) that an intuitive cognitive style would be also showed the importance of automatic related to reduced latent inhibition. Results processing in decision making – they demon- showed that those with higher scores on a strated that when making a decision that faith in intuition factor (consisting of intu- is complex and multi-attributed, people do ition items related to affect) tended to have better when conscious deliberation is inten- reduced latent inhibition. Further, latent tionally prevented. This idea is also a major inhibition was not associated with an intu- tenet of the unconscious thought theory ition factor consisting of items having to do (UTT), in which it is argued that decisions with holistic processing of information or a about simple issues can be better tackled by rational cognitive style. There was also a ten- conscious thought, whereas decisions about dency for those scoring high (as compared complex matter can be better approached to medium or low) on the faith in intu- with unconscious thought (Dijksterhuis & ition factor to benefit more from a preexpo- Nordgren, 2006, but see Aczel, 2009; Newell, sure condition where participants received Wong, & Cheung, 2009; Payne, Samper, the relevant stimuli in the first part of the Bettman, & Luce, 2008; Thorsteinson & task. Therefore, current research suggests Withrow, 2009). that decreased latent inhibition is unrelated to general intelligence or a rational cognitive style. Since decreased latent inhibition may 13 For more on the relations between intelligence make an individual more likely to perceive and the acquisition of expertise more generally, and make connections that others do not see, see Ackerman (Chapter 41, Intelligence and Exper- this ability in combination with high psycho- tise, this volume). In this section I focus instead on the relation between intelligence and intuition, metric intelligence can lead to the highest particularly from an individual differences perspec- levels of creative achievement. tive. P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

454 SCOTT BARRY KAUFMAN

Along similar lines, Hogarth (2005)distin- Epstein’s experiential system is related to guished between deliberate and tacit cog- intuition in the sense of “gut-feelings” that nitive processes. According to Hogarth, guide behavior. Based on his theory, Epstein complex decisions will benefit from tacit developed the Rational-Experiential Inven- processing whereas less complex decisions tory (REI; Pacini & Epstein, 1999), which will benefit from deliberate processing. An measures individual differences in the ten- additional component in Hogarth’s model dency to rely on each mode of thought. is the degree of bias in the original learn- His research program has discovered that ing environment. If the feedback presented the intelligence of each system is indepen- in the original learning environment regard- dent of, or very weakly correlated with, ing decision accuracy is clear and immedi- the intelligence of the other (Epstein & ate, the environment is considered “kind,” Meier, 1989), and each subscale (analyti- and accurate causal relationships can be cal and experiential) has unique predictive learned. Environments in which feedback is validity for a wide range of intelligent behav- unclear and not available in a timely man- iors (see Epstein, 2003, for a review). In ner are considered “wicked” and are consid- general, the rational scale is more strongly ered highly biased. In wicked learning envi- positively related to measures of intellectual ronments, the intuitive system is prone to performance such as scores on the Scholas- errors. According to Hogarth, intentional, tic Aptitude Test (SAT) and grade point deliberate thought is best suited to biased averages (GPA) than is the experiential learning environments where the complex- scale, whereas the experiential scale is more ity of the task is low, whereas intuitive pro- strongly positively related to extroversion, cessing is best suited to learning environ- agreeableness, favorable interpersonal rela- ments in which bias is low and complexity of tionships, empathy, creativity, emotionality, the task is high (see Epstein, 2003, and Kah- sense of humor, and art appreciation than is neman, 2009, for related ideas, including the the rational scale. The rational scale is more notion that the quality of an intuitive judg- strongly negatively associated with neuroti- ment is dependent upon the predictability cism, depression, anxiety, stress in college of the environment in which the judgment life, subtle racism, extreme conservatism, is made and the individual’s opportunity to alcohol abuse, and naıve¨ optimism than is learn the regularities in that environment). the experiential scale, whereas the experien- Recently, researchers have investigated tial scale is more strongly negatively associ- the role of individual differences in the use ated with distrust and intolerance than is the of intuition. With the aim of integrating the rational scale. Many of these relations held psychodynamic focus on unconscious pro- even after controlling for the NEO Five Fac- cessing with the cognitive focus on ratio- tor Inventory (NEO-FFI; Costa & McCrae, nal conscious thinking, Seymour Epstein put 1989), which measures the Big Five fac- forth the cognitive-experiential self-theory tors of personality. Other researchers have (CEST; Epstein, 1994), which was an out- used the REI to investigate human cogni- growth of ideas presented in Epstein (1973). tion. For instance, Klaczynski (2009) reviews The theory posits that humans have two par- a number of studies he and his collaborators allel but interacting modes of information conducted using the REI to investigate the processing. The rational system is analytic, development of dual processes across the life logical, abstract, experienced actively and span. consciously, is slower to process informa- Pretz (2008) has extended both the exper- tion, and requires justification via logic and imental work on intuition and the cognitive evidence. In contrast, the experiential system styles approach by looking at the effects of is holistic, affective, concrete, experienced individual differences in an analytical ver- passively, processes information automati- sus intuitive strategy and level of experience cally, and is self-evidently valid (experience on practical problem solving. Pretz reasoned alone is enough for belief). that the more experienced an individual is P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 455

with a task, the less complex the task and are consistent with Baylor’s (Baylor, 2001)U- the more decomposable the problem will shaped model of expertise and intuition and appear to that individual. Pretz noted that research showing the facilitation of intuition the relevant knowledge associated with an for complex, high-stakes decision making everyday problem-solving task is likely to be (Klein, 1999). Indeed, Pretz and Totz (2007) acquired through informal experience, and have developed a scale to measure individ- individuals with more experience will there- ual differences in the tendency to rely on fore have more tacit knowledge but will also three different forms of intuition: affective, be able to better articulate that knowledge. heuristic, and holistic. Another implication As a result, the expert can use metacognitive of Pretz’s study is that many social prob- skills to explicitly identify the main problems may be better suited to the cognitive lem, identify the most relevant information, unconscious, as they may be more complex and identify the consequences of various than nonsocial problems. Whereas individ- courses of action (Antonakis et al., 2002). ual differences in the cognitive unconscious In Pretz’s study, college students were can be adaptive for some social problems, instructed to use either holistic intuition there may be instances of social cognition (bringing to mind all relevant information in which the cognitive unconscious can lead and trusting hunches) or analysis (defin- to undesirable outcomes (see Implicit Social ing the problem, distinguishing the relevant Cognition section). from irrelevant information, and monitoring Another line of research has investigated the problem carefully) when solving vari- the intimate connection between intuition ous practical problems dealing with college and insight. Anecdotally, insight has played life. Pretz found that the effectiveness of the a crucial role in the generation of cre- strategy on task performance interacted with ative ideas. The great French mathemati- the participant’s level of experience: analysis cian Henri Poincare(´ 1921) described inci- worked better for more experienced individ- dents in which an answer came to him only uals whereas novices were slightly more suc- after his conscious attention was directed cessful when they employed a holistic, intu- away from the problem and he wasn’t itive strategy. A similar pattern was found consciously deliberating on the problem. looking at existing individual differences in Poincare´ argued that these moments of sud- strategy preference. den inspiration are the result of unconscious Pretz’s study suggests that among indi- thinking. Based on reflections of his creative viduals with an intermediate level of exper- thought process, he argued that the cre- tise, analytical problem solving can be help- ative process starts with conscious work on ful in perceiving the logic and structure a problem, followed by unconscious work, of the problem, and intuition can distract and then, if insight is successful, another the expert from this critical information. In stage of conscious work to verify that the contrast, intuitive, holistic thought may be ideas makes sense and to work out the impli- best suited for novices in a domain who see cations of the idea. Indeed, insight is consid- the task as ill-defined and need to bring to ered an important component of the cre- mind the relevant information. An impli- ative process (Wallas, 1926). cation of Pretz’s study is that intermediate Empirical work supports these anecdotes. experts should rely on an analytical strategy In reviewing a number of experiments when solving complex, practical problems. relating to implicit thought, intuition, and Full-blown experts who have fully automa- insights, Kihlstrom, Shames, and Dorfman tized their task may benefit from an intuitive (1996) have this to say about the nature of mode of thought. intuition: This distinction between holistic intuition (of the sort studied in Pretz’s study) From the experiments described in this and inferential intuition (full automatiza- chapter, it appears that the processes tion) was made by Hill (1987–1988); the ideas underlying intuitions closely resemble those P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

