Neural and Cognitive Plasticity: From Maps to Minds

Eduardo Mercado IIIUniversity at Buffalo, The State University of New York

Some species and individuals are able to learn cognitive skills more flexibly than others. Learningexperiences and cortical function are known to contribute to such differences, but the specific factors thatdetermine an organism’s intellectual capacities remain unclear. Here, an integrative framework ispresented suggesting that variability in cognitive plasticity reflects neural constraints on the precision andextent of an organism’s stimulus representations. Specifically, it is hypothesized that cognitive plasticitydepends on the number and diversity of cortical modules that an organism has available as well as thebrain’s capacity to flexibly reconfigure and customize networks of these modules. The author relates thisframework to past proposals on the neural mechanisms of intelligence, including (a) the relationshipbetween brain size and intellectual capacity; (b) the role of prefrontal cortex in cognitive control and themaintenance of stimulus representations; and (c) the impact of neural plasticity and efficiency on theacquisition and performance of cognitive skills. The proposed framework provides a unified account ofvariability in cognitive plasticity as a function of species, age, and individual, and it makes specificpredictions about how manipulations of cortical structure and function will impact intellectual capacity.

Keywords: cognitive neuroscience, comparative cognition, evolution, cortical column, neuromodulation

Intellectual capacity varies as a function of species, age, andindividual (see, e.g., Binet & Simon, 1905/1961; Neisser et al.,1996; Rumbaugh & Washburn, 2003; Spearman, 1904; Thorndike,1911). Behaviorally, this variability is evidenced by differences inthe capacity to learn a cognitive skill, the rate at which the skill islearned, and the highest performance levels that can be achieved(Ackerman & Cianciolo, 2000; Harlow, 1959; Li et al., 2004). Forexample, prodigies may rapidly acquire the ability to computemathematical functions, compose music, or play chess at levelsexceeding those of most other individuals (D. H. Feldman, 1993).Dolphins can rapidly learn to classify any pair of objects as beingeither the same or different (Mercado, Killebrew, Pack, Macha, &Herman, 2000), whereas pigeons have difficulty learning this task(Blaisdell & Cook, 2005; Katz & Wright, 2006), suggesting thatthese two species differ in their capacity to acquire this particularcognitive skill. Why do some individuals or species rapidly learna cognitive skill when others do not? This fundamental unan-swered question lies at the heart of understanding intellectualcapacity.

The answer to this question undoubtedly relates to brain func-tion, but the specific mechanisms through which neural circuitsfacilitate and constrain intellectual abilities remain poorly under-stood. Past efforts to correlate variability in behavioral flexibilityand intelligence with brain structure and function point to at leastthree neural properties that may limit intellectual capacity: (a) the

absolute or relative size of one or more brain regions (Abboitz,1996; Gibson, 2002; Healy & Rowe, 2007; Jerison, 1973, 2002;Roth & Dicke, 2005); (b) the capacity of neural circuits (especiallyin the frontal lobes) to actively maintain and control informationprocessing (Conway, Kane, & Engle, 2003; J. Duncan, 2001; Gray& Thompson, 2004; Koechlin, Ody, & Kouneiher, 2003; E. K.Miller & Cohen, 2001; Petrides, 1996); and (c) the capacity ofneural circuits to physically change their connections over time tomaximize efficiency (Garlick, 2002, 2003; Hebb, 1949; Li, 2003;Li, Brehmer, Shing, Werkle-Bergner, & Lindenberger, 2006;Pascual-Leone, Amedi, Fregni, & Merabet, 2005; Stiles, 2000).Put simply, compact neural circuits may afford less processingpower than can more extensive circuits; less memory capacity mayreduce the range of operations that neural circuits can perform; andhard-wired circuits may be less efficient than adaptable circuits.

In the present article, I attempt to provide an integrative, brain-based framework for investigating and understanding variability incognitive plasticity in humans and other animals by building onclassic theories of learning and intelligence. The brain’s capacityto differentiate representations is specified as the primary deter-minant of which cognitive skills an organism can learn as well ashow long it will take for an organism to learn a particular cognitiveskill. The framework then links this capacity to basic structural andfunctional properties of cortical networks,1 identifying three fac-tors that potentially can account for variability in intellectualcapacity within and across species. Each of these factors is affected

1 The emphasis on cortical function in this article should not be inter-preted as suggesting that animals without cerebral cortex (e.g., birds) areincapable of learning cognitive skills. The current review focuses oncortical processing in mammals because more data are available regardingtheir cognitive capacities and neural function. Neural circuits that serveanalogous functions to cortex may also constrain intellectual capacity innon-mammalian species.

This research was supported by National Institute of Mental HealthGrant MH 67952. I thank Itzel Orduna, Estella Liu, Sean Green, BarbaraChurch, Herbert Roitblat, and David Smith for their helpful comments ondrafts of the article.

Correspondence concerning this article should be addressed to EduardoMercado III, Department of Psychology, Park Hall, University at Buffalo,The State University of New York (SUNY), Buffalo, New York 14260.E-mail: emiii@buffalo.edu

Psychological Bulletin Copyright 2008 by the American Psychological Association2008, Vol. 134, No. 1, 109–137 0033-2909/08/$12.00 DOI: 10.1037/0033-2909.134.1.109

109

by both genetic and environmental variables (including the accu-mulation of knowledge), and each can in principle be experimen-tally manipulated to determine its impact on an organism’s intel-lectual capacity. The proposed explanatory frameworkincorporates processes of evolution, development, and individuallearning and synthesizes findings from the fields of comparativecognition, human intelligence, and systems neuroscience. Theframework accounts for known correlations between neuroanat-omy and measures of intelligence and also predicts additionalcorrelations that have not been considered in past studies. Addi-tionally, the framework makes several novel predictions about theimpacts of various experimental manipulations of brain structureand function on the acquisition of cognitive skills.

The main aim of the article is to provide a broad empirical andtheoretical framework from which specific testable hypothesesabout the mechanisms underlying intellectual capacity can beformulated. The broad scope of this review precludes comprehen-sive consideration and synthesis of all the relevant literature.Instead, I review an assortment of classic and modern empiricalfindings that are either representative of a particular researchdirection or that are particularly pertinent to evaluating the pro-posed framework. I provide references to more extensive, special-ized reviews throughout the article in an attempt to supplementnecessarily abbreviated discussions of topics about which volumeshave been written.

Cognitive Plasticity and Related Constructs

Researchers often use the term cognitive plasticity in discus-sions of cognitive development during early childhood and cog-nitive vitality in old age to refer to the modifiability of cognitionby social interactions and training experiences (Kramer, Bherer,Colcombe, Dong, & Greenough, 2004; Li, 2003). In the presentcontext, I develop a broader conception of this psychologicalconstruct that encompasses a wider range of learning experiencesand intelligence measures and that generalizes across species.

Many organisms can adapt their behavior on the basis of expe-rience. William James (1890) described the capacity that enableshumans to form “new sets of habits” or skills as plasticity. Skilllearning can be classified into two types: perceptual–motor andcognitive (Anderson, 1982; Fitts, 1964; Johnson-Laird, 1982;Rosenbaum, Carlson, & Gilmore, 2001). Whereas perceptual–motor skill learning requires adjusting motor responses on thebasis of sensory feedback, cognitive skill learning involves acquir-ing the ability to solve a problem in a way that was not previouslypossible, without necessarily requiring any new perceptual–motorskills (Rosenbaum et al., 2001; VanLehn, 1996). Cognitive skillscan thus be defined as abilities that an organism can improvethrough practice or observational learning and that involve judg-ments or processing beyond what is involved in learning or per-forming a perceptual–motor skill. For example, learning to solvethe Tower of Hanoi puzzle is challenging not because it requireslearning to place disks on pegs (a perceptual–motor skill) butbecause it requires the discovery of effective strategies for movingdisks to achieve a particular outcome (a cognitive skill). Followingthe terminology of James, the capacity to acquire cognitive skillscan be described as cognitive plasticity. Cognitive plasticity en-ables individuals to learn to solve problems in intellectual tasks—success is determined by the appropriate application of knowledge

rather than dexterity. This construct can be operationalized as therate and proficiency of cognitive skill learning (Figure 1).

Humans and other animals differ in their ability to acquirecognitive skills and to generalize them to novel contexts. Histor-ically, such differences have been attributed to variability in intel-ligence. Binet and Simon (1905/1961, p. 93) defined intelligenceas the “faculty of adapting one’s self to circumstances,” and someresearchers view the ability to flexibly solve problems as synon-ymous with intelligence (Hofman, 2003; Macphail, 1982). If in-telligence is viewed as a general capacity to acquire and applyknowledge (Jensen, 1989), then intelligent behavior is little morethan a demonstration of cognitive skills (Eysenck, 1982), andcognitive plasticity is conceptually equivalent to intelligence. Mostmodern conceptions of intelligence are derived from an individu-al’s performance on psychometric tests, however (for a review, seeNeisser et al., 1996), and most of these tests only indirectlymeasure cognitive skill learning (e.g., focusing on how problem-solving skills are applied rather than on how they are acquired).Consequently, in the following discussion, the term intelligence isused to describe an individual’s ability to perform cognitive skills,as indicated by scores on behavioral tests.

One recently proposed mechanism for differences in humanintelligence (and by proxy cognitive plasticity) is neural plasticity(Garlick, 2002). Neural plasticity refers to the capacity of neuralcircuits to change in response to fluctuations in neural or glialactivity. This phenomenon is typically associated with changes insynaptic connections between individual neurons but also includesprocesses such as the addition of new neurons (neurogenesis),increased myelination of axons, or changes in the size or shape ofa neuron. In a recent review discussing cortical plasticity in thehuman brain, Pascual-Leone and colleagues (2005) argued that“plasticity is an intrinsic property of the nervous system retainedthroughout a lifespan” and that “it is not possible to understandnormal psychological function or the manifestations or conse-quences of disease without invoking the concept of brain plastic-ity” (p. 378). Accordingly, Garlick (2002, 2003) proposed that ifbrains vary in the ability to adjust neural connections based on

Figure 1. Idealized cognitive skill acquisition curves for individualsvarying in cognitive plasticity. High cognitive plasticity is associated withrapid learning and high performance levels. These three curves couldreflect learning by different species, by different individuals varying inintelligence, or by an individual at different points in his or her lifespan.Dotted line � low plasticity; dashed line � moderate plasticity; solidline � high plasticity.

110 MERCADO

experience, then those individuals with higher levels of neuralplasticity might have greater intellectual capacity. Other research-ers have suggested that genetic factors impacting neural efficiencyor plasticity in the lateral prefrontal cortex determine individualdifferences in intelligence (Gray & Thompson, 2004; Shaw et al.,2006).

The framework presented below similarly assumes that neuralplasticity contributes to cognitive plasticity and intelligence butonly to the extent that reorganization increases the brain’s capacityto resolve stimulus representations. The term stimulus representa-tion is used here to refer to neural activity evoked either by sensoryreceptors or by the initiation of movements and thoughts. Stimulusrepresentations indicate that particular environmental and internalstates have occurred (or are about to occur), and they thus repre-sent those states. The present framework proposes that an organ-ism’s representational resolving power constrains what that organ-ism can learn about events. In particular, an individual that cannotdistinguish two stimulus representations cannot learn to responddifferentially to the events that are associated with those represen-tations. Consequently, an organism’s cognitive plasticity is limitedby the capacity of its brain to resolve stimulus representations,hereafter referred to as representational resolution. Furthermore,the current framework proposes that individual differences incognitive plasticity arise because representational resolution variesacross individuals and as a function of age.

Like many prior attempts to account for intellectual capacity, thepresent framework assumes that cortical networks are critical.Rather than focusing on increases in computational power orexecutive control functions afforded by cortical expansion, how-ever, the current framework emphasizes a basic building block ofcortical circuits: cortical modules. The term cortical module isused in neuroscience to describe anatomical and electrophysiolog-ical regularities in cortical circuits (Leise, 1990; Mountcastle,1998; Zaborszky, 2002). Specifically, iteratively repeated net-works of cortical neurons appear to contain compact circuits thatcan function independently or together; the term cortical module,as used in this article, refers to one of these compact circuits.Because cortical modules are structural in nature (and thereforeobservable), they do not correspond to the functionally specialized“mental modules” proposed by Fodor (1983). Although there is anot a specific neural computation or function that all corticalmodules are known to perform, there are abundant data showingthat cortical neurons in at least some modules respond selectivelyto sensory inputs and that in several cortical regions the spatialorganization of cortical modules reflects stimulus selectivity (seeMountcastle, 1998, for a review). These data suggest that somecortical modules facilitate the differentiation of sensory events.The present framework extends this idea to suggest that, in gen-eral, cortical modules facilitate the differentiation of stimulusrepresentations. Given that the framework also proposes that cog-nitive plasticity depends on representational resolution, variabilityin cortical modules that occurs as a function of an individual’sspecies, age, or heredity should be associated with variability inthat individual’s cognitive plasticity and intelligence.

The framework identifies three factors that likely determine theimpact of cortical modules on representational resolution: theiravailability, reconfigurability, and customizability. Availability re-fers to the number and diversity of cortical modules that areavailable for differentiating stimulus representations. This factor,

which is closely related to the complexity of cortical circuits,constrains the range of possible stimulus representations in waysthat are analogous to how pixel density and color depth limit thecrispness and precision of images displayed on a computer mon-itor. Reconfigurability refers to the brain’s ability to flexibly de-velop new configurations of cortical modules, and to rapidlyswitch between them, as a function of task demands. This factorimpacts the functional versatility of neural circuits; for example,different subsets of available cortical modules may be engagedduring different stages of learning. Customizability refers to thebrain’s capacity to dynamically adjust the selectivity of corticalmodules based on experience. This factor captures the effects thatneural plasticity mechanisms may have on the number and diver-sity of cortical modules. These three factors are interdependentbecause the range of circuit configurations depends on the numberand specificity of modules, and reconfiguring or customizing cor-tical modules can affect their availability.

Theoretical Precursors to the Framework

The present framework builds on several classical theories ofthe mechanisms underlying learning and intelligence: Harlow’s(1949, 1959) theory of learning sets; Spearman’s (1904) theory ofgeneral intelligence and general discrimination, and Hebb’s (1949,1980) theory of cell assemblies. These theories have greatly influ-enced modern ideas about the mechanisms of intellectual capacityas well as the impact of experience on these mechanisms. Bycurrent standards, the theories might appear overly simplistic andinadequate to deal with the mass of data that has been collected byneuroscientists, intelligence researchers, and cognitive psycholo-gists over the last 50 years. Nevertheless, the theories identify keyprinciples that can still provide important insights into the factorsthat constrain an organism’s intellectual abilities.

Harlow’s Learning Set Theory

Harlow’s (1949) discovery that monkeys trained on series ofnovel visual discrimination problems got progressively better atsolving new problems challenged Thorndike’s (1911) claims thatanimals learned only through trial and error. After extensive ex-perience learning to select one image of a pair for many differentpairs, monkeys eventually were able to determine which of the twoimages would be rewarded after performing a single trial with anew pair; that is, if they initially chose the correct image on thefirst trial, they would continue to choose it in subsequent trials—otherwise, they would choose the alternative. Harlow (1949) de-scribed this process as “learning to learn,” or forming a learningset. Harlow’s (1949) findings suggested that monkeys were notlimited to learning about the particulars of the world but that theycould also figure out general strategies for solving problems.Furthermore, cross-species correlations between performance onlearning set tasks and behavioral flexibility seemed to indicate thatmeasures of learning set formation could provide an objective testof animal intelligence (Harlow, 1959).