456 SCOTT BARRY KAUFMAN

which underlie implicit memory. In recog- Recent research, however, suggests that nition, people’s intuitions about the past – different components of the task may the feeling of familiarity, in the absence of differentially relate to controlled cognition. full recollection – seems to be based on the Reber, Ruch-Monachon, and Perrig (2007) perceptual fluency that comes with prim- first replicated earlier research on the ACT ing. . . . We actually think of these men- by finding that participants often under- tal states as implicit thoughts: instances in estimated their degree of closeness to the which an idea or image influences experience, thought, or action in the absence of answer; these subjective reports of close- conscious awareness of what that idea or ness exhibited a positive slope, suggesting image is. that participants possessed implicit knowledge about the task and indeed felt hunches As for the link between intuitions and about their progress that weren’t necessarily insight, they then go on to say: aligned with objective incremental progress. The researchers then distinguished between . . . it is clear that problem solutions, like performance level, processing style, implicit memories, are not discontinuous, all-or- knowledge, and subjective feeling of close- none affairs, remaining entirely uncon- ness to the solution on the ACT. While scious until they emerge full-blown into performance level correlated with verbal the full light of consciousness. There is a intelligence, processing style and implicit point, as they approach and cross what knowledge were not correlated with ver- Wallas (1926), following William James bal intelligence. Further, a faith in intu- (1890), called the “fringe” of consciousness, when we know they are coming, even when ition cognitive style and the Big Five per- we do not know what they are. This is sonality traits Openness to Experience and the point, between preparation and insight, Conscientiousness were all correlated with where intuitions occur. (p. 19) processing style, but not with implicit knowledge on the task. These results sug- Other researchers have investigated the con- gest that a promising research direction is to trolled and spontaneous cognitive mecha- decompose problem-solving tasks into their nisms that underlie insight (see Sternberg processing style and intuitive components & Davidson, 1995, for a review of research and investigate relations between individual on insight). A methodology that is often differences in these components and indi- employed is the Accumulated Clues Task vidual differences in various processes and (ACT), in which participants must discover thinking styles relating to intelligence and a word, but are given clues (e.g., words that the cognitive unconscious. are associated with the answer) along the way. After each clue is presented, participants are required to provide an answer. Domains The clues get increasingly helpful (are more Implicit Social Cognition related to the answer) and the answers given by the participants get objectively closer to There is an emerging consensus in the the answer in an incremental fashion that social cognition literature that many of our occurs before their subjective ratings of feel- social behaviors and judgments are made ing close to an answer, which they often automatically, without intention, effort, or report occurring to them in a sudden flash awareness (Bargh & Chartrand, 1999; Bargh of insight (Bowers, Farvolden, & Mermigis, & Morsella, 2008). Research on automatic 1995; Dorfman, Shames, & Kihlstrom, 1996). evaluation, impression formation, and auto- Research has shown that individual differ- matic characterization all demonstrate the ences in how long it takes participants to prevalence of automaticity in social life. It arrive at the correct answer correlate with is generally thought now that mere percep- verbal intelligence. tion of a stimulus can lead instantly and P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 457

automatically to a judgment without any such as race, gender, and even insects. In conscious reflection or reasoning. Indeed, studies that involve some measure of dis- until the 1980s, attitudes were mostly crimination toward a social group, both assumed to rely on consciously available explicit and IAT measures predict behav- information (Nosek, Greenwald, & Banaji, ior, with the IAT offering superior pre- 2007). diction (Greenwald, Poehlman, Uhlmann, Recently, researchers have investigated & Banaji, 2009). Furthermore, it has been individual differences in implicit social demonstrated that people with the strongest cognition, using a variety of measures automatic racial biases are most likely to “that avoid requiring introspective access, engage in a wide variety of discriminatory decrease the mental control available to pro- behavior, including overt behavior (Rudman duce the response, reduce the role of con- & Ashmore, 2007, but see van Ravenzwaaij, scious intention, and reduce the role of self- van der Maas, & Wagenmakers for an alter- reflective, deliberative processes” (Nosek et native account). al., 2007,p.267). Greenwald and Banaji Therefore, research on how individual (1995) have been among the most active differences in intelligence and the cognitive researchers investigating the role of implicit unconscious interact to produce stereotyp- cognition in various social psychology con- ing and attitude formation is of both theoret- structs such as attitudes, stereotypes, and ical and practical interest. Recent research self-esteem. In their research, they attempt utilizing fMRI techniques provides some to “reveal traces of past experience that peo- clues. Chee, Sriram, Soon, and Lee, 2000) ple might explicitly reject because it con- used fMRI to examine participants while flicts with values or beliefs, or might avoid these individuals were taking the IAT. The revealing because the expression could have researchers found that the left dorsolateral negative social consequences. Even more prefrontal cortex and to a lesser degree the likely, implicit cognition can reveal infor- anterior cingulate were most active dur- mation that is not available to introspec- ing conditions in which items from incon- tive access even if people were motivated to gruent categories (e.g., insect + pleasant) retrieve and express it” (Nosek et al., 2007, shared a response key than when items p. 266; see Wilson, Lindsey, & Schooler, from congruent categories (e.g., flower + 2000, for related ideas about attitudes). pleasant) shared a key. According to the One of the best-validated measures of researchers, this suggests that greater con- implicit social cognition is the Implicit Asso- trolled cognition was required in condi- ciation Test (IAT; Greenwald, McGhee, tions in which it was necessary to overcome & Schwartz, 1998). The IAT requires the the prepotent tendency to map emotion- participant to categorize various stimu- ally congruent items to the same response lus exemplars representing four concepts key. In another study, Phelps et al. (2000) (e.g., men, women, good, bad) using two had White participants view faces of unfa- response options. When concepts that share miliar Black and White males. Participants a response are strongly associated, it is who showed greater activation of the amyg- expected that the sorting task will be eas- dala (a region of the brain associated with ier for the participant (as indexed by faster fear and negative emotions) while viewing responses and fewer errors) than when the Black faces relative to White faces tended concepts are weakly associated. Thus, the to score higher on two measures of uncon- IAT affords insight into automatic associa- scious race evaluation: the IAT and the eye- tive processes that are introspectively inac- blink response. In a second experiment, the cessible. Over the last decade, the IAT has researchers did not find the same pattern been adapted for use in various disciplines of brain activation when the faces were (see Nosek et al., 2007, for a review) and to familiar and the participants regarded the assess implicit attitudes related to categories Black and White individuals positively. In a P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