Following Harlow’s (1959) lead, comparative psychologists be-gan to use cognitive plasticity (operationally defined as rate ofacquisition and asymptotic performance on learning sets) as ameasure of animal intelligence (e.g., Doty, Jones, & Doty, 1967;Schrier, 1966). Specifically, the species that developed learning

111NEURAL AND COGNITIVE PLASTICITY

sets the fastest and to the highest levels of performance wereconsidered to be the most intelligent. The objectivity of thesebehavioral measures also made it possible to correlate them withvarious measures of brain volume. For example, Riddell and Corl(1977) compared rankings of different species derived from sev-eral brain metrics with rankings based on learning set performance.They reported a near-perfect correlation (r � .87–.98), suggestingthat brain metrics provide the best possible predictor of differencesin intellectual capacity across species.

Critics were quick to note, however, that both the brain metricsand the learning set performance measures used in such analysesinvolved simplistic assumptions that did not always fit the knownfacts (Macphail, 1982). For instance, rankings based on eitherrelative brain size or learning set performance have resulted insome non-primates being ranked higher than some monkeys. Ad-ditionally, measures of learning set formation have not provided anunequivocal indication of intelligence because differences in per-formance may reflect differences in sensory, motor, or motiva-tional factors rather than differences in problem-solving abilities(Bitterman, 1965, 1975; Hodos & Campbell, 1969; Lefebvre,1996; Macphail, 1982; Slotnick, Hanford, & Hodos, 2000;Thomas, 1996). For example, bottlenose dolphins initially showedlittle ability to form learning sets for an auditory discriminationtask (Herman, Beach, Pepper, & Stalling, 1969), but minor pro-cedural changes later revealed learning capacities comparable withthose of primates (Herman & Arbeit, 1973). Such interpretationalcomplications ultimately squelched attempts to rank different spe-cies in terms of their intelligence on the basis of measures oflearning set performance and decreased interest in learning setformation.

Traditional learning set tasks are clearly limited as cross-speciesintelligence metrics. Nevertheless, learning set studies still canprovide important clues about how cognitive plasticity varieswithin and across species (Sutherland & Mackintosh, 1971). Todevelop a learning set, an organism must adjust how it approachessolving a problem. For instance, Harlow (1959) described learningacross a series of visual discrimination tasks as involving a gradualshift from an initial period of trial and error learning to a phase ofinsight learning in which the learner adopts a generalized rule:win–stay, lose–shift. From this perspective, learning to learn mightbe more accurately described as learning to recognize, becausewhat an individual ultimately learns is when to apply certainstrategies. Most explicit studies of learning set formation haveinvolved series of discrimination tasks like those used by Harlow,but there are many other experiments in which animals havelearned general strategies for solving problems (see Roitblat & vonFersen, 1992, for a review). For example, animals have learned toclassify objects as same or different (Premack, 1983a), to assignnumerals to quantities (Brannon & Terrace, 1998), and to repeatactions on command (Mercado, Murray, Uyeyama, Pack, & Her-man, 1998). Cognitive skill learning tasks such as these whichrequire animals to learn “abstract” rules likely involve learning setformation. The future development of more sophisticated learningset tasks may help to overcome the limitations associated withclassical approaches in the same way that development of humanintelligence tests has improved their utility.

The idea that learning set formation may be relevant to a widerange of learning phenomena in humans and other animals (ini-tially proposed by Harlow, 1959) has resurfaced in theories of

human learning and problem solving. For example, Halford andcolleagues (Halford, Bain, Maybery, & Andrews, 1998) proposedthat learning set acquisition often involves forming a representa-tion of task structure and mapping that structure onto novel prob-lems (much like analogical inference). From this perspective,learning sets are types of general knowledge abstracted frompersonal experiences that guide choices and actions. In this moregeneral framework, learning set formation involves recognizingthe constraints and relevant dimensions of a problem space, iden-tifying the relevant options for responding within that problemspace, and acquiring the skills (cognitive and perceptual–motor)necessary to achieve a satisfactory outcome (Adolph, 2002).

Harlow (1959) proposed that insightful learning and perfor-mance, as well as all concepts, develop only through the formationof learning sets. He showed experimentally that the capacity ofmonkeys and children to form learning sets increased systemati-cally as a function of their age, that their acquisition rate wascorrelated with asymptotic performance levels (as in Figure 1), andthat similar variability in acquisition patterns was apparent acrossprimate species (Harlow, 1949; 1959). He also showed that brain-injured monkeys that were missing cortex from one hemisphereacquired learning sets like normal monkeys, but with an accuracythat was consistently lower than normal (Harlow, 1949). Thesedata led him to conclude that

there is a capacity factor or factors which influence speed of learningset formation, asymptote of maximally efficient learning set perfor-mance, and even the ability to form effective learning sets. The factoror factors may also determine the minimal number of trials perproblem essential to establish the learning set, and probably determinethe maximal complexity of problems amenable to learning set forma-tion. (Harlow, 1959, p. 508)

He attributed this capacity factor or factors to cortical complexity.

Spearman’s Theory of General Intelligence and GeneralDiscrimination

Intelligence has long been associated with the ability to performwell in school. Spearman (1904) showed that different assessmentsof academic performance share a common source of variance,which he described as general intelligence (also referred to as g;Neisser et al., 1996). He also reported that measures of generalintelligence were highly correlated with measures of general dis-crimination capacities, substantiating Galton’s (1883) earlierclaims of a correspondence between intellectual capacity and sen-sitivity to fine sensory differences. Galton suggested that acuitywas the source of intellectual capacity because “the more percep-tive the senses are of difference, the larger is the field upon whichour judgment and intelligence can act” (Galton, 1883, p. 19). Incontrast, Spearman attributed this correspondence to a commonunderlying cause, which he described as the “Intellectual Func-tion.”

Other studies during this same period, however, refuted the ideathat perceptual acuity was related to intelligence (e.g., Wissler,1901; for a review, see Deary, 1994). Furthermore, researcherswere quick to note cases such as Helen Keller, in which keenintellect developed despite severely impaired sensory capacities.As a result, interest in the relationship between perceptual acuityand intelligence quickly waned.

112 MERCADO

A century later, confirmations of Spearman’s (1904) reportsregarding a link between intelligence and discrimination abilitieshave begun to appear, and Wissler’s (1901) results have comeunder increasing scrutiny (Deary, 1994). These recent studies havesuggested that constraints on intellectual capacities directly paral-lel, or are identical to, constraints on sensorimotor processes(Deary, 1995; Deary, Bell, Bell, Campbell, & Fazal, 2004; Helm-bold, Troche, & Rammsayer, 2006; Lindenberger & Baltes, 1994;Lindenberger, Scherer, & Baltes, 2001; Raz, Willerman, & Yama,1987; Rosenbaum et al., 2001; Watson, 1991). For example, Dearyand colleagues (2004) statistically compared several measures ofgeneral intelligence with measures of sensitivity to differences incolor and loudness and found that a general intelligence factorshared 85% of its variance with a general discrimination factor,replicating Spearman’s results.

Unlike Harlow (1949, 1959), Spearman viewed the effects oflearning on intellectual performance as an “irrelevancy” that had tobe eliminated to identify an individual’s “natural innate faculties”(Spearman, 1904, p. 227). In part, Spearman adopted this approachbecause

if (as stated by many persons on the strength apparently of a priorireasoning) we had assumed that the discrimination of minute differ-ences of sensation ‘is to be cultivated as the foundation of all intel-ligence,’ then we should have had to admit the variations due toPractice as perfectly relevant and we should have looked for a con-tinual expansion in people’s general ability in proportion to the laborthey had expended on distinguishing tones, shades, and weights fromone another. (Spearman, 1904, p. 228)

In short, Spearman was interested in measuring individual differ-ences in intellectual talent, whereas Harlow was interested inmeasuring how organisms acquire cognitive expertise. Despitethese different perspectives, both men arrived at the same conclu-sion: Some fundamental underlying factor constrains an individu-al’s intellectual capacity.

Hebb’s Theory of Cell Assemblies

Hebb (1949) effectively synthesized the views of Harlow (1949,1959) and Spearman (1904) by proposing that the term intelligencehas two different meanings. Hebb (1949) described “IntelligenceA” as referring to an individual’s innate potential and “IntelligenceB” as referring to a person’s intellectual performance. FromHebb’s (1949) perspective, both Harlow (1949) and Spearmanwere using tests of Intelligence B to make inferences about Intel-ligence A. The main question that Hebb (1949) was interested inanswering, however, was not how to measure intelligence orlearning capacity but how a patient could maintain performance onintelligence tests after having his or her prefrontal lobes or rightcortical hemisphere surgically removed. The explanation he pro-posed was that Intelligence B is a function of the concepts that apatient has developed through experience and that these conceptsare resilient to cortical damage because they correspond to theactivity of large assemblies of interconnected cells (neurons) inwhich many independent pathways can serve equivalent functions(Hebb, 1949, 1980).

Hebb’s (1949) theory of cell assemblies not only provided apossible explanation for the persistence of intellectual abilitiesafter extensive cortical damage but also for the correlations be-

tween sensory acuity and intelligence reported by Spearman(1904). Specifically, the theory suggested that even simple per-ceptual capacities depended on cortical assemblies organizedthrough sensory experiences and that thought processes corre-sponded to interactions between such experience-generated corti-cal assemblies. In Hebb’s words,

perception in the early stages is a consequence of a primitive learningprocess, but thereafter learning becomes a function of perception – thelearning, that is, that is most characteristic of the adult human being– and a function of cognitive structures that originate in perception.(Hebb, 1980, p. 98)

Thus, the cell assembly theory provided two potential sources forthe observed correlations between acuity and intelligence. First,assemblies underlying concepts were hypothesized to arise fromthe assemblies underlying percepts, as intimated by both Galton(1883) and Harlow (1959). Second, Hebb (1949) hypothesized thatall cell assemblies were constructed by dynamically modifyingcortical interconnections as a function of experience, that is, that asingle neural process leads to the formation of both concepts andpercepts. Consequently, an individual’s intrinsic capacity to reor-ganize cortical circuits could constrain both perceptual and intel-lectual abilities. Hebb’s (1949) ideas about cortical assemblies andthe role of experience in cortical organization continue to stronglyinfluence theories of cortical function (e.g., Harris, 2005) but havehad less impact on current explanations of the neural mechanismsunderlying intelligence.

An Integrative Framework for Understanding theRelationship Between Neural and Cognitive Plasticity

None of the theoretical viewpoints summarized above weredeveloped to explain differences in cognitive plasticity acrossindividuals or species. Nevertheless, they collectively provide auseful perspective from which to identify the mechanisms thatdetermine such differences. Harlow’s (1949) theory emphasizesthe role of trial and error learning in concept formation and howorganisms differ in their ability to acquire concepts as a function oftheir age and species. Hebb’s (1949) theory also emphasizes therole of learning in intellectual capacity but focuses on the progres-sive learning and organization of percepts rather than on associa-tive learning of stimulus–response relationships. Both Harlow(1949) and Hebb (1949) pointed to cortical complexity as a keydeterminant of what an organism can learn. Spearman’s (1904)theory treats the effects of age and experience as confounds thatcan strongly impact measures of intellectual performance, therebyinterfering with measurements of individual differences in intelli-gence. His ideas are similar to those of Hebb’s, however, in that hepostulates a close link between an individual’s perceptual andintellectual capacities. Neither Harlow (1949) nor Hebb (1949)attempted to account for human individual differences in theirtheories, instead focusing on larger differences in intellectual ca-pacity associated with gross differences in brain structure.

All three theorists recognized that intellectual capacity is dy-namic and that certain global factors (e.g., age and cortical com-plexity) normally constrain the range of abilities that an individualcan acquire. Hebb (1949), in particular, was one of the first toemphasize that age-dependent changes in both perceptual and

113NEURAL AND COGNITIVE PLASTICITY

intellectual abilities likely reflect experience-dependent neuralplasticity within cortical networks. Harlow (1949) and Hebb(1949) were also both impressed by the fact that massive damageto the cortex of adults often did not lead to dramatic deficits inintellectual capacity, suggesting that cortical networks were eitherhighly redundant or functionally equipotential. Integrating theseoverlapping theoretical ideas into a common neurally based frame-work provides a way to more fully assess their explanatory powerand to further their development.

The framework described below suggests that representationalresolution constrains an individual’s capacity to learn new cogni-tive skills and that an individual’s ability to resolve stimulusrepresentations largely depends on the functional properties of thecortical modules that he or she possesses. To a certain extent, theframework can be viewed as replacing Hebb’s (1949) assembliesof generic brain “cells” with assemblies of diversified corticalmodules that are customized through evolution and individuallearning to differentiate the changing states of an organism’sinternal and external environment. As noted earlier, the frameworkidentifies three properties of cortical modules that likely impactcognitive plasticity. First, it is proposed that the availability ofcortical modules (their number and diversity) limits cognitiveplasticity. This neural constraint is related to issues of brain size,because cortical volume limits the maximum number and diversityof cortical modules. Second, it is suggested that cognitive plastic-ity entails the ability to flexibly develop new circuit configurations

and to switch between them, that is, that assemblies of corticalmodules are reconfigurable. This neural constraint focuses on aparticular representational control process among several that maymaintain and control stimulus representations, and it thus builds oncurrent ideas about the role of cognitive control in intellectualcapacity. Finally, it is hypothesized that the extent to which net-works of cortical modules can be customized (dynamically re-tuned) constrains cognitive plasticity; such retuning depends onmechanisms of neural plasticity.

The neural substrates that provide the foundation for the explan-atory framework and their hypothesized roles are shown in Figure2. I have already briefly described how the availability, reconfigu-rability, and customizability of cortical modules may constrain anindividual’s representational resolution. The following sectionsreview evidence showing how the structural and functional orga-nization of cortical and subcortical networks relates to these threeproperties of cortical modules and that the capacity of corticalcircuits to resolve stimulus representations may strongly determinethe differences in cognitive plasticity seen within and across spe-cies and individuals. It is important to note that, although thecurrent explanatory framework is motivated by numerous empir-ical findings from neuroscience research, it has yet to be system-atically investigated in relation to differences in intellectual capac-ity within and across individuals or species, and thus it remainsspeculative.

Figure 2. Schematic illustration of how cortical and subcortical networks may participate in the differentiationof stimulus representations.

114 MERCADO

Representational Resolution: Cortical ModulesDifferentiate Stimulus Representations

Cognitive skill learning undoubtedly involves cortical process-ing (Pauli et al., 1994; Poldrack, Prabhakaran, Seger, & Gabrieli,1999; Raichle, 1998; Reichle, Carpenter, & Just, 2000). Prefrontalactivity, interactions between cortical and subcortical regions, andthe overall size and structure of cortical regions can thus stronglyimpact cognitive plasticity. Evolutionary considerations suggestfurther that the expansion of cortical networks reduces constraintson cognitive plasticity (Gibson, 2002; Sternberg & Kaufman,2002). One way that cortical processing may enhance cognitiveplasticity is by expanding an organism’s capacity to representinternal and external events. This could be achieved either byextending the temporal window within which events can be pro-cessed or by providing multiple representations of a single event inparallel (analogous to photos taken at various angles and magni-fications). The availability of multiple “looks” at an event canprovide an individual with more choices. Longer looks providemore opportunities for contingencies to reveal themselves and thuscan also be viewed as additional representations of an event. Toassess whether cortical networks expand an organism’s represen-tational resolution, it is important to consider how the structure ofthese networks varies across species.

Differentiation through progressive segregation is a ubiquitousfeature of sensory processing circuits in the brain (Ebbesson, 1984;Hofman, 2003; Kaas & Collins, 2001; Metzner & Juranek, 1997).On an evolutionary time scale, one sees the emergence of discretesensory modalities as well as the division of unimodal sensorysystems into parallel, somewhat redundant circuits with increas-ingly selective processing capacities (D. A. Clark, Mitra, & Wang,2001; Krubitzer & Kaas, 2005; Northcutt & Kaas, 1995; Rakic,1988; Striedter, 2006). At a developmental scale, relatively dif-fusely connected sensory cortical networks that respond to a widerange of stimuli give way to more selective, precisely connectedsystems that are “sensitized” based on environmental conditions,especially during the so-called sensitive periods (Bence & Levelt,2005; Johnson & Vecera, 1996; Stiles, 2000; Zhang et al., 2001).Finally, as individual organisms learn from their experiences, onesees further refinement of cortical sensitivities and dynamic topo-graphic reorganization of cortical networks (Buonomano & Mer-zenich, 1998; Weinberger, 2004). This progression suggests a pushtoward neural mechanisms that enable greater differentiation andrefinement of stimulus representations.

Inputs from sensory receptors are typically remapped multipletimes in the mammalian central nervous system; different map-pings afford different stimulus–response configurations (Metzner& Juranek, 1997). Cortical extent and the number of corticalneurons thus appear to constrain an organism’s capacity to differ-entiate stimulus representations. For example, a cat’s ability togeneralize learning about one tone to other tones of differentfrequencies can be predicted from the topography of frequencysensitivities in the cat’s auditory cortex (R. F. Thompson, 1965;see also Orduna, Mercado, Gluck, & Merzenich, 2005). In asimilar vein, the spatial topography of auditory and somatosensorycortical sensitivities in dogs can be inferred simply from observinghow they generalize conditioned responses (Pavlov, 1927). Thenumber of segregated functional maps in cortex also predicts anorganism’s capacity to differentiate stimulus representations. For

instance, echolocating bats possess a larger number of specializedfeature maps in auditory cortex, and primates have numerousspecialized maps in visual cortex (see also R. O. Duncan &Boynton, 2003). Thus, the structural features of cortical networksdetermine the separability of stimulus representations, therebyconstraining what an organism can learn. Recent studies of audi-tory processing in humans similarly have suggested that corticalnetworks are critical for segregating auditory scenes and that suchstimulus differentiation is intimately linked to more general intel-lectual abilities (Naatanen, Tervaniemi, Sussman, Paavilainen, &Winkler, 2001).

Evidence of a correspondence between discriminative capacitiesand more complex learning abilities is ubiquitous in comparativeresearch. For example, the rate at which learning sets are acquiredis predicted by the rate of discrimination learning in early sessions(Harlow, 1949, 1959). Additionally, stimulus salience, which re-flects an organism’s sensitivities to different stimuli, is a keyparameter in several theories of animal learning (e.g., Mackintosh,1975; Rescorla & Wagner, 1972). These theories identify salienceas a major determinant of learning rate and asymptotic perfor-mance in a wide range of learning tasks.

Cognitive aging research also has revealed links between per-ceptual abilities, intellectual capacity, and cognitive plasticity. Forexample, the Berlin aging study found that sensory and sensori-motor variables predicted 59% of the reliable variance in measuresof general intelligence (Baltes & Lindenberger, 1997; Linden-berger & Baltes, 1997; Lindenberger, Scherer, & Baltes, 2001).Several hypotheses have been proposed to account for this corre-spondence, including the ideas that sensory deficits cause drops inIQ (Anstey & Smith, 1999) and that global changes in neuralprocessing efficiency affect both processes (Lindenberger &Baltes, 1994). The current framework integrates these two hypoth-eses by suggesting that neural processing capacity is constrainedby representational resolution.

The relationship between representational resolution and cogni-tive plasticity hypothesized in the current framework differs fromthe relationship between perceptual acuity and intelligence sug-gested by Galton (1883) and Spearman (1904) in that specificcausal mechanisms (features of cortical networks) are proposed toaccount for this relationship. Because these mechanisms relate tobrain structure, cases like Helen Keller’s, in which sensory deficitsare dissociated from intelligence, are not problematic. Corticalmodules will continue to be available despite major losses ofsensory inputs, and therefore sensory deficits should minimallyimpact the brain’s overall representational resolving power. Notsurprisingly, the framework predicts that cognitive skills that re-quire information from a lost sensory modality will be impossibleto learn since the brain will be unable to resolve the relevantexternal events. Note that unlike Spearman’s theory, the presentframework does not focus on discrimination abilities per se butinstead on the brain’s capacity to differentiate stimulus represen-tations. Such differentiation includes not only discriminating sen-sory events but also recognizing and identifying internal andexternal events.

Several researchers have pointed to dissociations between par-ticular measures of perceptual acuity and measures of specificintellectual capacities as evidence against a close link betweenacuity and intelligence (e.g., see Acton & Schroeder, 2001;Thorndike, Lay, & Dean, 1909). Such evidence is not contradic-

115NEURAL AND COGNITIVE PLASTICITY

tory to the current framework because the framework presumesthat different subsets of cortical modules mediate the learning andperformance of different cognitive skills. Thus, the frameworkpredicts that perceptual distinctions made with a greater numberand diversity of cortical modules will be better predictors of anindividual’s cognitive plasticity than will distinctions requiringlittle or no cortical processing. However, the framework does notpredict that an individual’s ability to make any given perceptualdistinction will be strongly correlated with his or her performanceon intelligence tests.

Availability: The Number and Diversity of CorticalModules Constrains Capacity

Size is a dominant factor in the organization of cortical net-works. More space means more room for connections and thepossibility of larger cell bodies. More neurons mean more possi-bilities for cell differentiation. Larger brains necessitate longerconnections between cortical regions, but this in turn allows formore precise projections across regions. For example, animalswith larger cortical regions will be better able to separate circuitsinvolved in different aspects of sensorimotor coordination. Thetopography of cortical sensitivities and the cytoarchitectonic vari-ability across cortical regions (e.g., Brodmann’s areas) provideevidence that there is such a structural division of labor (see Figure3). Sensory cortices, in particular, show disproportionate allocationof cortical regions to the processing of dense regions of receptorspace (Allman, 1998; Catania & Kaas, 1995). Interestingly, vari-ation in the topography of cortical sensitivities across individualscan be as large or larger than variation across closely relatedspecies (Merzenich, 1985). Inputs from other brain regions cominginto the cortex are perpendicular to the cortical surface and areorganized into columns, which reduces terminal overlap and in-creases the possibility of functional segregation. The six layers ofneurons within cortex also allow for finer sensory processing(Swanson, 2003) and more expandable topographic structure(Striedter, 2006). Finer grained topography in cortical networksincreases opportunities for precise decomposition of inputs; pre-cision necessitates (or at least benefits from) topographical orga-nization. Thus, the most important role of increasing cortical sizeis to reduce constraints on cortical organization and, in particular,constraints on the differentiation of inputs.

Columnar geometry is a defining feature of cortical architecture(Hubel & Wiesel, 1963; Mountcastle, 1957, 1997, 1998). Colum-nar organization varies predictably across individuals and species(Buxhoeveden & Casanova, 2002), with the size of cortical mac-rocolumns reflecting the distribution of thalamic inputs (Stevens,2001, 2002). Macrocolumns are discrete, polygonal structures(roughly hexagonal) consisting of varying numbers of minicol-umns (ca. 60–80), each of which may contain 80–100 neurons(Favorov & Diamond, 1990). Macrocolumns in cat visual cortexare significantly larger than those in rhesus monkeys, but themonkey’s cortical columns contain more cells. So, the monkey’slarger brain is associated with smaller but more complex minicol-umns. In primates, the complexity of columnar composition invisual cortex is associated with higher visual acuity. However, inmost cases it is unclear how variations in the structure of corticalcolumns relate to functional capacity either across or within spe-

cies. Groups of cortical columns may function together as a co-herent unit; cortical modules are one such unit.

The number of cortical modules in the mammalian brain hasincreased through evolution (Krubitzer & Kaas, 2005). This evo-lutionary expansion relates directly to the number of minicolumns(Rakic, 1995). The number of columns developed by an organismreflects the number of progenitor cells early in development (Ra-kic, 1988); the evolutionary expansion of cortex may similarlydepend on the number of divisions that are possible (Rakic, 1995).From this perspective, cortical modules containing essentially thesame circuits are increased in number through duplication (Allman& Kaas, 1971; Kaas, 1984). Consistent with this idea, multiple“redundant” maps of sensory surfaces with similar topographiesare seen in the cortices of all vertebrates (Metzner, 1999). Simi-larly, subcortical sensory pathways that provide inputs to sensorycortices reveal parallel decomposition (Swanson, 2003). For ex-ample, the mammalian cochlear nucleus divides auditory inputsinto six parallel streams (Young, 1997). Evolutionary processesclearly select for cortical expansion, but duplication and segrega-tion of cortical modules alone does not appear to be sufficient forgenerating functional advances. Differentiation of cortical modulesis also necessary.

Two evolutionary mechanisms have been proposed to accountfor the divergent properties of cortical modules and associatedincreases in the number of cortical subdivisions. Kaas (1984)proposed that replicated cortical modules sharing a basic structuralorganization became specialized across generations. In contrast,Ebbesson (1980, 1984) postulated a parcellation process in whichthe selective loss of inputs to cortical regions ultimately producedmore specialized cortical subdivisions. In both Kaas’s (1984) andEbbesson’s (1980, 1984) frameworks, unitary cortical processingmodules divide (through mechanisms analogous to either mitosisor meiosis, respectively) to create multiple modules with functionssimilar to, but more specialized than, the original module. Forexample, in “primitive” marsupials, cortical regions responsive todifferent sensory modalities overlap with each other and withmotor cortex, whereas in more advanced marsupials, these areasare more segregated (Abboitz, 1996). Similarly, primary sensorycortices in primates are physically farther apart than in othersmaller mammals and are surrounded by a larger number ofspecialized secondary sensory fields. Subcortically, greater sen-sory segregation in the thalamic nuclei of frogs relative tosalamanders is associated with an increased ability to discriminatemoving configurations (Ewert, 1984). Animals with highlyevolved sensory capacities (such as echolocating bats and electro-locating fish) possess the most segregated and functionally spe-cialized sensory systems (Covey, 2005; Metzner & Juranek, 1997).The evolutionary isolation and differentiation of cortical modulesthus appears to facilitate finer distinctions between stimuli, leadingto more appropriate behavior and more selective motor control.

It is less clear how the number and diversity of functionalcortical modules vary across the lifespan. However, the expansionand shrinkage of cortex during early development and aging re-spectively suggest that the number of cortical modules mightsimilarly change. Ontogeny recapitulates phylogeny in that colum-nar organization in fetal tissue is quite uniform. As individualsensory systems develop, there is selective pruning and sharpeningof cortical response features. Surgically manipulating inputs tosensory cortex early in development can generate novel structural

116 MERCADO

Figure 3. “Cytoarchitectural studies of the lateral surfaces of the cerebral cortices of a series of primates andan insectivore, arranged by E. G. Jones in descending order of size by reference to the magnifications given byeach of the authors in their original articles. At the top is the map for man, by Brodmann (1912); the sequencefrom the above left to below right, in order: orangutan, gibbon, and macaque (Mauss, 1910), then lemur and aninsectivore (Brodman, 1914)” (Mountcastle, 1998, p. 24). Different patterns on the surface of each brain showcortical regions that neuroanatomists have classified as having distinctive cellular features (e.g., the distributionof neuronal types within columns varies across areas). Note that the number of distinctive regions decreases asthe size of the brain and cortex decreases. From Perceptual Neuroscience: The Cerebral Cortex (p. 24), by V. B.Mountcastle, 1998, Cambridge, MA: Harvard University Press. Copyright, 1998 by the President and Fellowsof Harvard College. Reprinted with permission.

117NEURAL AND COGNITIVE PLASTICITY

organizations of cortical modules, demonstrating that environmen-tal demands can increase the diversity of modules available (Kahn& Krubitzer, 2002; Sharma, Angelucci, & Sur, 2000). Similarly,sensory deprivation during development leads to greater overlap incortical connections and less selectivity to inputs (Chang & Mer-zenich, 2003; Zhang et al., 2001, 2002), suggesting that the num-ber and capacity of cortical modules also vary during developmentdepending on an organism’s experiences. It appears that as corticalmodules develop they become progressively more selective,thereby increasing an organism’s ability to differentiate stimulusrepresentations.

Currently, there is only indirect evidence that the number anddiversity of cortical modules constrain intellectual capacity. This ispartly because it is difficult to manipulate these variables whilemeasuring the resulting effects on cognitive plasticity. Acrossspecies, the number and diversity of cortical modules are corre-lated with the size of cortex (Jerison, 2002). Consequently, previ-ously identified correlations between intellectual capacities andcortical extent also apply, at least in part, to cortical modules.Across individuals, Harlow’s (1949) experiments, which showedthat monkeys with less cortex form learning sets more slowly, aswell as Lashley’s (1929) studies, which showed that a rat’s capac-ity to learn to navigate a maze is correlated with the amount ofcortical tissue remaining after surgery, provided some support forthe idea that the absolute number of available cortical modules isan important determinant of learning capacity (see also R. Thomp-son, Crinella, & Yu, 1990).

Reconfigurability: Cognitive Plasticity Depends onFlexible Neural Circuitry

Most researchers now agree that different brain regions servedifferent roles at different stages of cognitive skill learning (Crone,Donohue, Honomichl, Wendelken, & Bunge, 2006; Doeller, Opitz,Krick, Mecklinger, & Reith, 2006; Pauli et al., 1994; Poldrack etal., 1999). In many cases, performance increases are correlatedwith decreased activity in some brain regions and increased activ-ity in others (Raichle, 1998). The involvement of any particularsubset of brain regions in cognitive skill acquisition depends ontask difficulty, an individual’s level of expertise, and the particulartask being performed. In some cases, subcortical structures such asthe basal ganglia, hippocampus, or cerebellum seem to be partic-ularly important, with different regions showing the most activityat different stages of acquisition (Poldrack et al., 2001; Poldrack etal., 1999; Raichle, 1998). In other cases, cortical networks in thefrontal lobes or parietal lobes show the greatest involvement (Pauliet al., 1994). To further complicate matters, the specific set ofcortical regions that becomes active during acquisition depends onthe specific strategies that an individual uses during learning(Reichle et al., 2000). This particular finding suggests that, evenwithin the same cognitive skill learning task, different neuralcircuits could mediate similar performances.

At first glance, the reconfigurability of cortical circuits mightseem to suggest that cortical modules are functionally interchange-able or equipotential. However, there is substantial evidence thatthe cortical modules within these distributed circuits serve uniquespecialized functions (e.g., Crone, Wendelken, Donohue, &Bunge, 2006). Mesulam (1998) has described the capacity forbrain regions to dynamically shift affiliations from one functional

network to another as selectively distributed processing. The cir-cuits involved at any given stage of learning may represent “pathsof least resistance;” that is, strategy selection may normally in-volve selecting the most efficient circuits (Reichle et al., 2000).

A core hypothesis of the present framework is that cognitiveplasticity depends on selectively distributed processing. Specifi-cally, the framework presumes that specialized cortical modulesthroughout the brain can be used in a variety of combinations (andsomewhat interchangeably) to enable the acquisition and perfor-mance of cognitive skills, as well as that variability in cognitiveplasticity across individuals of different species and ages reflectsvariability in their capacity to reconfigure cortical modules. Al-though the mechanisms underlying dynamic circuit reconfigura-tion are not well understood, lampreys and even crustaceans showsome capacity to rapidly reconfigure their neural circuits (Ayers,Carpenter, Currie, & Kinch, 1983; Harris-Warrick & Marder,1991), suggesting that this may be a fundamental feature of neuralprocessing across many species.