458 SCOTT BARRY KAUFMAN

related study, Cunningham et al. (2004)had may be the better predictor of intelligent participants view Black and White faces behavior (since it helps override generaliza- either subliminally or supraliminally dur- tions that can lead to explicit prejudice and ing fMRI. When presented subliminally, stereotyping). Such research would further the amygdala was more active for Black illustrate the need for measuring individual faces relative to White faces. This effect differences in both controlled and automatic was reduced when the faces were presented cognitive processes in order to predict vari- supraliminally. Further, control regions in ous forms of intelligent behavior. the prefrontal cortex (which are also activated during working memory and psycho- Creative Cognition metric intelligence tests) showed greater activation for Black faces than White faces Creativity requires both novelty and useful- when presented supraliminally. Race bias as ness (Kaufman, 2007). The Creative Cog- assessed by the IAT was related to a greater nition Approach endeavors to identify and difference in amygdala activation for Black investigate the role of mental processes in faces relative to White faces, and activity creative cognition at various stages in the in the prefrontal cortex predicted a reduc- creative process (Finke, Ward, & Smith, tion in amygdala activation from the sublim- 1992, 1995; Ward, Smith, & Finke, 1999). inal to the supraliminal condition. Accord- Creative cognition researchers have identi- ing to the researchers, this provides evidence fied two main phases of creative invention for neural distinctions between automatic that occur in a cyclical fashion in ordinary and controlled processing of social groups, individuals. During the generative phase, suggesting that controlled processes (which the individual generates numerous candi- support performance on measure of psy- date ideas or solutions and forms a men- chometric intelligence) may modulate auto- tal representation (referred to as a preinven- matic evaluation. tive structure). Then during the exploratory These results suggest that individual dif- stage, the individual examines the can- ferences in measures of controlled cogni- didate mental representations and ideas tion may predict the extent to which auto- and consciously and sometimes painstak- matic evaluations influence behavior. To ingly works out their implications. Cognitive expand the range of individual differences in unconscious processes activated through implicit social cognition investigated, it may defocused attention most likely play more be useful to construct new implicit learning of a role during the generative stage, tasks that consist of stimuli relating to the whereas controlled cognitive processes acti- learning of real-world contingencies in the vated through focused attention most likely social domain. Tasks that already exist that play more of a role during the exploratory could be adapted include the task used by stage. The highest levels of creativity, how- Lewicki, Hill, and Sasaki (1989), in which ever, most likely require the ability for both participants implicitly learn to judge the modes of thought and the flexibility to intelligence of individuals from brain scans switch modes of thought throughout the or the adaptation of that task employed by creative process. Woolhouse and Bayne (2000), in which par- On the one hand, behavioral and brain ticipants implicitly learn to judge the job studies suggest that creative people are char- suitability of job candidates based on their acterized by a lack of inhibition (Eysenck, personality profile. Such research can help 1995; Martindale, 1999), and case studies distinguish between situations in which indi- repeatedly show that creative people do vidual differences in the cognitive uncon- describe the creative process as effortless scious contribute to intelligent behavior (for and lacking in deliberation (Csikzentmiha- example, when a person is engaging in an lyi, 1996). However, studies also show that area of expertise or generating novel ideas), creative individuals defocus their attention and situations in which controlled cognition when approaching a creative task but they P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 459

are capable of focusing their attention when of association between words can lead to it comes time to make the ideas practical an advantage over time in the total number (Martindale, 1999). In recent years, Oshin of potentially relevant conceptual associa- Vartian and colleagues have extended this tions that can be considered. The researchers research by showing in a series of clever argued that the task they used involved experiments that creative people are able to unambiguous task instructions and associa- adjust their focus of attention, depending on tions and that it is just these conditions in the demands of the task. which those with better divergent thinking In one study, Vartanian, Martindale, and skills focus their attention, which can result Kwiatkowski (2007) found a negative corre- in a faster reaction time. lation between creative potential (measured An interesting question raised by Var- by fluency scores) and speed of informa- tanian, Martindale, and Matthews (2009) tion processing on two tasks that did not is whether the mechanism that regulates involve interference or ambiguity, and a pos- the focus of attention is itself automatic itive correlation between creative potential or requires self-control. They have argued and speed of information processing on two that the unambiguous nature of their task tasks that did require the inhibition of inter- led to automatic regulation of attention. fering information. Therefore, subjects with They point to evidence that in other circum- greater creative potential were better able stances, top-down processing can also play to slow down or speed up their information an important role in creative cognition. Var- processing, depending on the task demands. tanian, Martindale, and Kwiatkowski (2003) A follow-up study found similar results and investigated the role of strategic flexibility extended the earlier results in a sample of in creative problem solving. They adminis- high school students in Russia (2008). The tered a rule discovery task and found that same pattern was found between creative participants with higher creative potential potential (as measured by fluency, flexibil- (as measured by fluency scores) were better ity, and originality) and response latency as at discovering the rules. Further, the strat- in the earlier study, and the findings held, egy of generating disconfirmatory hypothe- correcting for IQ. In a third study, partici- ses played an important role for successful pants were instructed to judge whether two participants in the later stages of hypothe- concepts were related or unrelated (Varta- sis testing after the first feedback was given. nian, Martindale, & Matthews, 2009). The Having already formed a representation of rationale was that creativity is frequently the problem space after feedback, successful defined as the novel and useful association participants were flexibly able to switch to of concepts that are not traditionally related. a more successful strategy following initial Therefore, this important cognitive process feedback. Similar results have been found relies at least in part on a person’s ability by Gilhooly, Fiortou, Anthony, and Wynn to quickly assess the degree of relationship (2007), who found that using think-aloud between concepts. The researchers manip- protocols that alternative uses for a task had ulated the degree of association between generated earlier in the course of the task word pairs. Participants with greater cre- drew primarily on memory-based strategies, ative potential (assessed by a measure of whereas uses generated later drew on a more divergent thinking) exhibited a faster reac- limited range of strategies requiring exec- tion time when judging the relatedness utive processes, such as imagining the dis- of the concepts. Psychometric intelligence assembly of the object and using the parts didn’t account for additional variance above or recombining the parts into other objects and beyond divergent thinking scores in pre- that could be applied in other ways. Simi- dicting the variability in reaction time per- lar to the results of the Vartanian, Martin- formance. The researchers conclude that the dale, and Kwiatkowki (2003) study, novelty ability of individuals with higher creative of responses was affected by the ability to potential to more quickly judge the degree use a specific strategy later in the course of P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