Evidence indicating that individuals with greater intellectualcapacity may be able to more flexibly configure their neuralcircuits has come primarily from electroencephalographic (EEG)studies in humans (Jausovec & Jausovec, 2000; Thatcher, North, &Biver, 2005). For example, children with higher IQs show reducedbackground EEG coherence, indicating more differentiated activ-ity across cortical regions (Gasser, Jennen-Steinmetz, & Verleger,1987). Conversely, low IQ adults show less differential neuralactivity across brain regions when performing a sentence verifica-tion task than do individuals with higher IQs (Neubauer,Freudenthaler, & Pfurtscheller, 1995), and older individuals withlower IQs show less evidence of switching between brain regionsduring cognitive skill learning (Grady, Springer, Hongwanishkul,McIntosh, & Winocur, 2006). Although such correlational studieshave not shown that differences in cortical reconfigurability lead todifferences in cognitive plasticity, they at least establish that thishypothesis is consistent with the known facts and suggest futureexperiments that could be conducted to establish a causal link (seebelow).

Customizability: Neuromodulatory Circuits ModulateCortical Differentiation

So far, I have emphasized the role that cortical structure anddynamics can play in differentiating stimulus representations andin constraining cognitive plasticity. It is well known, however, thatevolutionarily older subcortical systems ultimately control howcortical networks respond to all events. At the most basic level,subcortical regions modulate cortical responsiveness to inputsbased on the sleep–wake cycle. When an organism is awake,subcortical, neuromodulatory systems containing neurons thatproject throughout cortex continuously gate processing of infor-mation. Several of these systems are known to impact the responseproperties of neurons in sensory cortex. Among these, the basalforebrain is thought to be particularly important for the activation,formation, and maintenance of stimulus representations. Neuronsin the basal forebrain are the main source of acetylcholine incortical networks and also are a source of GABA (gamma-aminobutyric acid, an inhibitory neurotransmitter) and variousneuropeptides that can impact cortical activity (Zaborszky, Pang,Somogyi, Nadasdy, & Kallo, 1999). A wide range of data suggests

118 MERCADO

that basal forebrain neurons affect not only how cortical networksrespond to particular events but also how and when learningexperiences change cortical sensitivities (for recent reviews, seeSarter, Hasselmo, Bruno, & Givens, 2005; Suga & Ma, 2003;Weinberger, 2003; Zahm, 2006). Thus, subcortical, neuromodula-tory systems that modulate activity levels in cortex appear to becritical to determining how experience remodels cortical architec-ture and how events are represented.

The idea that connections between cortical neurons may in-crease in strength or efficiency as a function of experience hasbeen around since the first observations that such connections exist(Ramon y Cajal, 1904; Yuste & Bonhoeffer, 2001). Historically,this form of neural plasticity was used mainly to explain howassociative links between external stimuli and reflexive responsesare created (Pavlov, 1927) as well as how memories are stored(Hebb, 1949). It was thus somewhat of a surprise when researchersWiesel and Hubel (1963) discovered that experience stronglyimpacted sensitivities in developing sensory cortical networks (seeBence & Levelt, 2005; and Stiles, 2000, for recent reviews), andeven more of a surprise when similar changes were observed inadult sensory cortex after injuries or learning experiences (re-viewed by Buonomano & Merzenich, 1998; Calford, 2002; Cooke& Bliss, 2006; D. E. Feldman & Brecht, 2005; Kaas, 1991; Wall,Xu, & Wang, 2002). The capacity of cortical networks to changeas a function of experience is called cortical plasticity.

A ubiquitous finding in studies of developmental and learning-related cortical plasticity is that the distribution of sensitivitiesacross sensory maps is systematically altered on the basis of thesensory events that the organism experiences. Often, this involvesexpansion of the cortical region that responds selectively to abehaviorally relevant stimulus (e.g., Recanzone, Schreiner, & Mer-zenich, 1993); that is, the number of cortical modules that respondto the stimulus increases. Importantly, the size of the corticalregion that is responsive to sensory inputs is a good predictor ofthe organism’s ability to differentiate those inputs, as caricaturedin the iconic, big-lipped, human somatosensory cortex homuncu-lus. Experiences with stimuli can also increase the selectivity withwhich cortical neurons respond to a particular stimulus (Polley,Kvasnak, & Frostig, 2004), effectively increasing the precision ofavailable modules. Both kinds of changes may increase an organ-ism’s capacity to differentiate stimuli. Basal forebrain neuronslikely contribute to cortical plasticity because both cortical expan-sion and retuning can be generated by externally controlling basalforebrain activity during the presentation of sensory events (Bakin& Weinberger, 1996; Kilgard & Merzenich, 1998; Mercado, Bao,Orduna, Gluck, & Merzenich, 2001).

The roles that basal forebrain neurons are thought to play incortical plasticity and sensory processing have been instantiated ina variety of qualitative (Suga & Ma, 2003; Weinberger et al., 1990)and computational models (Mercado, Myers, & Gluck, 2001; Yu& Dayan, 2005). The basic idea is that neuromodulators (espe-cially acetylcholine) released from basal forebrain projections tocortical neurons enhance responses to sensory inputs by increasingthe probability that the inputs will generate action potentials. Thisin turn facilitates neural plasticity such that future responses tothose particular inputs will also be enhanced. Collectively, theseexperience-dependent neuromodulatory effects are thought to in-crease the ability of sensory cortical networks to efficiently pro-cess behaviorally relevant stimuli.

One reason to expect that basal forebrain modulation might bean important factor in cognitive plasticity is that this region isextensively innervated by neurons in the prefrontal cortex, whichis a cortical region often associated with variability in intelligence(e.g., Gray & Thompson, 2004; Shaw et al., 2006). The connec-tions between prefrontal cortex and the basal forebrain appear tobe organized into large numbers of specialized channels, such thatspecific regions of the prefrontal cortex project to predictable basalforebrain regions (Zaborszky, 2002; Zahm, 2006). These projec-tions constitute distributed parallel circuits enabling multilevelconcurrent processing that may provide a way for prefrontal cir-cuits either to modulate the selectivity and plasticity of corticalresponses to ongoing stimulus representations (Sarter, Hasselmo,Bruno, & Givens, 2005; Zaborszky, 2002) or to modulate learningbased on estimates about the uncertainty of ongoing events (Yu &Dayan, 2005). Processing information through parallel channels inthe sensory cortices, prefrontal cortex, and basal forebrain couldincrease cognitive plasticity by enhancing the selective adjustmentof components of stimulus representations (i.e., customizing cor-tical modules) or by facilitating rapid switching between process-ing modes (i.e., reconfiguring cortical circuits) when circum-stances change (Mesulam, 1990; Sarter, Bruno, & Turchi, 1999).In this way, the real-time, situation-dependent customization andreconfiguration of cortical modules may emulate evolutionary,developmental, and learning-dependent mechanisms for differen-tiating stimulus representations—all of these processes involvedifferentiation through segregation.

Whereas the relationship between basal forebrain modulationand sensory cortical plasticity is well established, evidence directlyrelating cognitive plasticity to either cortical reorganization orbasal forebrain function is sparse and open to interpretation. On anevolutionary scale, animals with the most expansive cortices (e.g.,humans and cetaceans), typically possess the largest basal fore-brain regions (Marino et al., 2001; Zaborszky et al., 1999) andoften show greater flexibility at learning cognitive skills (Marino,2002). Additionally, basal forebrain damage in rats impairs theirability to form learning sets in a discrimination reversal task,suggesting that this neuromodulatory system is important for“learning to learn” (Bailey & Thomas, 2001). In humans withAlzheimer’s disease, loss of basal forebrain projections is associ-ated with cognitive decline (e.g., see Fibiger, 1991; Wenk, 1997;Whitehouse et al., 1982), but it is less clear how these deficitsimpact cognitive skill learning or intellectual capacity. Generally,it is difficult to interpret the behavioral and cognitive effects ofbasal forebrain damage because this region influences so manydifferent neural systems and because no studies have been con-ducted directly relating the activity of basal forebrain neurons tothe performance or learning of cognitive skills.

Explanations of Variability in Intellectual Capacity: Pastand Present

Three neural mechanisms often proposed to account for differ-ences in intellectual capacity across individuals and species arebrain size, prefrontal cortex function, and brain speed or efficiency(for reviews, see Deary, 2000; MacPhail, 1982; Mackintosh,1998). In human studies, correlations between IQ and frontal lobeactivity/structure have taken center stage as researchers attemptedto identify unique circuits in the prefrontal cortex that enhance

119NEURAL AND COGNITIVE PLASTICITY

intellectual capacities (J. Duncan, 2001; Gray & Thompson, 2004;Koechlin et al., 2003; Shaw et al., 2006). This neuropsychologicalapproach has largely overshadowed research relating global dif-ferences in processing speed, efficiency, and adaptability to indi-vidual differences in intelligence (however, see Deary & Caryl,1997; Garlick, 2002; Haier et al., 1992). Studies of cross-speciesdifferences in intelligence traditionally focused on differences inneural processing power, as estimated from the size of variousstructures in the brain (Jerison, 1973; Rensch, 1956). Each of theseapproaches has clarified the mechanisms underlying intellectualcapacity, but none has provided a full account of how variations inneural circuits give rise to variations in cognitive plasticity.

What follows is a brief review of past evidence linking neuralmechanisms to intellectual capacity. This review is selective, fo-cusing primarily on evidence that bears on the current proposalthat cognitive plasticity varies as a function of the availability,reconfigurability, and customizability of cortical modules. Severalrecent articles (Deary & Caryl, 1997; Gibson, 2002; Gray &Thompson, 2004; Jerison, 2002; McDaniel, 2005; Roth & Dicke,2005) and books (e.g., Deary, 2000; Mackintosh, 1998; Rumbaugh& Washburn, 2003) provide more extensive reviews of the evi-dence collected to date. The integrative framework presented hereextends these past perspectives, accounting for known correlationsbetween brain anatomy and measures of intelligence as well assome data that have complicated interpretation of these correla-tions. A key difference of the current approach from most pastefforts is that it makes specific predictions about how environmen-tal conditions, learning, and knowledge acquisition can impact theneural mechanisms underlying cognitive plasticity. These predic-tions, some of which are presented in the following sections,suggest that both previously discarded and new approaches tounderstanding intellectual capacity can contribute to futureprogress.

Availability: The Role of Brain Size and Structure

Scientists and philosophers have related brain characteristics tocognitive capacities for several millennia. Historically, qualitativeimpressions of intellectual capacities were correlated with increas-ingly precise measurements of brain tissue (Jerison, 1973; Rensch,1956). Larger brain regions generally provide room for morecomplex circuitry, more dendritic expansion, more synapses,thicker myelin, more neurons, and larger neurons (Deary & Caryl,1997), all of which could increase functional capacity. The fol-lowing discussion reviews recent attempts to relate differences inbrain size to differences in intelligence. Collectively, these studiesprovide strong support for the idea that brain size in some wayconstrains cognitive plasticity.

Intelligence in nature: Innovations and social complexity. Re-cent attempts to correlate brain features with intellectual capacityacross species often have involved testing closely related specieson tasks that reflect their behavior in the wild and then relatingdifferences in performance to differences in the relative size ofparticular brain regions (reviewed by Sherry, 2006). The premiseunderlying such research is that enlargement of particular brainregions provides an organism with selective advantages whenfaced with specific ecological challenges. For example, largerhippocampal regions are thought to give certain species of birds anadvantage in spatial tasks related to foraging (Biegler, McGregor,

Krebs, & Healy, 2001; Volman, Grubb, & Schuett, 1997), and alarger than normal cerebellum mediates the sophisticated electro-location abilities of certain fish (Heiligenberg, 1990). From thisperspective, variability in relative brain region size across speciesreflects differences in adaptive specializations.

Unlike earlier studies of learning set acquisition, newer researchoften has focused on measuring performance across species, mak-ing it difficult to directly relate the results of these studies tovariations in cognitive plasticity. For instance, across several dif-ferent orders of birds, the relative size of a bird’s forebrain ispredictive of the likelihood that a particular species will engage ininnovative foraging tactics (Lefebvre, Whittle, Lascaris, & Finkel-stein, 1997); that is, bigger forebrains predict greater behavioralflexibility (see also Ratcliffe, Fenton, & Shettleworth, 2006;Reader & Laland, 2002; Sol, Duncan, Blackburn, Cassey, & Le-febvre, 2005). However, innovation is a subjective measure thatincludes everything from eating an atypical food item (e.g., humanvomit) to using an automatic sensor to open a bus station door(Lefebvre, Whittle, Lascaris, & Finkelstein, 1997), and the extentto which different birds depend on reflexive actions versus learnedperceptual–motor or cognitive skills during such foraging anom-alies is essentially unknown. Consequently, little can be said at thisstage about the relationship between cognitive plasticity and inno-vative foraging behavior.

The most extensive application of this new approach has been instudies of primate behavior. Unlike most other neuroecologicalstudies of intellectual abilities in animals, studies of primate cog-nition have explicitly focused on understanding how intelligenceevolved in humans—an unabashedly anthropocentric approach(Byrne, 1995; Byrne & Whiten, 1988; Corballis, 2007; Corballis &Lea, 1999; Sternberg & Kaufman, 2002). In primates, the relativeand absolute sizes of cortex are high compared with those of otherbrain regions (D. A. Clark et al., 2001; Dunbar, 1992). Differencesin the relative size of brain regions (especially the cortex) havebeen correlated with a wide range of abilities in primates thoughtto require high behavioral flexibility, including deception (Byrne& Corp, 2004), social complexity (Dunbar, 2003; Lindenfors,2005), tool use, social learning, and innovative foraging (Reader &Laland, 2002). The goal of such analyses has been to test thehypothesis that human intelligence evolved as an adaptive special-ization for dealing with complex social situations. Unfortunately,the gross nature of the brain–behavior comparisons used in thesestudies makes them difficult to interpret (Bolhuis & Macphail,2001; Healy & Rowe, 2007; Holekamp, 2007). It thus remainsunclear whether the observed differences in behavior reflect dif-ferences in cognitive plasticity or other factors, such as the relativeeffectiveness of foraging techniques in different species.

One approach to addressing limitations in neuroecological stud-ies has been to measure additional brain features (e.g., neuronaldensity, myelination) that might more closely relate to the mech-anisms that underlie differences in intellectual capacity (Gibson,2002; Roth & Dicke, 2005). Pragmatic and ethical issues limit theextent to which such approaches can be used to compare brainsacross different primate species. However, recent technologicaland scientific advances have made it possible to precisely relatedifferences in brain tissue to standardized measures of intellectualcapacity in the one species for which such standardized psycho-metric measures exist: humans.

120 MERCADO

Size differences in humans. As noted earlier, most modernconceptions of human intelligence are derived from measures ofindividual performance on psychometric tests (for a review, seeNeisser et al., 1996). Typically, these tests only indirectly measurecognitive plasticity, focusing instead on how individuals applyproblem-solving skills and knowledge. Recent dynamic testingprocedures do measure cognitive skill acquisition, however (Gri-gorenko & Sternberg, 1998). Research comparing performance onstandard intelligence tests with performance on these dynamictests has shown a strong positive relationship (Embretson, 1992;Embretson & Prenovost, 2000). Thus, measures of an individual’sIQ seem to provide a good indication of an individual’s cognitiveplasticity.

Recent analyses based on structural MRI measurements of braintissue, as well as postmortem measures of volume, have shownsignificant positive correlations between brain volume and variousIQ measures (Flashman, Andreasen, Flaum, & Swayze, 1997;Gignac, Vernon, & Wickett, 2003; Jensen & Sinha, 1993; Mc-Daniel, 2005; G. F. Miller & Penke, 2007; Witelson, Beresh, &Kigar, 2006), suggesting that humans with bigger brains are morelikely to score higher on IQ tests. In addition to definitivelyestablishing this basic relationship, structural MRI analyses sug-gest that men and women differ in terms of how the physicalfeatures of their brains relate to IQ measures. For example, inwomen the volume of white matter is more strongly correlatedwith IQ, whereas in men gray matter volume shows the strongestcorrelations (Haier, Jung, Yeo, Head, & Alkire, 2005). Further-more, this technique makes it possible to compare intellectualcapacity with variations in the size of different brain regions(Colom, Jung, & Haier, 2006; Haier, Jung, Yeo, Head, & Alkire,2004; Wilke, Sohn, Byars, & Holland, 2003).

Consistent positive correlations between brain size and intellec-tual capacity, within and across species, indicate that levels ofcognitive plasticity may be constrained by the size and structure ofseveral different brain regions. The present explanatory frameworksuggests that size-related changes in both cortical and subcorticalnetworks drive such correlations. In particular, it hypothesizes thatlarger brains afford a larger number and greater diversity ofcortical modules as well as more options for configuring andcustomizing these modules. This in turn increases the organism’scapacity to differentiate stimulus representations, thereby reducingconstraints on cognitive skill learning.

Past ideas about why bigger brains are often better have focusedon the amount of “extra” cortical tissue that certain large-brainedspecies possess (Jerison, 1973) or on the overall computationalpower as estimated from the number of neurons or synapsesavailable (Roth & Dicke, 2005). In contrast, the current frameworksuggests that the amount of tissue and neurons an organism pos-sesses is less critical than how the circuits within that tissue areorganized and that it is resolving power that determines intellectualability rather than general information processing capacity. A keycorollary of this idea is that experience-dependent changes in theresponse features of cortical modules can impact how well differ-ent stimulus representations can be differentiated, which can inturn affect an organism’s intellectual capacity.

Future directions and predictions: Beyond volumetric compar-isons. Many recent studies correlating brain size and intellectualcapacity discount the relevance of absolute brain size and focusinstead on relative measures (e.g., ratio-based brain metrics such as

brain weight/body weight and neocortex volume/brain volume).This bias is driven partly by allometric considerations, partlybecause some non-primates have bigger brains than those of pri-mates and partly because relative brain size measures are excellentpredictors of species differences in behavioral flexibility (Healy &Rowe, 2007). The predictive value of ratio-based brain metricsdoes not imply, however, that they are more informative than areabsolute measures (Marino, 2006), and there are surprisingly fewconvincing explanations as to why the relative distribution ofdifferent tissues within a brain should predict intellectual capacitybetter than will the absolute amount of each kind of tissue. Jerison(1973) suggested that a fixed cortical mass has fixed informationprocessing capacities and that greater configurational complexity(which should vary as a function of the absolute amount of cortex)provides for greater behavioral plasticity. Gibson (2002) similarlyargued that the enlargement of the cortex, cerebellum, and basalganglia provided humans with increased information processingcapacities. The assumption that absolute measures of brain size areless relevant than relative measures is further called into questionby recent data showing that equal increases in the sizes of differentbrain regions do not always correspond to equal increases in thenumber of constituent neurons (Herculano-Houzel, Mota, & Lent,2006). In other words, constant size ratios across brain regions donot imply fixed levels of neural complexity, making it difficult toevaluate ratio-based brain metrics independently of absolute brainsize.

Global brain metrics have often been criticized as a grossoversimplification that ignores what is known about the organiza-tion of the brain (Healy & Rowe, 2007; Mace, Harvey, & Clutton-Brock, 1980), and more recent studies often instead measure thesize of particular brain regions. This approach does not address thecriticism, however, since brain regions invariably can be subdi-vided into yet smaller regions. Furthermore, measurements ofbrain region size are often expressed relative to the size of otherbrain regions, limiting their usefulness in the same ways notedabove for relative measures of brain size. Perhaps more fundamen-tally, most existing volumetric measures of brain tissue do notreveal the impacts of experience on an organism’s brain. Thus,these measures predict that if two chimpanzees possess equalamounts of brain tissue (or neurons, or synapses), then they willpossess equivalent intellectual capacities, even if one of the chim-panzees lived in a lightless crate its whole life, whereas the otherhad extensive learning experiences in an enriching environment.These limitations suggest the need for new neuroanatomical mea-sures that more directly reflect the mechanisms underlying intel-lectual capacity.

The present framework presumes that two brains could beidentical in terms of their absolute or relative size but could differgreatly in terms of the number and diversity of cortical modules.Furthermore, two brains with a similar number and diversity ofcortical modules could still confer different levels of cognitiveplasticity depending on how selectively neuromodulatory neuronsin subcortical regions can enhance cortical differentiation of stim-ulus representations. Thus, bigger brains are often predictive ofhigher levels of cognitive plasticity, but additional brain indicesbeyond size (particularly structural measures from the cortex andthe basal forebrain) should provide better estimates of cognitiveplasticity. The framework predicts that species with large brainsbut a limited ability to learn cognitive skills (e.g., cows) possess

121NEURAL AND COGNITIVE PLASTICITY

cortical modules that are more homogeneous (and/or less numer-ous) than those of species with high cognitive plasticity. Similarly,it predicts that variations in brain size within species, such as isseen across genders in humans, or in different breeds of dogs, neednot strongly constrain intellectual abilities because the absolutesize of cortical modules can potentially co-vary with brain size;that is, two differently sized brains can potentially contain acomparable number and diversity of cortical modules. Finally, theframework predicts that the number and diversity of cortical mod-ules will vary with experience and that such differences will bereflected in the spatial distribution and selectivity of cortical col-umns. If so, then measures of cortical topography can potentiallyprovide a way to differentiate an impoverished chimpanzee’s brainfrom that of an enriched chimp.

To the extent that cortical maps constrain what is perceptuallyand behaviorally possible (and consequently what is learnable),functional measures of cortical topography may prove to be auseful alternative index of cognitive plasticity and intelligenceacross individuals and species. Of course, it currently is impracti-cal to functionally map cortical sensitivities in most individuals, soindirect methods of assessing cortical sensitivities, such as thebehavioral techniques developed by Pavlov (1927), may be nec-essary. For example, species differ in their behavioral thresholdsfor resolving auditory frequencies, called auditory frequency dif-ference limens (Fay, 1974). Mammals with lower thresholds fordiscriminating differences between frequencies usually have largernumbers of auditory cortical fields with more selective responseproperties (Northcutt & Kaas, 1995). Although quantitative com-parisons of frequency difference limens and measures of cognitiveplasticity across species have never been made, it is interesting tonote that those mammals that are most sensitive to frequencydifferences (primates and delphinids) are also the mammals thatform learning sets most rapidly. This suggests that indirect behav-ioral measures of an organism’s psychophysical thresholds orcortical sensitivities may predict that organism’s cognitive skilllearning capacities at least as well as measures of tissue volume. Inaccord with the proposals of Galton (1883) and Spearman (1904),the present framework predicts that measures of some perceptualthresholds (which reflect representational resolution of externalstimuli) should be predictive of an individual’s cognitive plasticityand intellectual capacity. Future work directly examining correla-tions between psychophysical thresholds, cognitive plasticity, andvarious anatomical and physiological brain metrics is needed toevaluate these predictions within and across species.

Currently, the lack of objective techniques for measuring intel-lectual capacity in non-humans almost necessitates an anthropo-centric focus. The classical learning set approach developed byHarlow (1959), which initially seemed to solve this problem, fellout of favor in part because differences in sensorimotor capacitiesacross species often confounded the interpretation of cross-speciesdifferences in learning. If cognitive plasticity is constrained by anorganism’s capacity to resolve stimulus representations, however,then this turns such critiques of learning set research on theirheads. Specifically, differences in the rate of learning set formationmay actually result from differences in representational resolutionrather than from methodological artifacts. In other words, organ-isms may behave most intelligently, and learn cognitive skills mosteffectively, when they are able to best differentiate the stimulusrepresentations that they need to respond to or learn about. For

example, rats show little ability to form learning sets when re-quired to discriminate visual stimuli (Warren, 1965) but showabilities comparable with those of some primates when tested witholfactory stimuli (Slotnick et al., 2000). A major advantage oflearning set tasks as a measure of cognitive plasticity is that theyprovide information about the dynamics of cognitive skill acqui-sition as well as performance indices.

Another advantage of learning set tasks is that, with furtherdevelopment, they potentially can be used to study differences inintellectual capacity not only across species but also across indi-viduals and throughout the lifespan. Studies of cognitive plasticityin human aging already make use of this approach. For example,training elderly individuals to improve memory recall through theuse of mnemonics involves the formation of learning sets. Similarresearch with aging animals tested on traditional learning setmeasures has shown striking parallels with research on aginghumans (Boutet et al., 2005; Harlow, 1959; Itoh, Izumi, & Kojima,2001; Rumbaugh, 1970; Rumbaugh & McCormack, 1967). Mea-sures of individual differences in human intelligence that dependon measures of learning capacity (e.g., dynamic tests of IQ; Gri-gorenko & Sternberg, 1998) may also prove useful for futurestudies of age-related changes in cognitive plasticity.

New learning set tasks may be particularly useful for measuringindividual differences in the intellectual capacities of non-humans.Comparative researchers have rarely attempted to systematicallymeasure variability in intelligence within a species (however, seeR. Thompson et al., 1990; Thorndike, 1935), although individualdifferences in learning capacity are noted in many comparativeresearch reports. Measures of individual differences in learning setformation within species would avoid many of the confoundsassociated with cross-species comparisons of learning set forma-tion and could facilitate analyses of the neural mechanisms under-lying cognitive plasticity (Yokoyama, Tsukada, Watanabe, &Onoe, 2005). Research on individual differences in intellectualcapacity in non-humans can also assess the generality of theoret-ical explanations for such differences that have been derived fromhuman studies. A basic prediction of the present framework is thatindividual differences in cognitive plasticity can be related tovariability in cortical structure and physiology, independent ofspecies. Future research focused on improving quantitative mea-sures of cognitive plasticity in non-humans is essential to testingthis prediction.

Correlational studies can provide only weak tests of the mech-anisms underlying intellectual capacity, even with more refinedmeasures of intellectual capacity and brain structure, because onecan never infer causation from correlation. New measures ofcognitive plasticity in non-humans could make it possible to di-rectly test how various brain features impact cognitive plasticitybecause neural variables can be more flexibly controlled in non-humans. In particular, techniques are available for (a) varying thesize of brain regions; (b) varying the number and diversity ofcortical modules; (c) genetically and surgically engineering chi-mera; (d) varying the rate of an organism’s development andaging; (e) temporarily disengaging discrete brain regions; (f) ma-nipulating neural plasticity mechanisms; and (g) controlling basalforebrain activity. Additionally, artificial selection can be used togenerate strains of animals of a particular species that differ intheir perceptual and cognitive abilities. For example, cultural do-mestication has generated a wide variety of dogs, with large

122 MERCADO

differences in brain size and “adaptive specializations” (e.g.,bloodhounds vs. sheep dogs). In the laboratory, rodents have beenbred to solve mazes well or poorly (Heron, 1935; Plomin, 2001),and flies have been bred to possess specific learning deficits(Waddell & Quinn, 2001). Genetic engineering has also been usedto create mice with heightened learning capacities (Chen et al.,2003; Tang et al., 1999). These artificially diversified populationsof individuals provide unique opportunities for examining therelationship between neural structure and intellectual capacity (seealso R. Thompson et al., 1990).

The current framework suggests that if animals are selectivelybred or engineered to increase the diversity and number of corticalmodules, then this should increase their cognitive plasticity. Ifthese features are selectively changed in subregions of the brain,then this could selectively increase the capacity to learn a subset ofcognitive skills; for example, a dog breed might be developed thathas exceptional numerical capacities. Obviously, not all cognitiveskills can be studied in all species. A basic assumption of theframework, however, is that different cognitive skills depend onfundamentally similar modes of cortical processing and are con-strained in similar ways. Comparative experiments can thus clarifyhow variability in cortical modules contributes to cognitive plas-ticity and intellectual capacity as well as the relationship betweenabsolute brain size, structural complexity, and behavioral flexibil-ity.

Reconfigurability: Maintenance and Control ofRepresentations

In the last century, one cortical region in particular has becomestrongly associated with the superior intelligence of humans: thefrontal cortex. The frontal lobes are currently considered by manyto be the seat of cognitive control (Koechlin et al., 2003; Wood &Grafman, 2003), supporting humans’ intellectual capacitiesthrough the flexible coordination of decision processes and mem-ory (Blair, 2006; J. Duncan, 2001; Pasternak & Greenlee, 2005).The relative size of the human frontal lobes in comparison withthat of most other mammals is sufficient evidence for many toconclude that this brain region may be a unique source of humanintellectual capacity. There are many theories about how corticalcircuits in the frontal lobes contribute to intellectual capacity (e.g.,Blair, 2006; J. Duncan, 2001; Fuster, 2001; E. K. Miller & Cohen,2001; West, 1996), none of which questions the idea that frontalcircuits are key to intellectual performance.

Localizing intelligence: A preoccupation with prefrontal cortex.Structural MRI measures have linked differences in the volume ofgray matter in frontal cortex to measures of general intelligence(Haier et al., 2005; P. M. Thompson et al., 2001), suggesting thatthe makeup of frontal circuits constrains intellectual capacity(Gray & Thompson, 2004). Consistent with this interpretation,complex changes in the thickness of frontal cortex during devel-opment are also correlated with changes in IQ (Shaw et al., 2006).Frontal lobe activity (especially in the prefrontal cortex) also tendsto be associated with measures of intelligence (e.g., Gray, Chabris,& Braver, 2003; Gray & Thompson, 2004). For example, in onepositron emission tomography study that compared cortical bloodflow as participants performed three tasks used to measure intel-ligence versus three control tasks, the greatest relative increase inblood flow occurred in the lateral prefrontal cortex for all three

IQ-related tasks (J. Duncan et al., 2000). Some researchers haveargued that the consistent association of parts of the frontal cortexwith measures of intelligence is evidence for a highly specializedfrontal system that provides humans with the ability to flexiblycontrol their cognitive acts (Bar-On, Tranel, Denburg, & Bechara,2003; Koechlin et al., 2003).

On the other hand, although regions within the frontal cortexoften show the largest correlations with measures of intelligence instructural imaging studies, the volume of several other cortical andsubcortical regions is also predictive of IQ (Andreasen et al., 1993;Colom et al., 2006). For example, gray matter volume within thetemporal lobe is positively correlated with measures of IQ (Haieret al., 2004), and decreased overall cerebellar volume is associatedwith a lower IQ (Allin et al., 2001). Additionally, fMRI measuresshow that activation during the performance of intelligence testscan be broadly distributed across both cortical and subcorticalregions (Esposito, Kirkby, Van Horn, Ellmore, & Berman, 1999;Gray et al., 2003; Lee et al., 2006; Prabhakaran, Rypma, &Gabrieli, 2001; Prabhakaran, Smith, Desmond, Glover, & Gabrieli,1997). Furthermore, damage to the frontal lobes often has littleimpact on the performance of well-learned cognitive skills or onthe IQ of adults (Milner, 1982), and children born with grosslydeficient or missing frontal lobes can do quite well in school(Ackerly, 1964). In one instance, a 16-year-old that had bothprefrontal lobes surgically removed showed normal intellectualcapacities and was able to live independently (Hebb, 1949). Thesefindings argue against the notion that frontal circuits providehumans with unique, specialized processing that is critical tocognitive skill learning or performance.