460 SCOTT BARRY KAUFMAN

problem solving, supporting the view that that can modulate between the different creative people switch strategies during the brain activations and deactivations depend- course of a task but also suggesting that top- ing on the demands of the task. down processing can play an important role in creative problem solving. Vartanian, Mar- tindale, and Kwiatkowski (2003) suggested a Conclusion bi-directional model of creativity in which the focus of attention is modulated accord- In his 1957 presidential address to the Amer- ing to top-down as well as bottom-up pro- ican Psychological Association, Lee Cron- cesses, with the use of bottom-up process- bach pleaded his case for uniting the bur- ing determined by the stage of the problem geoning field of cognitive psychology, with (bottom-up processing primarily during the its focus on the experimental psychology of earlier stages, and top-down processing pri- higher order information processing, with marily during the later stages). Both Var- the study of individual differences in Spear- tanian, Martindale, and Kwiatkowski (2003) man’s g. Cronbach’s call set off a great and Vartanian (2009) mentioned that an deal of research that would demonstrate important future line of research will be that the newer theories regarding the nature to investigate the underlying mechanism(s) of intelligence and the burgeoning field that enable the modulation of information- of information-processing psychology were processing strategies during the course of indeed quite compatible. The work by Hunt creative problem solving. (Hunt, Frost, & Lunneborg, 1973) and Stern- Drawing more on the memory and brain berg (1977) helped lay the foundation for the literature, Bristol and Viskontas (2006) came experimental study of intelligent reasoning to similar conclusions. They proposed that processes that are deliberate and effortful. creative individuals are good at modulat- Subsequent research has tended to focus on ing inhibitory processes, so that they have both lower level as well as higher level cor- both the capability for cognitive control and relates of general intelligence. the capacity for disinhibition and can switch One particular set of cognitive processes fluidly from one mode to another. In par- that has not been investigated as thoroughly ticular, they argue that creative individu- as the others from an individual differences als can defocus their attention at the early perspective is the set related to the cogni- stages of creative cognition so that they tive unconscious. This situation of mutual grasp the whole set of potential covaria- neglect has had the unfortunate conse- tions; then, during the retrieval and elabo- quence of limiting our picture of the nature ration stage, they can control attention so of both human intelligence and the cognitive that they can inhibit prepotent responses unconscious, thus potentially limiting our and thereby allow remote associations to understanding of the role of individual dif- enter into consciousness without intrusions. ferences in information processing in com- Therefore, the researchers argue that cre- plex cognition more generally. The study ative individuals are both able to overcome of individual differences in the cognitive cognitive inhibition and are capable of sup- unconscious can increase our understanding pressing undesired responses. They claim of the nature of intelligence by helping us that this skill requires the ability to acti- find boundary conditions for so-called gen- vate the dorsolateral prefrontal cortex and eral intelligence (g) and by doing so, dis- inhibit retrieval-related processes that may covering where g breaks down. Similarly, interfere with accessing remote associations, the study of individual differences in gen- as well as to deactivate the dorsolateral pre- eral intelligence and its associated cognitive frontal cortex, depending on the context mechanisms can elucidate the nature of the of the task and the goals of the individual. cognitive unconscious by helping to clarify They also left as an interesting open question and delineate automatic, spontaneous, and determining the precise brain mechanisms rapid information-processing mechanisms. P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 461

By charting new terrains, researchers can Anderson, M. (2008). What can autism and increase understanding of the determinants dyslexia tell us about intelligence? Quarterly of intelligent behavior. A potentially fruit- Journal of Experimental Psychology, 61, 116– ful line of research is to adapt already exist- 128. ing experimental paradigms and construct Antonakis, J., Hedlund, J., Pretz, J., & Stern- 2002 new tests that tap the cognitive uncon- berg,R.J.( ). Exploring the nature and acquisition of tacit knowledge for military lead- scious. Individual differences in such tasks ership (Research Note 2002–04). Alexandria, may not be strongly related to psycho- VA: Army Research Institute for the Behav- metric intelligence but may still explain ioral and Social Sciences. intelligent behavior independent of psycho- Baker, C. I., Olson, C. R., & Behrmann, M. metric intelligence. Researchers can then (2004). Role of attention and perceptual investigate the precise cognitive and neural grouping in visual statistical learning. Psycho- mechanisms that underlie measures of the logical Science, 15, 460–466. cognitive unconscious and develop interven- Barch, D. M. (2005). The cognitive neuroscience tions to raise these skills in everyone. By of schizophrenia. Annual Review of Clinical 321 353 fostering collaborations across the various Psychology, 1, – . 2004 areas of psychology and related disciplines, Bargh, J. A. ( ). The four horsemen of automaticity: Awareness, intention, efﬁciency, and and incorporating dual-process theory into control in social cognition. In R. S. Wyer & our thinking, we should be able to come T. K. Srull (Eds.), Handbook of social cognition to a fuller, more complete understanding of (pp. 1–41). Hillsdale, NJ: Erlbaum. human intelligence. Bargh, J. A., & Chartrand, T. L. (1999). The unbearable automaticity of being. American Psychologist, 54, 462–479. Acknowledgments Bargh, J. A., & Morsella, E. (2008). The unconscious mind. Perspectives on Psychological Sci- I would like to thank Edward Cokely, ence, 3, 73–79. 1988 Seymour Epstein, Jean Pretz, and Robert Baruch, I., Hemsley, D. R., & Gray, J. A. ( a). Sternberg for their insightful suggestions on Latent inhibition and “psychotic proneness” in normal subjects. Personality and Individual an earlier draft. Differences, 9, 777–783. Baruch, I., Hemsley, D. R., & Gray, J. A. (1988b). Differential performance of acute and chronic References schizophrenics in a latent inhibition task. Jour- nal of Nervous and Mental Disease, 176, 598– Achter, J. A., Lubinski, D., Benbow, C. P., & 606. Eftekhari-Sanjani, H. (1999). Assessing voca- Baylor,A.L.(2001). A U-shaped model for the tional preferences among gifted adolescents development of intuition by level of expertise. adds incremental validity to abilities: A dis- New Ideas in Psychology, 19, 237–244. criminant analysis of educational outcomes Binet, A., & Simon, T. (1916). The development over a 10-year interval. Journal of Educational of intelligence in children (E. S. Kite, Trans.). Psychology, 91, 777–786. Baltimore, MD: Williams & Wilkens. Aczel, B. (2009). Attention and awareness in Bowers, K. S., Farvolden, P., & Mermigis, L. human learning and decision making (Unpub- (1995). Intuitive antecedents of insight. In lished doctoral dissertation). University of S. M. Smith, T. B. Ward & R. A. Finke (Eds.), Cambridge, Cambridge, UK. The creative cognition approach (pp. 27–51). Anderson, J. R. (1982). Acquisition of cognitive Cambridge, MA: MIT Press. skill. Psychological Review, 89, 369–406. Bristol, A. S., & Viskontas, I. V. (2006). Dynamic Anderson, M. (2005). Marrying intelligence and processes within associative memory stores: cognition: A developmental review. In R. J. Piecing together the neural basis of creative Sternberg & J. E. Pretz (Eds.), Cognition & cognition. In J. C. Kaufman & J. Baer (Eds.), intelligence: Identifying the mechanisms of the Creativity and reason in cognitive development mind (pp. 268–288). Cambridge, UK: Cam- (pp. 60–80). New York, NY: Cambridge Uni- bridge University Press. versity Press. P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