An alternative view of prefrontal function is that some circuitsin the frontal lobes may serve to monitor and facilitate cognitiveacts rather than to control and select them (Bunge, Kahn, Wallis,Miller, & Wagner, 2003; Bunge et al., 2005; J. Duncan, 2001).Consistent with this interpretation, individuals with a high IQ tendto rely on frontal cortex less as they gain mastery of a cognitiveskill (Gevins & Smith, 2000). Further evidence that a majorfunction of frontal cortical circuits is to facilitate processing inother brain regions came from studies of monkeys (Fuster, 2001;Genovesio, Brasted, Mitz, & Wise, 2005; Levy & Goldman-Rakic,2000; E. K. Miller & Cohen, 2001; Petrides, 1996; Romo &Salinas, 2003). In particular, measurements of prefrontal activity inmonkeys have been key in establishing that this region maintainsactivity that relates to the specific identity of remembered stimuli(Constantinidis, Franowicz, & Goldman-Rakic, 2001). These find-ings closely parallel results from human studies showing a closerelationship between intelligence and working memory (Acker-man, Beier, & Boyle, 2005; Broder, 2003; Conway et al., 2003;Engle, Tuholski, Laughlin, & Conway, 1999). Humans are not theonly animals that can maintain information in memory, and soresearchers have looked to other frontal lobe functions to explainthe exceptional intellectual capacities of humans. Two humanabilities that are thought to be fundamental to human cognition andlearning are language and imitation. Both abilities have beenassociated with specialized circuits in the frontal lobes, particularlyin Broca’s area and surrounding regions (Arbib, 2005; Grodzinsky,2006; Rizzolatti & Arbib, 1998), and both may contribute signif-icantly to an individual’s cognitive plasticity (a more detaileddiscussion of how these abilities contribute is provided below).

123NEURAL AND COGNITIVE PLASTICITY

The extent to which information about past events can contrib-ute to ongoing learning and decision making processes depends onhow well representations can be neurally maintained in the ab-sence of stimulation. Thus, one way that prefrontal cortex mayenhance cognitive plasticity is by contributing to the maintenanceof stimulus representations. It is important to note, however, thatother cortical regions also show a capacity to maintain stimulusrepresentations similar to that seen in prefrontal cortex (Romo &Salinas, 2003). For example, when monkeys identify visual imagesas being either the same or different as images they have seen inthe recent past, activity in their premotor cortex is quite similar tothat observed in prefrontal cortex (Wallis & Miller, 2003). Broadsimilarities in patterns of cortical activity seen in recordings fromvarious regions in monkeys performing cognitive skills suggestthat numerous cortical regions contribute to intellectual perfor-mance, consistent with data from human neuroimaging studies.

The contributions of frontal lobe circuits to representationalmaintenance are just one way that this brain region may impactcognitive plasticity. Prefrontal networks are also thought to func-tion in the control of attention and response inhibition, therebyconstraining the ways in which representations can be used (Bar-kley, 1997; Blair, 2006; Craik & Bialystok, 2006; Hedden &Gabrieli, 2006; West, 1996). Control of representational processesmay enable an organism to concentrate, consider various alterna-tives, suppress default responses, and access knowledge of similarpast situations. The influence that cognitive control can have oncognitive plasticity is most clearly suggested by changes in intel-lectual capacity observed across the human lifespan.

Cognitive development and cognitive aging. The saying “youcan’t teach an old dog new tricks” illustrates the common percep-tion that there is a strong link between aging and cognitive plas-ticity. Experimentally confirming this adage is difficult, however,because changes in intellectual capacity are confounded with, orare directly a result of, decreased sensory resolution, processingspeed, motor control, and memory abilities. Measuring cognitiveplasticity during early development is similarly complicated bysystematic improvements in these processes. Nevertheless, psy-chometric measures of intelligence consistently indicate that anumber of cognitive capacities increase during development, reachan asymptote, and then decline in later years (see, e.g., Li et al.,2004), as everyday experience would suggest. A simplistic inter-pretation of these findings is that cognitive plasticity initiallyincreases with age, then stabilizes, and finally decreases.

The development and functional capacity of several frontal loberegions show a similar trend, with decrements in frontal functionaccounting for at least some of the decrements seen in intellectualcapacity associated with aging (Bugg, Zook, DeLosh, Davalos, &Davis, 2006). Development and deterioration of frontal lobe struc-ture and function across the lifespan are often associated withrespective corresponding increases and decreases in cognitive con-trol (reviewed by Craik & Bialystok, 2006; Hedden & Gabrieli,2006; West, 1996). This is not to say that the ability to selectivelyprocess representations is the only, or even the major, factor thatdetermines age-related changes in intellectual capacity. Processingspeed and working memory capacity also decrease later in thelifespan (Hertzog, 1991; Salthouse, 1992, 2000), and both arecorrelated with individual differences in intelligence (Conway,Cowan, Bunting, Therriault, & Minkoff, 2002; Fry & Hale, 1996,2000). Memory search abilities (a type of cognitive control) appear

to be closely related to processing speed in older adults, but not inyounger adults (Baron & Mattila, 1989; Hertzog, Cooper, & Fisk,1996; Verhaeghen & Marcoen, 1996). Thus, processing speed,memory capacity, and memory control all deteriorate in parallel inelderly individuals, which likely contributes to decreases in intel-lectual capacity (Salthouse, 1996).

Developmental and lifespan researchers often focus on identi-fying the range of cognitive plasticity in humans and the types ofexperiences that facilitate or constrain plasticity (Baltes, 1987; Li,2003; Stiles, 2000). In the context of aging research, cognitiveplasticity is typically discussed in relation to techniques for fore-stalling cognitive decline (Kramer et al., 2004) and is operationallydefined as within-person variability in performance during trainingon cognitive tasks (e.g., Verhaeghen & Marcoen, 1996; Yang,Krampe, & Baltes, 2006). Expertise and practice seem to counter-act declines in certain cognitive skills, suggesting that age-relatedchanges occur in parallel across multiple cognitive domains ratherthan reflecting a global change in cognitive capacity (see Krameret al., 2004).

Many of the cognitive tasks that researchers have used to studycognitive skill learning during aging explicitly measure the effectsof training on memory abilities. For example, researchers havemeasured memory search performance in young and old individ-uals and the effects of practice on their performance (Fisk, Cooper,Hertzog, & Anderson-Garlach, 1995; Hertzog et al., 1996). In abasic memory search task, participants are instructed to hold itemsin memory and then to report on whether a probe item is in the set.Other memory tasks require participants to use mnemonics tofacilitate recall of episodic memories (Baltes & Kliegl, 1992;Singer, Lindenberger, & Baltes, 2003; Verhaeghen & Marcoen,1996). Measures of training-related increases in memory searchabilities in cognitive aging research are comparable in many re-spects with dynamic tests of intelligence (Embretson & Prenovost,2000; Grigorenko & Sternberg, 1998) as well as measures oflearning set formation in comparative research. Each techniqueinvolves using patterns of acquisition to assess differences inintellectual capacity. Interestingly, different species show the samebasic pattern of increasing, stabilizing, and declining cognitivecapacities as a function of age that is seen in humans, despite largedifferences in lifespan duration and in frontal lobe structure (deMagalhaes & Sandberg, 2005; Kausler, 1994). For example, stud-ies in monkeys (Harlow, 1959; Itoh et al., 2001; Rumbaugh, 1970;Rumbaugh & McCormack, 1967), and dogs (Boutet et al., 2005),show that their capacity to acquire learning sets increases duringdevelopment and then decreases during aging. In addition, thecounteracting effects of training experiences on cognitive declineare also seen in other animals (Milgram, 2003; Milgram, Siwak-Tapp, Araujo, & Head, 2006). What varies across species is not thedynamics of cognitive plasticity across the lifespan but rather theduration of the developmental process, the nature of the cognitiveskills learned, and asymptotic performance levels. According tothe present framework, variations in an organism’s ability to learncognitive skills throughout the lifespan reflect age-related varia-tions in its ability to differentiate stimulus representations.

The general finding from these and related studies of cognitiveskill learning is that older adults do not benefit from practice to thesame extent as younger adults, although the progress they show isqualitatively similar (Fisk & Rogers, 1991; Rogers, Fisk, & Hert-zog, 1994; Verhaeghen, Marcoen, & Goossens, 1992). Generally,

124 MERCADO

older individuals are slower to learn and seldom reach the levels ofperformance achieved by younger adults, suggesting that as agroup their cognitive plasticity is lower. There are different pat-terns of learning, however, depending on the specific skill beingpracticed and on initial levels of performance (Beier & Ackerman,2005; Verhaeghen & Marcoen, 1996; Verhaeghen et al., 1992).For example, younger adults show a greater capacity than agedadults to learn timed memory search skills, but elderly individualsshow greater improvement than young individuals when speededresponses are not required (Baron & Mattila, 1989). The observa-tion that individuals show different dynamics when acquiringdifferent cognitive skills, despite similarities in learning trajecto-ries, suggests that cognitive plasticity is a local constraint obeyingglobal rules. In the present explanatory framework, what thismeans is that representational resolution can vary as a function ofthe situation that an organism faces (leading to differences incognitive plasticity) but that the underlying mechanisms that de-termine the organism’s ability to differentiate stimulus represen-tations will be the same across all situations. Thus, an individual’sintellectual capacity can vary greatly depending on the types ofproblems to be solved (as suggested by proponents of theoriesdescribing multiple intelligences, e.g., Gardner, 1983), but therewill still be a common source of variance across these domains (assuggested by Spearman, 1904) since ultimately representationalresolution is limited by the structure and adaptability of the indi-vidual’s cortical networks.

The current framework characterizes the development and de-terioration of cognitive control as variability in the reconfigurabil-ity of cortical modules. It does not specify the prefrontal cortex (orany other brain region) as a locus of control and, therefore, doesnot predict major losses in cognitive plasticity or intellectualcapacities if this region is no longer present, for example, after thefrontal lobes are surgically removed. Instead the framework hy-pothesizes that prefrontal circuits can modulate the representa-tional resolution of cortical circuits by (a) facilitating the customi-zation of cortical modules via their projections to the basalforebrain and (b) enhancing the reconfigurability of cortical mod-ules by inhibiting some while maintaining the activity of others.Thus, prefrontal circuits may supplement existing cortical mech-anisms for differentiating stimulus representations without addingany unique functions that are characteristic of (or essential for) theintellectual abilities of humans.

Future directions and predictions. Neural circuits involving afixed number and diversity of cortical modules can vary greatly interms of their reconfigurability depending on how those modulesinteract. Case studies of human patients have indicated that whensubstantial portions of cortex (including prefrontal cortex) areremoved or were never present, cognitive plasticity in some indi-viduals remains relatively high. The neural changes that occur tocompensate for large losses of tissue are essentially unknown.Neuroimaging studies in such patients, especially during the learn-ing of cognitive skills, could provide new insights into how theircortical function differs from normal.

Relatively little is known about how the suite of neural circuitsinvolved in cognitive skill acquisition varies across individuals orspecies. Instead, neuroimaging research has focused on identifyinggeneral activation patterns that are consistent across many indi-viduals (e.g., group averages of relative changes in activity).Within-subject fMRI studies can provide new ways of identifying

factors that constrain an individual’s intellectual capacity. Forexample, real-time, fMRI-based neurofeedback experiments, inwhich participants voluntarily control blood flow in different cor-tical regions (deCharms et al., 2004, 2005), can be used to exper-imentally test the framework’s predictions about the reconfigu-rability and availability of cortical modules. Neuroimaging studieshave revealed specific brain regions that are often involved asparticipants learn particular cognitive skills. With practice, partic-ipants can improve their ability to control regional activationthrough feedback from fMRI images (deCharms et al., 2004; Yooet al., 2006). Using this technique, participants could graduallyincrease the number of co-activated cortical modules within one ormore brain regions typically associated with acquisition or perfor-mance of a cognitive skill, thereby potentially increasing theavailability of relevant cortical modules before the individualsattempt to learn that cognitive skill. The present framework pre-dicts that individuals that improve their ability to selectively acti-vate the relevant regions should benefit from this practice duringcognitive skill learning (i.e., this practice should increase theircognitive plasticity)—in effect, these individuals will have en-hanced their ability to select relevant cortical modules.

The present framework similarly predicts that if individuals canlearn to selectively control activity in their prefrontal cortex orbasal forebrain, then this should enhance cognitive plasticityacross a wide range of tasks. Studies demonstrating that targetedmanipulation of cortical activation can improve (or degrade) anindividual’s capacity to learn cognitive skills before training onthat skill has begun would strongly support the proposed relation-ship between cortical function and cognitive plasticity.

Customizability: Neural Efficiency and Plasticity

Whereas recent investigations of links between human intelli-gence and neural circuitry heavily emphasize prefrontal function,earlier work suggested that differences in global efficiency mightaccount for variability in intellectual capacity across individuals(see Deary & Caryl, 1997; Mackintosh, 1986, for reviews). Thebasic idea is that faster brains that can process information moreefficiently should have greater capacity, which is reflected inhigher IQs. This line of reasoning implies that circuits distributedthroughout the brain operate differently (e.g., more rapidly orefficiently) in more intelligent individuals, rather than just a fewcritical circuits operating differently. As noted earlier, behavioralevidence has shown a close correspondence between processingspeed and intelligence measures across the lifespan. This behav-ioral work is paralleled by studies showing correlations betweentemporal features of brain potentials (e.g., evoked potentials) andIQ (Caryl, 1994; Deary & Caryl, 1997). For example, shorterlatency P300s (a positive deflection within evoked potentials thatpeaks around 300 ms) are associated with higher IQs (Bazana &Stelmack, 2002; McGarry-Roberts, Stelmack, & Campbell, 1992).Other evidence suggesting that faster brains may be more likely toyield higher IQs has come from positive correlations betweenintelligence measures and measures of both nerve conductionvelocity (Reed, 1993, 2004) and temporal processing (Deary &Stough, 1996; Fink & Neubauer, 2005; Helmbold et al., 2006).

Neuroimaging studies showing increased frontal lobe activationin individuals with lower IQs have reinforced the notion thatefficiency yields capacity (e.g., Haier et al., 1992, 1988; Neubauer,

125NEURAL AND COGNITIVE PLASTICITY

Fink, & Schrausser, 2002). Specifically, these findings suggest thatinefficient use of frontal circuits (evidenced by high levels ofactivity) is associated with lower intellectual abilities. Further-more, one can actually see gradual decreases in frontal activationas an individual’s proficiency at performing a cognitive skillincreases (Ramsey, Jansma, Jager, Van Raalten, & Kahn, 2004).This disengagement of frontal cortical regions is associated withan increased capacity to perform multiple tasks in parallel, lendingcredence to the idea that efficiency affords capacity. Thus, al-though intellectual engagement often is associated with increasesin frontal activity (e.g., J. Duncan et al., 2000; Koechlin, Basso,Pietrini, Panzer, & Grafman, 1999; Koechlin et al., 2003), itappears that higher activation of frontal circuits per se does notprovide an individual with above average intellectual capacities.

Several mechanisms can potentially account for variability inneural efficiency across individuals, including arborization of cor-tical neurons, cerebral glucose metabolism, nerve conduction ve-locity, sex hormones, and others (reviewed by Neisser et al., 1996).Faster processing with minimal circuitry does not guarantee en-hanced learning capacity, however, and thus it remains unclearhow differences in neural efficiency might impact cognitive plas-ticity. Slower processing might limit the availability and fidelity ofstimulus representations, which should reduce cognitive plasticityaccording to the present explanatory framework.