462 SCOTT BARRY KAUFMAN

Brown, J. B., Aczel, B., Jimenez, L., Kaufman, Costa,P.T.,&McCrae,R.R.(1989). The NEO- S. B., Mackintosh, N., & Plaisted, K. (2010). PI/NEO-FFI manual supplement. Odessa, FL: Intact implicit learning in autism spectrum Psychological Assessment Resources. conditions. Quarterly Journal of Experimental Cowan, N. (1988). Evolving conceptions of Psychology, 1, 1–24. memory storage, selective attention, and Carey, S., & Spelke, E. (1994). Domain-specific their mutual constraints within the human knowledge and conceptual change. In L. A. information-processing system. Psychological Hirschfeld & S. A. Gelman (Eds.), Mapping Bulletin, 104, 163–191. the mind: Domain specificity in cognition and Cronbach,L.J.(1957). The two disciplines of sci- culture (pp. 169–200). New York, NY: Cam- entific psychology. American Psychologist, 12, bridge University Press. 671–684. Carroll, J. B. (1993). Human cognitive abilities: Csikzentmihalyi, M. (1996). Creativity: Flow and A survey of factor-analytic studies. Cambridge, the psychology of discovery and invention.New UK: Cambridge University Press. York, NY: Harper Collins. Carson, S. H., Peterson, J. B., & Higgins, D. M. Cunningham, W. A., Johnson, M. K., Raye, (2003). Decreased latent inhibition is associ- C.L.,Gatenby,J.C.,Gore,J.C.,&Banaji, ated with increased creative achievement in M. R. (2004). Separable neural components in high-functioning individuals. Journal of Per- the processing of black and white faces. Psy- sonality and Social Psychology, 85, 499–506. chological Science, 15, 806–813. Chabris, C. F. (2007). Cognitive and neurobiolog- Dennett, D. C. (1992). Consciousness explained. ical mechanisms of the Law of General Intel- New York, NY: Back Bay Books. ligence. In M. J. Roberts (Ed.), Integrating the Dijksterhuis, A., & Nordgren, L. F. (2006). A mind: Domain general vs. domain specific pro- theory of unconscious thought. Perspectives on cesses in higher cognition (pp. 449–491). New Psychological Science, 1, 95–109. York, NY: Psychology Press. Dorfman, J., Shames, V. A., & Kihlstrom, Chabris, C.F., & Simons, D. (2010). The invisible J. F. (1996). Intuition, incubation, and insight: gorilla: And other ways our intuitions deceive us. Implicit cognition in problem solving. In New York, NY: Crown Archetype. G. D. M. Underwood (Ed.), Implicit cognition Chaiken, S. (1987). The heuristic model of per- (pp. 257–296). New York, NY: Oxford Uni- suasion. In M. P. Zanna, J. M. Olson, & C. P. versity Press. Herman (Eds.), Social influence: The Ontario Dorfman, L., Martindale, C., Gassimova, V., & symposium (vol. 5, pp. 3–39). Hillsdale, NJ: Vartanian, O. (2008). Creativity and speed of Erlbaum. information processing: A double dissociation Chee, M. W. L., Sriram, N., Soon, C. S., & Lee, involving elementary versus inhibitory cogni- K. M. (2000). Dorsolateral prefrontal cortex tive tasks. Personality and Individual Differ- and the implicit association of concepts and ences, 44, 1382–1390. attributes. Neuroreport: For Rapid Communi- Epstein, S. (1973). The self-concept revisited or a cation of Neuroscience Research, 11, 135–140. theory of a theory. American Psychologist, 28, Cokely, E. T., & Kelley, C. M. (2009). Cognitive 404–416. abilities and superior decision making under Epstein, S. (1991). Cognitive-experiential self- risk: A protocol analysis and process model theory: An integrative theory of personality. evaluation. Judgment and Decision Making, 4, In R. Curtis (Ed.), The relational self: Conver- 20–33. gences in psychoanalysis and social psychology Cokely, E. T., Parpart, P., & Schooler, L.J. (2009). (pp. 111–137). New York, NY: Guilford Press. On the link between cognitive control and Epstein, S. (1994). Integration of the cognitive heuristic processes. In N. A. Taatgen & H. and the psychodynamic unconscious. Ameri- v. Rijn (Eds.), Proceedings of the 31st Annual can Psychologist, 49, 709–724. Conference of the Cognitive Science Society Epstein, S. (2001). Manual for the Constructive (pp. 2926–2931). Austin, TX: Cognitive Sci- Thinking Inventory. Odessa, FL: Psychological ence Society. Assessment Resources. Cokely, K., Kelley, C. M., & Gilchrist, A. L. Epstein, S. (2003). Cognitive-experiential self- (2006). Sources of individual differences in theory of personality. In T. Millon & M. J. working memory: Contributions of strategy Lerner (Eds.), Comprehensive handbook of psy- to capacity. Psychonomic Bulletin & Review, 13, chology (Vol. 5, pp. 159–184). Personality and 991–997. social psychology. Hoboken, NJ: Wiley. P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 463

Epstein, S. (2010). Demystifying intuition: What Sequenzlernen [The role of working mem- it is, what it does, and how it does it. Psycho- ory in implicit sequence learning]. Zeitschrift logical Inquiry, 21, 295–312. fur¨ Experimentelle Psychologie, 42, 545– Epstein, S., & Meier, P. (1989). Constructive 575. thinking: A broad coping variable with spe- Gabora, L., & Kaufman, S. B. (2009). Evolution- ciﬁc components. Journal of Personality and ary approaches to creativity. In J. C. Kauf- Social Psychology, 57, 332–349. man & R. J. Sternberg (Eds.), Cambridge Ericsson, K. A., & Charness, N. (1994). Expert handbook of creativity. Cambridge, UK: Cam- performance: Its structure and acquisition. bridge University Press. Manuscript in pre- American Psychologist, 49, 725–747. paration. Ericsson, K. A., & Kintsch, W. (1995). Long-term Gardner, H. (1993). Frames of mind: The theory of working memory. Psychological Review, 102, multiple intelligences (2nd ed.). New York, NY: 211–245. Basic Books. Ericsson, K. A., & Lehmann, A. C. (1996). Gardner, H. (1999). Intelligence reframed: Multiple Expert and exceptional performance: Evi- intelligences for the 21st century.NewYork,NY: dence of maximal adaptation to task con- Basic Books. straints. Annual Review of Psychology, 47, 273– Gebauer, G. F., & Mackintosh, N. J. (2007). Psy- 305. chometric intelligence dissociates implicit and Evans, J. S. B. T. (2008). Dual-processing explicit learning. Journal of Experimental Psy- accounts of reasoning, judgment, and social chology: Learning, Memory, and Cognition, 33, cognition. Annual Review of Psychology, 59, 34–54. 255–278. Gebauer, G. F., & Mackintosh, N. J. (2009). Evans, J. S. B. T., & Frankish, K. (2009). In two Implicit learning and intelligence: A princi- minds: Dual processes and beyond.NewYork, pal component analysis. Manuscript submit- NY: Oxford University Press. ted for publication. Eysenck, H. J. (1995). Creativity as a product of Gigerenzer, G. (2007). Gut feelings: The intel- intelligence and personality. In D. Saklofske ligence of the unconscious. New York, NY: & M. Zeidner (Eds.), International handbook Viking. of personality and intelligence: Perspectives on Gigerenzer, G., & Brighton, H. (2009). Homo individual differences (pp. 231–247). New York, heuristicus: Why biased minds make better NY: Plenum Press. inferences. Topics in Cognitive Science, 1, 107– Feist, G. J. (2001). Natural and sexual selection in 143. the evolution of creativity. BulletinofPsychol- Gigerenzer, G., & Regier, T. (1996). How do we ogy and the Arts, 2, 11–16. tell an association from a rule? Comment on Feldman, J., Kerr, B., & Streissguth, A. P. Sloman (1996). Psychological Bulletin, 119, 23– (1995). Correlational analyses of procedural 26. and declarative learning performance. Intelli- Gilhooly, K. J., Fiortou, E., Anthony, S. H., & gence, 20, 87–114. Wynn, V. (2007). Divergent thinking: Strate- Finke,R.A.,Ward,T.B.,&Smith,S.M.(1992). gies and executive involvement in generating Creative cognition: Theory, research, and appli- novel uses for familiar objects. British Journal cations. Cambridge, MA: MIT Press. of Psychology, 98, 611–625. Fletcher, J., Maybery, M. T., & Bennett, S. (2000). Gladwell, M. (2007). Blink: The power of think- Implicit learning differences: A question of ing without thinking.NewYork,NY:BackBay developmental level? Journal of Experimental Books. Psychology: Learning, Memory, and Cognition, Greenwald, A. G., & Banaji, M. R. (1995). 26, 246–252. Implicit social cognition: Attitudes, self- Fodor, J. (1983). The modularity of mind.Boston, esteem, and stereotypes. Psychological Review, MA: MIT Press. 102, 4–27. Fox, M. C., Roring, R. W., & Mitchum, A. Greenwald, A. G., McGhee, D. E., & Schwartz, L. (2009). Reversing the speed-IQ correla- J. L. K. (1998). Measuring individual differ- tion: Intra-individual variability and atten- ences in implicit cognition: The Implicit Asso- tional control in the inspection time paradigm. ciation Test. Journal of Personality and Social Intelligence, 37(76–80). Psychology, 74, 1464–1480. Frensch, P. A., & Miner, C. S. (1995). Zur Rolle Greenwald, A. G., Poehlman, T. A., Uhlmann, des Arbeitsgedachtnisses¨ beim impliziten E., & Banaji, M. R. (2009). Understanding and P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

464 SCOTT BARRY KAUFMAN

using the Implicit Association Test: III. Meta- Jensen,A.R.(1998). The g factor: The science of analysis of predictive validity. Journal of Per- mental ability. Westport, CT: Praeger. sonality and Social Psychology, 97, 17–41. Jiang, Y., & Chun, M. M. (2001). Selective atten- Haidt, J. (2001). The emotional dog and its ratio- tion modulates implicit learning. Quarterly nal tail: A social intuitionist approach to moral Journal of Experimental Psychology, 54A, 1105– judgment. Psychological Review, 108, 814–834. 1124. Hammond, K. R., & Hamm, R. M., Grassia, J., & Jimenez,´ L., & Mendez, C. (1999). Which atten- Pearson, T. (1987). Direct comparison of the tion is needed for implicit sequence learning? efficacy of intuitive and analytical cognition Journal of Experimental Psychology: Learning, in expert judgment. IEEE Transactions on Sys- Memory, and Cognition, 25, 236–259. tems, Man, and Cybernetics, SMC, 17, 753–770. Johnson,M.K.,&Hirst,W.(1993). MEM: Mem- Hassin, R. R., Uleman, J. S., & Bargh, J. A. (2005). ory subsystems as processes. In A. F. Collins, The new unconscious. New York, NY: Oxford S. E. Gathercole, M. A. Conway, & P. E. University Press. Morris (Eds.), Theories of memory (pp. 241– Hawkins, J. (2005). On intelligence.NewYork, 286). Hillsdale, NJ: Erlbaum. NY: Holt. Jung,C.G.(1921/1971). Psychological types (H. G. Hemphill, J. F. (2003). Interpreting the mag- Baynes, Trans.). Princeton, NJ: Princeton Uni- nitudes of correlation coefficients. American versity Press. (revised by R.F.C. Hull). Psychologist, 58, 78–79. Kahneman, D., & Klein, G. (2009). Conditions Hertzog, C., & Robinson, A. E. (2005). Metacog- for intuitive expertise. American Psychologist, nition and intelligence. In O. Wilhelm & R. W. 64, 515–526. Engle (Eds.), Handbook of understanding and Kahneman, D., & Frederick, S. (2002). Represen- measuring intelligence (pp. 101–123). Thousand tativeness revisited: Attribute substitution in Oaks, CA: Sage. intuitive judgment. In T. Gilovich, D. Grif- Hill, O. W. (1987–1988). Intuition: Inferential fin, & D. Kahneman (Eds.), Heuristics and heuristic or epistemic mode? Imagination, biases: The psychology of intuitive judgment Cognition and Personality, 7, 137–154. (pp. 49–81). New York, NY: Cambridge Uni- Hirschfeld, L. A., & Gelman, S. A. (1994). Map- versity Press. ping the mind: Domain specificity in cognition Kane, M. J., Bleckley, M. K., Conway, A. R. A., and culture. New York, NY: Cambridge Uni- &Engle,R.W.(2001). A controlled-attention versity Press. view of working-memory capacity. Journal Hogarth, R. M. (2005). Deciding analytically of Experimental Psychology: General, 130, 169– or trusting your intuition? The advantages 183. and disadvantages of analytic and intuitive Kaufman, S. B. (2007). Creativity. In C. R. thought. In T. Betsch & S. Haberstroh Reynolds & E. Fletcher-Janzen (Eds.), Encyclo- (Eds.), Routines of decision making (pp. 67–82). pedia of special education (3rd ed.). New York, Mahwah, NJ: Erlbaum. NY: Wiley. Howard-Jones, P. A., & Murray, S. (2003). Kaufman, S. B. (2007). Investigating the role of Ideational productivity, focus of attention, domain general mechanisms in the acquisi- and context. Creativity Research Journal, 15, tion of domain specific expertise. High Ability 153–166. Studies, 18, 71–73. Humphreys, L. G., Lubinski, D., & Yao, G. Kaufman, S. B. (2009a). Beyond general intelli- (1993). Utility of predicting group member- gence: The dual-process theory of human intelli- ship and the role of spatial visualization in gence (Doctoral dissertation). Yale University, becoming an engineer, physical scientist, or New Haven, CT. artist. Journal of Applied Psychology, 78, 250– Kaufman, S. B. (2009b). Faith in intuition is asso- 261. ciated with decreased latent inhibition in a Hunt, E., Frost, N., & Lunneborg, C. (1973). sample of high-achieving adolescents. Psychol- Individual differences in cognition: A new ogy of Aesthetics, Creativity, and the Arts, 3, approach to intelligence. In G. H. Bower 28–34. (Ed.), The psychology of learning and motiva- Kaufman, S. B., DeYoung, C. G., Gray, J. R., tion: Advances in research and theory. Oxford, Brown, J., & Mackintosh, N. (2009). Asso- UK: Academic Press. ciative learning predicts intelligence above James, W. (1890). The principles of psychology. and beyond working memory and processing New, York, NY: Dover. speed. Intelligence, 37, 374–382. P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 465