The wave of scientific studies of neural efficiency has recededin the past decade, only to be replaced by burgeoning interest in therole that neural plasticity plays in behavioral flexibility. The sim-plest hypothesis about how neural plasticity relates to cognitiveplasticity is that there is a direct relationship: Greater neuralplasticity means greater intelligence. Variability in levels of neuralplasticity across individuals could affect how rapidly an individ-ual’s brain adapts to new situations, which could translate intodifferences in intellectual capacity (Garlick, 2002). Different abil-ities require different connections across circuits; neural plasticityguided by experience establishes these patterns of connections.Because many neural circuits exhibit neural plasticity, variations inthis capacity could potentially affect a wide variety of abilities.Different levels of neural plasticity across individuals could impactthe number and complexity of neural connections as well as therate at which new experiences lead to reorganization. These factorscould, in turn, affect processing speed and neural efficiency. Ge-netic manipulations that enhance neural plasticity in mice do seemto facilitate certain kinds of learning (Tang et al., 1999), consistentwith the idea that neural plasticity constrains learning capacity.

These recent proposals about how neural plasticity relates toindividual differences in intelligence have extended the classicalview, derived from comparative neuroanatomy studies, that brainstructure and complexity constrain behavioral flexibility (Jerison,1973; Lashley, 1949; Rensch, 1956). The hypothesis that variabil-ity in neural plasticity determines differences in intellectual capac-ity currently has limited explanatory power because (a) there is nostraightforward way to measure differences in the rate at whichneural circuits can change with sufficient precision to make cross-individual comparisons possible; (b) optimal levels of neural plas-ticity likely vary considerably depending on the nature of theintellectual tasks to be learned, the existing circuitry that might beengaged, and the maturity of the individual’s brain; and (c)changes in learning rate do not generally account for differences inwhat it is possible to learn—learning capacity is much more

dependent on how inputs are represented and on how a networklearns.

The current integrative framework addresses these limitationsby suggesting a specific role for neural plasticity beyond justmaking connections stronger or more efficient. Specifically, theframework proposes that cortical plasticity enables an organism toreallocate and retune cortical modules to increase overall repre-sentational resolution, thereby increasing cognitive plasticity. Boththe structure of cortical networks and the selectivity with whichbasal forebrain projections can modulate activity in particularcortical regions are hypothesized to limit how precisely corticalmodules can be customized. From this perspective, it is not theoverall capacity to change cortical connection strengths that con-strains an individual’s intellectual capacity (Garlick, 2002) but thebrain’s capacity to dynamically adjust the modifiability of discretesets of cortical modules as well as the likelihood that particularneural regions will become active during the acquisition or per-formance of a cognitive skill. In this way, the number and diversityof cortical modules that an organism engages during learning orperformance of a particular cognitive skill can change as a functionof experience. As noted previously, this proposal can be viewed asan extension of Hebb’s (1949) hypothesis that assemblies of neu-rons accumulate functional capacity based on experience. Thecontribution of the current framework to Hebb’s hypothesis is theidea that structural features of an organism’s brain constrain howeffectively such assemblies can be formed as well as the sugges-tion that variations in these structural features within and acrossindividuals and species may account for differences in intellectualcapacity.

Future directions and predictions: Brain building. Thepresent framework predicts that improving an individual’s repre-sentations of external stimuli can increase his or her capacity tolearn cognitive skills involving those stimuli. One way to increaserepresentational resolution might be through intensive training.Recent work showing that training programs designed to rehabil-itate children with language learning impairments can improve notonly children’s auditory perceptual acuity (Merzenich et al., 1996)but also their language comprehension skills (Tallal et al., 1996)provides some support for this prediction. Such training-relatedbehavioral changes are associated with changes in cortical activa-tion patterns (Temple et al., 2003), which may reflect changes inthe number of cortical modules engaged during performance of theskill.

This prediction can be further tested by controlling an animal’sexperiences during development (Hebb, 1949). Enriching a rat’senvironment, for example, increases dendritic branching in cortex(Rosenzweig, 1966; van Praag, Kempermann, & Gage, 2000;Volkmar & Greenough, 1972) as well as the strength of physio-logical responses (Engineer et al., 2004). With age, the topograph-ical organization of cortical maps deteriorates in rats, in parallelwith decreases in perceptual acuity and behavioral flexibility (Coq& Xerri, 2000, 2001; Leblanc, Weyers, & Soffie, 1996; Ohta,Matsumoto, & Watanabe, 1993; Turner, Hughes, & Caspary,2005). Enriched environments can attenuate this effect (Coq &Xerri, 2001). Similarly, training programs and enriched environ-ments can help elderly humans to delay declines in cognitivecapacity (see Kramer et al., 2004, for a review). The presentframework suggests that such experience-dependent changes re-flect continuous customization of cortical modules. Thus, it pre-

126 MERCADO

dicts that experimental manipulations that block cortical plasticitymechanisms should reduce the beneficial impacts of enrichment,and manipulations that enhance these mechanisms should amplifythe benefits of enrichment.

Impacts of Imitation and Language Abilities on CognitivePlasticity

The variety and complexity of cognitive skills possessed byadult humans suggest that cognitive plasticity is greater in humansthan in most other organisms. Historically, there have been heateddebates about why this is so. One group of thinkers has focused onexplaining why there is such a great intellectual divide betweenhumans and other animals (e.g., Dennett, 1997; Fodor, 1975). Thisgroup has proposed explanations for human intellectual abilitiesranging from divine providence (Descartes, 1637/1999) to theexistence of specialized neural language centers (Pinker, 1994);these proposals can be glossed as “enhanced cognitive plasticityresults from extra parts.” A second group of thinkers (exemplifiedby Darwin, 1871) has proposed that the intellectual separationbetween humans and other animals is not as great as it might atfirst appear (for recent reviews, see Heyes & Huber, 2000). Thisgroup recognizes that cognitive performance in humans far ex-ceeds that of other animals but questions whether the basic cog-nitive processes that underlie these performances are substantiallydifferent from those seen in many other organisms. From theDarwinian perspective, it is not the case that humans learn cogni-tive skills differently from other animals by using special human-specific brain circuits. Rather, humans may simply learn differentcognitive skills and may do so more rapidly and efficiently thancan most other species, using the same basic kinds of neuralcircuits that many organisms possess. Darwin’s ideas ultimatelygave rise to the fields of comparative psychology, animal learning,and comparative cognition, which have produced a great deal ofbehavioral data pertinent to understanding cognitive plasticity.Although clear examples of cognitive skill learning by non-humans remain relatively rare, there are now sufficient demonstra-tions of animals learning to count, classify, construct, and describeobjects to reject the claim that non-humans are mindless machines(Rumbaugh & Washburn, 2003).

Past explanations of the neural mechanisms underlying intelli-gence reflect earlier beliefs about why the intellectual capacities ofhumans differ from those of other animals. For instance, much ofthe recent effort to relate brain characteristics to human cognitionand intelligence has focused on identifying and localizing uniquefeatures of hominid brains that yield new and special abilities suchas language or mental time travel. This approach melds well withneo-Cartesian ideas about how specialized extra parts providehumans with intellectual capacities beyond those of all otheranimals. Parallel efforts to relate differences in intelligence acrossspecies or individuals to variability across numerous brain regionsare more consonant with the Darwinian proposal that such differ-ences vary along a continuum. A basic premise of the currentframework is that both approaches can potentially provide impor-tant insights about the neural mechanisms that constrain cognitiveplasticity.

One aspect of human intellectual abilities that there is littledisagreement about is that culture and language play importantroles. Both language and imitation abilities are associated with

specialized functions in specific subregions of frontal cortex (Ar-bib, 2005; Grodzinsky, 2006; Rizzolatti & Arbib, 1998; Rizzolatti& Craighero, 2004), consistent with the hypothesis that uniquecircuits in the frontal lobes are critical to the intellectual abilities ofhumans. In contrast, the current framework proposes that frontalcircuits simply supplement existing mechanisms for customizingand reconfiguring cortical modules. Consequently, it implicitlyendorses the Darwinian proposal that humans and other animalslearn cognitive skills in essentially similar ways and that theintellectual capacities of humans do not depend on any “extraparts” that are unique to humans. The following discussion pro-vides a speculative account of how culture and language mayamplify human cognitive plasticity by increasing the capacity ofhuman brains to differentiate stimulus representations.

Generalized Imitation

Several prominent psychologists have theorized that childrenlearn extensively by imitating the actions of others (Bandura,1977; N. Miller & Dollard, 1941; Piaget, 1953). It has also beensuggested that imitation is fundamental to cultural transmission ofinformation and the development of human societies (Blackmore,2000; Dawkins, 1976; Dennett, 1995). From these perspectives,imitation abilities are critical to the development of intellectualcapacities in humans. A common correlate of these proposals isthat humans rely on learning through imitation to a much greaterextent than do other animals and that this is one of the reasons thathumans are able to learn cognitive skills that other animals do not.

Researchers continue to debate whether animals other thanhumans have comparable capacities for learning through imitation(Laland & Janik, 2006; Rendell & Whitehead, 2001; Whiten,Horner, Litchfield, & Marshall-Pescini, 2004). Although numer-ous species have demonstrated the ability to imitate some actions(for a review, see Zentall, 2006), only a few such as bottlenosedolphins (Herman, 2002) and great apes (Custance, Whiten, &Bard, 1995; Hayes & Hayes, 1952) have shown the ability toflexibly imitate actions on command. In these studies, trainerstaught individuals to apply a general rule that can be glossed as“copy the action most recently observed” (Zentall & Akins, 2001).The ability to imitate on command is called generalized imitation.

A key aspect of generalized imitation is that it requires anindividual to retain information about past events and then usethese memories to select appropriate actions. In particular, gener-alized imitation involves using long-term or reference memories torecognize when a copying strategy is appropriate as well as usingworking memory for actions recently observed. The dependence ofgeneralized imitation on memory abilities suggests that memorycapacity and fidelity may strongly constrain the contribution ofimitation to cognitive skill learning. Thus, the likelihood that ananimal with imitation abilities might achieve levels of cognitiveplasticity comparable with those of a child depends on how well itsneural circuits can generate, actively maintain, and flexibly accessrepresentations of internal and external events. From the perspec-tive of the current framework, this means that animals with corticalmodules that are less available and less reconfigurable will benefitless from imitation abilities.

If children learn cognitive skills by imitating others, then themechanisms that constrain imitation will in turn limit levels ofcognitive plasticity. For example, if frontal circuits serve to in-

127NEURAL AND COGNITIVE PLASTICITY

crease the availability and reconfigurability of cortical modules (asproposed in the current framework), then this would in turn in-crease the role that imitation could play in cognitive skill learning,thereby increasing a child’s cognitive plasticity. Put another way,children can capitalize on a non-unique ability (generalized imi-tation) because they have greater representational resolution thando many other animals. The present framework thus assumes thatfrontal circuits contribute to children’s greater representationalresolution, but it allows for the possibility that comparable levelsof resolution might be achieved even if these circuits are notpresent. Thus, a child without frontal lobes may still be able tolearn a wide range of cognitive skills through imitation or othermeans quite well and may have an IQ comparable with that ofchildren with frontal lobes.

Language as a Learning Tool

Language use requires practice but not specific perceptual–motor acts (e.g., sentences can be written, gestured, or spokenaccording to regional customs). It thus seems reasonable to de-scribe language use as a cognitive skill. Toddlers are able to uselanguage to solve problems related to predicting and controllingthe behavior of others. In addition to serving communicativefunctions, language enables humans to reveal the contents of theirminds, to others and to themselves. Language thus provides hu-mans with unique ways of changing the contents of their minds(e.g., instructive learning, exposition, argumentation, deliberation,etc.; Carruthers, 2002; A. Clark, 1998). In modern humans, manycognitive skills are dependent on language abilities, as are manypsychometric tests designed to measure intelligence. This does notimply, however, that language is the linchpin of human intellectualcapacities. Language appears to be just one of many factors thatfacilitates cognitive skill learning by humans, including imitationabilities, a long life span, and environmental (especially cultural)affordances (Premack, 2004).

Learning to associate events, desires, and actions with wordschanges the problem-solving landscape by altering what is expe-rienced. Language can augment memory (e.g., mnemonics facili-tate recall) and simplify the environment. Language can alsoinfluence how events are perceived and how learning generalizes.For example, self-explanation facilitates generalization of problemsolving by children (Brown & Kane, 1988; Crowley & Siegler,1999). Language thus can serve as a cognitive tool that facilitatesknowledge acquisition and problem solving (A. Clark, 1998;Lieberman, 2000; Vicente & Martinez-Manrique, 2005); that is,language provides a catalyst for cognitive plasticity.

The idea that language is a tool that can enhance learningabilities is not a new one and is not limited to cognitive skilllearning by humans. Premack, in his classic studies of artificiallanguage learning by chimpanzees, reported that chimpanzees thatmastered artificial languages were able to learn to perform con-ceptual tasks such as a relational matching that “uneducated”chimpanzees showed no ability to learn (Premack, 1983b). It waslater shown that artificial language learning per se was not thecritical factor and that experience with symbol systems was suf-ficient to enhance learning capacity (R. K. Thompson, Oden, &Boysen, 1997). In a similar vein, Boysen and colleagues (Boysen,Bernston, Hannan, & Cacioppo, 1996) reported that chimpanzeesthat showed no ability to learn to choose the smaller of two

different amounts of candy were able to learn this task when thecandies were replaced with Arabic numerals. Findings such asthese suggest that the availability of symbolic representations, suchas those present within a language, can help an organism toovercome default modes of action (i.e., reconfigure cortical mod-ules), thereby enhancing cognitive plasticity.

One way that both language and imitation abilities may augmentcognitive plasticity in humans is by expanding the dimensions ofstimulus representations. For example, speech supplements visualscenes with acoustic labels, which are in turn linked to (a) themotor programs used to produce speech; (b) visual representationsof words; and (c) actions required to write those words. Conse-quently, seeing someone jump can activate representations acrossseveral modalities in parallel. Similarly, visual representations ofscenes involving actions that can be imitated (like jumping) maybe supplemented by the motor representations that would be re-quired to copy the actions (Wilson, 2001; Wilson & Knoblich,2005). Increasing the dimensions of the input space can simplifylearning by placing additional constraints on the problem space,particularly if a supplementary dimension provides a simpler (i.e.,more resolvable) stimulus representation. Thus, linguistic and im-itative capacities effectively expand the range of stimulus repre-sentations that an individual can bring to bear when attempting tosolve a problem. The role of language in stimulus differentiation ismost clearly evident during early language acquisition, as generaluse labels are gradually supplemented by more specific identifyingwords. In the present framework, it is this increased capacity todifferentiate stimulus representations that increases cognitive plas-ticity in humans rather than any specialized imitation or languagecompetence per se.

Conclusion

The present integrative framework holds that individuals withthe highest cognitive plasticity are those that have the greatestnumber and diversity of cortical modules as well as the greatestcapacity to dynamically retune and reconfigure those modules intointeracting representational networks on the basis of ongoingevents and past experiences. The explanation for why such indi-viduals have greater cognitive plasticity is that cortical circuits canincrease an organism’s ability to resolve stimulus representations(including representations of actions and thoughts), affording ad-ditional opportunities for the organism to learn new solutions toproblems. Several brain regions contribute to these processes.Basal forebrain neurons projecting throughout cortex can impactthe number of cortical modules engaged in stimulus processing aswell as the degree to which modules are customized during learn-ing experiences. Prefrontal cortical networks may monitor activityin cortical modules and modulate basal forebrain activity, impact-ing the duration of neural activity in either region as well asinteractions between cortical modules. The framework suggeststhat to predict differences in cognitive plasticity, both acrossspecies and across individuals within different species, it is criticalthat one has a clear understanding of the role that cortical networksplay in cognitive skill learning. Without this, it may be impossibleto predict with any precision how processing speed, or adaptabil-ity, or even specialized circuits will impact learning in a givenspecies or individual. With a clearer understanding of how corticalprocessing constrains cognitive plasticity, the reasons why intelli-

128 MERCADO

gence varies across individuals and species should become moretransparent.