Kaufman,S.B.,DeYoung,C.G.,Gray,J.R., In N. Colangelo & S. G. Assouline (Eds.), Jimenez,´ L., Brown, J. B., & Mackintosh, N. Talent development IV: Proceedings from the (2009). Implicit learning as an ability. Cogni- 1998 Henry B. and Jocelyn Wallace National tion, 116, 321–340. Research Symposium on Talent Development Keren, G., & Schul, Y. (2009). Two is not always (pp. 219–227). Scottsdale, AZ: Gifted Psychol- better than one: A critical evaluation of two- ogy Press. system theories. Perspectives on Psychological Lubow, R. E. (1989). Latent inhibition and condi- Science, 4, 533–550. tioned attention theory.Cambridge,UK:Cam- Kihlstrom, J. F. (1987). The cognitive uncon- bridge University Press. scious. Science, 237(4821), 1445–1452. Lubow, R. E., & Gewirtz, J. C. (1995). Latent Kihlstrom, J. F., Shames, V. A., & Dorfman, J. inhibition in humans: Data, theory, and impli- (1996). Intimations of memory and thought. cations for schizophrenia. Psychological Bul- In L. M. Reder (Ed.), Implicit memory and letin, 117, 87–103. metacognition. Hillsdale, N.J.: Erlbaum. Lubow, R. E., Ingberg-Sachs, Y., Zalstein-Orda, Klaczynski, P. A. (2009). Cognitive and social N., & Gewirtz, J. C. (1992). Latent inhibition cognitive development: Dual-process research in low and high “psychotic-prone” normal sub- and theory. In S. B. T. Evans & K. Frank- jects. Personality and Individual Differences, 13, ish (Eds.), In two minds: Dual processes and 563–572. beyond. New York, NY: Oxford University Maas,H.L.J.,Dolan,C.V.,Grasman,P.P. Press. P., Wicherts, J. M., Huizenga, H. M., & Rai- Klaczynski, P. A., & Cottrell, J. M. (2004). A dual- jmakers, M. E. J. (2006). A dynamical model process approach to cognitive development: of general intelligence: The positive mani- The case of children’s understanding of sunk fold of intelligence by mutualism. Psycholog- cost decisions. Thinking & Reasoning, 10, 147– ical Review, 113, 842–861. 174. Margolis, H. (1987). Patterns, thinking, and cogni- Klein, G. (1999). Sources of power: How people tion: A theory of judgment. Chicago, IL: Uni- make decisions. Cambridge, UK: Cambridge versity of Chicago Press. University Press. Martindale, C. (1995). Creativity and connection- Lewicki, P., & Hill, T. (1987). Unconscious pro- ism. In S. M. Smith, T. B. Ward, & R. A. cesses as explanations of behavior in cognitive, Finke (Eds.), The creative cognition approach personality, and social psychology. Personality (pp. 249–268). Cambridge, MA: MIT Press. and Social Psychology Bulletin, 13, 355–362. Martindale,C.(1999). Biological bases of creativ- Lewicki, P., Hill, T., & Czyzewska, M. (1992). ity. In R. J. Sternberg (Ed.), Handbook of cre- Nonconscious acquisition of information. ativity (pp. 137–152). Cambridge, UK: Cam- American Psychologist, 47, 796–801. bridge University Press. Lewicki, P., Hill, T., & Sasaki, I. (1989). Self- McGeorge, P., Crawford, J. R., & Kelly, S. W. perpetuating development of encoding biases. (1997). The relationships between psychome- [Empirical Study]. Journal of Experimental tric intelligence and learning in an explicit and Psychology: General, 118, 323–337. an implicit task. Journal of Experimental Psy- Lieberman, M. D. (2007). The X- and C-systems: chology: Learning, Memory, and Cognition, 23, The neural basis of automatic and controlled 239–245. social cognition. Social neuroscience: Integrat- Metcalfe, J., & Mischel, W. (1999). A hot/cool- ing biological and psychological explanations of system analysis of delay of gratiﬁcation: social behavior, 290–315. Dynamics of willpower. Psychological Review, Litman,L.,&Reber,A.S.(2005). Implicit cog- 106, 3–19. nition and thought. In K. J. Holyoak & R. Miller, G. A. (2003). The cognitive revolution: G. Morrison (Eds.), The Cambridge handbook A historical perspective. Trends in Cognitive of thinking and reasoning (pp. 431–453). New Sciences, 7, 141–144. York, NY: Cambridge University Press. Newell, B. R., Wong, K.Y., & Cheung, J. C. H. Loftus, E. F., & Klinger, M. R. (1992). Is the (2009). Think, blink or sleep on it? The impact unconscious smart or dumb? American Psy- of modes of thought on complex decision chologist, 47, 761–765. making. Quarterly Journal of Experimental Psy- Lohman, D. F. (2001). Fluid intelligence, induc- chology, 62, 707–732. tive reasoning, and working memory: Where Nosek, B. A., Greenwald, A. G., & Banaji, M. R. the theory of multiple intelligences falls short. (2007). The Implicit Association Test at age P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

466 SCOTT BARRY KAUFMAN

7: A methodological and conceptual review. Reber, A. S. (1993). Implicit learning and tacit In J. A. Bargh (Ed.), Social psychology and the knowledge: An essay on the cognitive uncon- unconscious: The automaticity of higher mental scious. New York, NY: Oxford University processes (pp. 265–292). New York, NY: Psy- Press. chology Press. Reber, A. S., Walkenfeld, F. F., & Hernstadt, R. Pacini, R., & Epstein, S. (1999). The relation of (1991). Implicit and explicit learning: Individ- rational and experiential information process- ual differences and IQ. Journal of Experimental ing styles to personality, basic beliefs, and the Psychology: Learning, Memory, and Cognition, ratio-bias phenomenon. Journal of Personality 17, 888–896. and Social Psychology, 76, 972–987. Reber, R., Ruch-Monachon, M.-A., & Perrig, Packard, M. G., Hirsh, R., & White, N. M. W. J. (2007). Decomposing intuitive compo- (1989). Differential effects of fornix and cau- nents in a conceptual problem solving task. date nucleus lesions on two radial maze tasks: Consciousness and Cognition: An International Evidence for multiple memory systems. Jour- Journal, 16, 294–309. nal of Neuroscience, 9, 1465–1472. Reyna, V. F. (2004). How people make decisions Payne, J. W., Samper, A., Bettman, J. R., & Luce, that involve risk: A dual-processes approach. M. F. (2008). Boundary conditions on uncon- Current Directions in Psychological Science, 13, scious thought in complex decision making. 60–66. Psychological Science, 19, 1118–1123. Rudman,L.A.,&Ashmore,R.D.(2007). Dis- Peterson, J. B., Smith, K. W., & Carson, S. crimination and the Implicit Association Test. (2002). Openness and extraversion are asso- Group Processes & Intergroup Relations, 10, ciated with reduced latent inhibition: Repli- 359–372. cation and commentary. Personality and Indi- Salthouse, T. A., McGuthry, K. E., & Hambrick, vidual Differences, 33, 1137–1147. D. Z. (1999). A framework for analyzing and Phelps, E. A., O’Connor, K. J., Cunningham, W. interpreting differential aging patterns: Appli- A.,Funayama,E.S.,Gatenby,J.C.,Gore, cation to three measures of implicit learning. J. C., et al. (2000). Performance on indirect Aging, Neuropsychology, and Cognition, 6, 1–18. measures of race evaluation predicts amygdala Schneider, W., & Shiffrin, R. M. (1977). Con- activation. Journal of Cognitive Neuroscience, trolled and automatic human information 12, 729–738. processing: I. Detection, search, and attention. Piaget, J. (1952). The origins of intelligence in chil- Psychological Review, 84, 1–66. dren. Madison, CT: International Universities Seger, C. A. (1994). Implicit learning. Psychologi- Press. cal Bulletin, 115, 163–196. Poincare,´ H. (1921). The foundations of science (G. Simon, H. A. (1979). Models of thought.New B. Halstead, Trans.). New York, NY: Science Haven, CT: Yale University Press. Press. Smith, S. M., Ward, T. B., & Finke, R. A. (1995). Poldrack, R. A., & Packard, M. G. (2003). Com- The creative cognition approach. Cambridge, petition among multiple memory systems: MA: MIT Press. Converging evidence from animal and human Spearman, C. (1904). “General intelligence,” brain studies. Neuropsychologia, 41, 245–251. objectively determined and measured. Ameri- Pretz, J. E. (2008). Intuition versus analysis: Strat- can Journal of Psychology, 15, 201–293. egy and experience in complex everyday prob- Stadler, M. A., & Frensch, P. A. (1997). Handbook lem solving. Memory & Cognition, 36, 554– of implicit learning. Thousand Oaks, CA: Sage. 566. Stanovich, K. (1999). Who is rational? Studies of Pretz, J. E., & Totz, K. S. (2007). Measuring indi- individual differences in reasoning.Mahwah, vidual differences in affective, heuristic, and NJ: Erlbaum holistic intuition. Personality and Individual Stanovich, K. E. (2004). The robot’s rebellion: Find- Differences, 43, 1247–1257. ing meaning in the age of Darwin. Chicago, IL: Pretz, J. E., Totz, K. S., & Kaufman, S. B. (2010). University of Chicago Press. The effects of mood, cognitive style, and cog- Stanovich, K. E. (2009). Distinguishing the reﬂec- nitive ability on implicit learning. Learning tive, algorithmic, and autonomous minds: Is and Individual Differences, 20, 215–219. it time for a tri-process theory. In J. S. B. T. Reber, A. S. (1989). Implicit learning and tacit Evans & K. Frankish (Eds.), In two minds: Dual knowledge. Journal of Experimental Psychol- processes and beyond. Oxford, UK: Oxford ogy: General, 118, 219–235. University Press. P1:PHB Trim:7in × 10in Top: 0.5in Gutter: 0.875in CUUS1280-22 cuus1280/Sternberg ISBN: 978 0 521 51806 2 February 11, 2011 17:57

INTELLIGENCE AND THE COGNITIVE UNCONSCIOUS 467

Stanovich, K. E., & West, R. F. (1997). Reasoning Vartanian, O., Martindale, C., & Kwiatkowski, independently of prior belief and individual J. (2003). Creativity and inductive reasoning: differences in actively open-minded thinking. The relationship between divergent think- Journal of Educational Psychology, 89, 342–357. ing and performance on Wason’s 2–4–6 task. Stanovich, K. E., & West, R. F. (2000). Individ- Quarterly Journal of Experimental Psychology, ual differences in reasoning: Implications for 34, 1370–1380. the rationality debate? Behavioral and Brain Vartanian, O., Martindale, C., & Kwiatkowski, Sciences, 23, 645–726. J. (2007). Creative potential, attention, and Stanovich,K.E.,&West,R.F.(2002). Individual speed of information processing. Personality differences in reasoning: Implications for the and Individual Differences, 43, 1470–1480. rationality debate? In T. Gilovich, D. Grifﬁn, Vartanian, O., Martindale, C., & Matthews, J. & D. Kahneman (Eds.), Heuristics and biases: (2009). Divergent thinking ability is related The psychology of intuitive judgment (pp. 421– to faster relatedness judgments. Psychology 440). New York, NY: Cambridge University of Aesthetics, Creativity, and the Arts, 3, 99– Press. 103. Sternberg,R.J.(1977). Component processes in Wallas (1926). The art of thought. New York, NY: analogical reasoning. Psychological Review, 84, Harcourt Brace. 353–378. Ward, T. B., Smith, S. M., & Finke, R. A. Sternberg, R. J., & Davidson, J. E. (1995). The (1999). Creative cognition. In R. J. Stern- nature of insight. Cambridge, MA: MIT Press. berg (Ed.), Handbook of creativity (pp. 189– Sternberg,R.J.,&Pretz,J.E.(2005). Cognition 212). New York, NY: Cambridge University & intelligence: Identifying the mechanisms of the Press. mind. New York, NY: Cambridge University Wilson, T. (2004). Strangers to ourselves: Discover- Press. ing the adaptive unconscious. Boston, MA: Har- Thorsteinson, T. J., & Withrow, S. (2009). Does vard University Press. unconscious thought outperform conscious Wilson, T. D., Lindsey, S., & Schooler, T. Y. thought on complex decisions? A further (2000). A model of dual attitudes. Psychologi- examination. Judgment and Decision Making, cal Review, 107, 101–126. 4, 235–247. Wilson, T. D., & Schooler, J. W. (1991). Think- Turke-Browne, N. B., Junge, J. A., & Scholl, B. ing too much: Introspection can reduce the J. (2005). The automaticity of visual statistical quality of preferences and decisions. Journal learning. Journal of Experimental Psychology: of Personality and Social Psychology, 60, 181– General, 134, 552–564. 192. Unsworth, N., & Engle, R. W. (2005). Individ- Woolhouse, L. S., & Bayne, R. (2000). Person- ual differences in working memory capacity ality and the use of intuition: Individual dif- and learning: Evidence from the serial reaction ferences in strategy and performance on an time task. Memory & Cognition, 33, 213–220. implicit learning task. European Journal of Per- van Ravenzwaaij, D., van der Maas, H.L., sonality, 14, 157–169. & Wagenmakers, E.J. (in press). Does the Zajonc, R. B. (1980). Feeling and thinking: Pref- name-race implicit association test measure erences need no inferences. American Psychol- racial prejudice? Experimental Psychology, doi: ogist, 35, 151–175. 10.1027/1618-3169/a000093. Zhiyan,T.,&Singer,J.L.(1997). Daydreaming Vartanian, O. (2009). Variable attention facili- styles, emotionality and the Big Five person- tates creative problem Psychology of Aesthetics, ality dimensions. Imagination, Cognition and Creativity, and the Arts, 3, 57–59. Personality, 16, 399–414.