One can think of this explanatory framework as replacing theprevalent “brain-as-computer” metaphor with a comparablysimple-minded simile: Cortex is like a set of tools. The subset oftools (cortical modules) relevant for building the representationsrequired to perform a particular cognitive skill may change asconstruction progresses. The size, shape, precision, and availabil-ity of each tool impacts its usefulness. The availability of custom-ized precision tools, in particular, may greatly facilitate construc-tion. Species and individuals with more tools and a greatercapacity to customize tools will be less constrained in terms of therepresentations that they can build. Development and experienceinitially increases an individual’s set of tools, but as agingprogresses, the likelihood that a tool is lost or broken increases. Ofcourse, replacing one questionable trope with another is hardlyprogress. Ultimately, the capacity of this framework to generatenew knowledge will determine its value.

Advances in psychometrics have transformed the way in whichindividual differences in human intelligence are conceptualizedand analyzed, whereas the development of comparable methodsfor measuring intellectual capacities in other species has scarcelybegun. Although new neuroimaging techniques provide powerfultools for correlating human capacities with brain structure andfunction, experiments that directly manipulate neural mechanismsto generate differences in cognitive plasticity are crucial for estab-lishing how cortical function contributes to intellectual capacity.Results from such experiments can not only provide new perspec-tives on the evolutionary processes that determine cognitive plas-ticity but may also clarify the role that development and aging playand suggest new strategies for overcoming intellectual deficits andmaximizing individual potential in humans and other animals. Forexample, if cognitive plasticity is constrained by a decreasednumber or diversity of cortical modules, as might happen after astroke, then techniques for externally controlling neuromodulation(e.g., through localized stimulation or drug infusions) could pro-vide a way of redistributing and customizing the available modulesto compensate for this loss. Experiments investigating the relation-ship between neural plasticity and intellectual capacity may thusprovide new insights into strategies for retaining or amplifyingcognitive plasticity in patients with brain disease or damage.

Human intelligence studies and comparative research havetaught us that identifying and measuring the intellectual capacitiesof different animals or individuals is a major empirical challengewith broad scope. Many of the difficulties associated with mea-suring and understanding variability in intelligence have arisen, atleast in part, as a historical consequence of government-sponsoredracial discrimination, and even genocide, in the name of eugenics.Research on the biological mechanisms of intelligence in the pasthas sparked heated debates regarding whether different groupsvary in intelligence and whether efforts should be made to increasethe average intelligence of the population (for a more detaileddiscussion, see Gray & Thompson, 2004; Mackintosh, 1986).Although the present framework attempts to identify biologicalmechanisms that constrain intellectual capacity, it does not directlyaddress how genetics impact those mechanisms. This frameworkalso does not rank different species, individuals, age groups, orracial groups in terms of their superiority or suggest that somebrains are inherently better than others. Rather, the current frame-

work provides a unified approach to investigating how variationsin neural circuits contribute to variations in cognitive plasticity,independent of the possible benefits and costs associated with suchvariability.

To the extent that intellectual capacity contributes to the mind,the contents of an organism’s mind reflect its capacity to learncognitive skills. Although the contents of different species’ mindsmay vary considerably, humans and other animals face manysimilar challenges when attempting to learn cognitive skills, andthe learning mechanisms that give rise to those contents appear toshare many features. For example, cognitive skill learning invari-ably depends on repeated encounters with similar situations, theability to rapidly recognize such similarities, and the capacity toselect appropriate responses. Additionally, in all species that havebeen studied, opportunities and capacities for learning new skillsincrease during development, stabilize in adulthood, and thendecrease during senescence. The debate between researchers whobelieve human intelligence is a unique outcome of specializedprocessors in the human brain and those who believe that theneural mechanisms mediating cognitive skills in humans areshared by other animals is unlikely to be resolved any time soon.Nevertheless, it can hardly be questioned that gaining a clearerpicture of how cortical networks function would increase ourunderstanding of how individuals become able to perform intelli-gent acts. In this way, investigating the relationship between neuraland cognitive plasticity in humans and other animals can providenew insights into how learning impacts intelligence, and howhuman minds arise.

References

Abboitz, F. (1996). Does bigger mean better? Evolutionary determinants ofbrain size and structure. Brain Behavior and Evolution, 47, 225–245.

Ackerly, S. S. (1964). A case of paranatal bilateral frontal lobe defectobserved for thirty years. In J. M. Warren & K. Akert (Eds.), The frontalgranular cortex and behavior (pp. 192–218). New York: McGraw-Hill.

Ackerman, P. L., Beier, M. E., & Boyle, M. O. (2005). Working memoryand intelligence: The same or different constructs? Psychological Bul-letin, 131(1), 30–60.

Ackerman, P. L., & Cianciolo, A. T. (2000). Cognitive, perceptual-speed,and psychomotor determinants of individual differences during skillacquisition. Journal of Experimental Psychology: Applied, 6, 259–290.

Acton, G. S., & Schroeder, D. H. (2001). Sensory discrimination as relatedto general intelligence. Intelligence, 29, 263–271.

Adolph, K. E. (2002). Learning to keep balance. Advances in ChildDevelopment and Behavior, 30, 1–40.

Allin, M., Matsumoto, H., Santhouse, A. M., Nosarti, C., AlAsady, M. H.,Stewart, A. L., et al. (2001). Cognitive and motor function and the sizeof the cerebellum in adolescents born very pre-term. Brain, 124(Pt 1),60–66.

Allman, J. M. (1998). Evolving brains. New York: Scientific AmericanLibrary/W. H. Freeman.

Allman, J. M., & Kaas, J. H. (1971). A representation of the visual field inthe caudal third of the middle temporal gyrus of the owl monkey. BrainResearch, 31, 85–105.

Anderson, J. R. (1982). Acquisition of cognitive skill. Psychological Re-view, 89, 369–406.

Andreasen, N. C., Flaum, M., Swayze, V., 2nd, O’Leary, D. S., Alliger, R.,Cohen, G., et al. (1993). Intelligence and brain structure in normalindividuals. American Journal of Psychiatry, 150(1), 130–134.

Anstey, K. J., & Smith, G. A. (1999). Interrelationships among biological

129NEURAL AND COGNITIVE PLASTICITY

markers of aging, health, activity, acculturation, and cognitive perfor-mance in late adulthood. Psychology and Aging, 14(4), 605–618.

Arbib, M. A. (2005). From monkey-like action recognition to humanlanguage: An evolutionary framework for neurolinguistics. Behavioraland Brain Sciences, 28(2), 105–124.

Ayers, J., Carpenter, G. A., Currie, S., & Kinch, J. (1983, September 23).Which behavior does the lamprey central motor program mediate?Science, 221(4617), 1312–1314.

Bailey, A. M., & Thomas, R. K. (2001). The effects of nucleus basalismagnocellularis lesions in Long-Evans hooded rats on two learning setformation tasks, delayed matching-to-sample learning, and open-fieldactivity. Behavioral Neuroscience, 115(2), 328–340.

Bakin, J. S., & Weinberger, N. M. (1996). Induction of a physiologicalmemory in the cerebral cortex by stimulation of the nucleus basalis.Proceedings of the National Academy of Sciences of the United States ofAmerica, 93, 11219–11224.

Baltes, P. B. (1987). Theoretical propositions of life-span developmentalpsychology: On the dynamics between growth and decline. Develop-mental Psychology, 5, 611–626.

Baltes, P. B., & Kliegl, R. (1992). Further testing of limits of cognitiveplasticity: Negative age differences in a mnemonic skill are robust.Developmental Psychology, 28, 121–125.

Baltes, P. B., & Lindenberger, U. (1997). Emergence of a powerfulconnection between sensory and cognitive functions across the adult lifespan: A new window to the study of cognitive aging? Psychology andAging, 12(1), 12–21.

Bandura, A. (1977). Social learning theory. New York: General LearningPress.

Barkley, R. A. (1997). Behavioral inhibition, sustained attention, andexecutive functions: Constructing a unifying theory of ADHD. Psycho-logical Bulletin, 121, 65–94.

Baron, A., & Mattila, W. R. (1989). Response slowing of older adults:Effects of time-limit contingencies on single- and dual-task perfor-mances. Psychology and Aging, 4(1), 66–72.

Bar-On, R., Tranel, D., Denburg, N. L., & Bechara, A. (2003). Exploringthe neurological substrate of emotional and social intelligence. Brain,126(Pt 8), 1790–1800.

Bazana, P. G., & Stelmack, R. M. (2002). Intelligence and informationprocessing during an auditory discrimination task with backward mask-ing: An event-related potential analysis. Journal of Personality andSocial Psychology, 83(4), 998–1008.

Beier, M. E., & Ackerman, P. L. (2005). Age, ability, and the role of priorknowledge on the acquisition of new domain knowledge: Promisingresults in a real-world learning environment. Psychology and Aging, 20,341–355.

Bence, M., & Levelt, C. N. (2005). Structural plasticity in the developingvisual system. Progress in Brain Research, 147, 125–139.

Biegler, R., McGregor, A., Krebs, J. R., & Healy, S. D. (2001). A largerhippocampus is associated with longer-lasting spatial memory. Proceed-ings of the National Academy of Sciences of the United States ofAmerica, 98(12), 6941–6944.

Binet, A., & Simon, T. (1961). New methods for the diagnosis of theintellectual level of subnormals (E. S. Kite, Trans.). In J. J. Jenkins &D. G. Paterson (Eds.), Studies of individual differences: The search forintelligence (pp. 90–96). New York: Appleton-Century-Crofts. (Origi-nal work published 1905)

Bitterman, M. E. (1965). The evolution of intelligence. Scientific Ameri-can, 212, 92–100.

Bitterman, M. E. (1975, May 16). The comparative analysis of learning.Science, 188, 699–709.

Blackmore, S. (2000). The power of memes. Scientific American, 283(4),pp. 64–66, 68–71, 73.

Blair, C. (2006). How similar are fluid cognition and general intelligence?A developmental neuroscience perspective on fluid cognition as an

aspect of human cognitive ability. Behavioral and Brain Sciences, 29(2),109–125.

Blaisdell, A. P., & Cook, R. G. (2005). Two-item same–different conceptlearning in pigeons. Learning and Behavior, 33(1), 67–77.

Bolhuis, J. J., & Macphail, E. M. (2001). A critique of the neuroecology oflearning and memory. Trends in Cognitive Sciences, 5(10), 426–433.

Boutet, I., Ryan, M., Kulaga, V., McShane, C., Christie, L. A., Freedman,M., et al. (2005). Age-associated cognitive deficits in humans and dogs:A comparative neuropsychological approach. Progress in Neuropsycho-pharmacology and Biological Psychiatry, 29(3), 433–441.

Boysen, S. T., Bernston, G. G., Hannan, M. B., & Cacioppo, J. T. (1996).Quantity-based interference and symbolic representations in chimpan-zees (Pan troglodytes). Journal of Experimental Psychology: AnimalBehavior Processes, 22(1), 76–86.

Brannon, E. M., & Terrace, H. S. (1998, October 23). Ordering of thenumerosities 1 to 9 by monkeys. Science, 282(5389), 746–749.

Broder, A. (2003). Decision making with the “adaptive toolbox”: Influenceof environmental structure, intelligence, and working memory load.Journal of Experimental Psychology: Learning, Memory, and Cogni-tion, 29(4), 611–625.

Brown, A. L., & Kane, M. J. (1988). Preschool children can learn totransfer: Learning to learn and learning from example. Cognitive Psy-chology, 20(4), 493–523.

Bugg, J. M., Zook, N. A., DeLosh, E. L., Davalos, D. B., & Davis, H. P.(2006). Age differences in fluid intelligence: Contributions of generalslowing and frontal decline. Brain and Cognition, 62(1), 9–16.

Bunge, S. A., Kahn, I., Wallis, J. D., Miller, E. K., & Wagner, A. D.(2003). Neural circuits subserving the retrieval and maintenance ofabstract rules. Journal of Neurophysiology, 90(5), 3419–3428.

Bunge, S. A., Wallis, J. D., Parker, A., Brass, M., Crone, E. A., Hoshi, E.,et al. (2005). Neural circuitry underlying rule use in humans and non-human primates. Journal of Neuroscience, 25(45), 10347–10350.

Buonomano, D. V., & Merzenich, M. M. (1998). Cortical plasticity: Fromsynapses to maps. Annual Review of Neuroscience, 21, 149–186.

Buxhoeveden, D. P., & Casanova, M. F. (2002). The minicolumn hypoth-esis in neuroscience. Brain, 125(Pt 5), 935–951.

Byrne, R. W. (1995). The thinking ape: Evolutionary origins of intelli-gence. Oxford, England: Oxford University Press.

Byrne, R. W., & Corp, N. (2004). Neocortex size predicts deception rate inprimates. Proceedings of the Royal Society of London, Series B: Bio-logical Sciences, 271(1549), 1693–1699.

Byrne, R. W., & Whiten, A. (1988). Machiavellian intelligence: Socialexpertise and the evolution of intellect in monkeys, apes, and humans.New York: Oxford University Press.

Calford, M. B. (2002). Dynamic representational plasticity in sensorycortex. Neuroscience, 111(4), 709–738.

Carruthers, P. (2002). The cognitive functions of language. Behavioral andBrain Sciences, 25(6), 657–674.

Caryl, P. G. (1994). Early event-related potentials correlate with inspectiontime and intelligence. Intelligence, 18, 15–46.

Catania, K. C., & Kaas, J. H. (1995). Organization of the somatosensorycortex of the star-nosed mole. Journal of Comparative Neurology,351(4), 549–567.

Chang, E. F., & Merzenich, M. M. (2003, April 18). Environmental noiseretards auditory cortical development. Science, 300(5618), 498–502.

Chen, A., Muzzio, I. A., Malleret, G., Bartsch, D., Verbitsky, M., Pavlidis,P., et al. (2003). Inducible enhancement of memory storage and synapticplasticity in transgenic mice expressing an inhibitor of ATF4 (CREB-2)and C/EBP proteins. Neuron, 39(4), 655–669.

Clark, A. (1998). Magic words: How language augments human compu-tation. In P. Carruthers & J. Boucher (Eds.), Language and thought:Interdisciplinary themes (pp. 162–183). Cambridge, England: Cam-bridge University Press.

130 MERCADO

Clark, D. A., Mitra, P. P., & Wang, S. S. (2001, May 10). Scalablearchitecture in mammalian brains. Nature, 411(6834), 189–193.

Colom, R., Jung, R. E., & Haier, R. J. (2006). Distributed brain sites for theg-factor of intelligence. NeuroImage, 31(3), 1359–1365.

Constantinidis, C., Franowicz, M. N., & Goldman-Rakic, P. S. (2001). Thesensory nature of mnemonic representation in the primate prefrontalcortex. Nature Neuroscience, 4(3), 311–316.

Conway, A. R., Cowan, N., Bunting, M. F., Therriault, D. J., & Minkoff,S. R. B. (2002). A latent variable analysis of working memory capacity,short-term memory capacity, processing speed, and general fluid intel-ligence. Intelligence, 30, 163–183.

Conway, A. R., Kane, M. J., & Engle, R. W. (2003). Working memorycapacity and its relation to general intelligence. Trends in CognitiveSciences, 7(12), 547–552.

Cooke, S. F., & Bliss, T. V. (2006). Plasticity in the human central nervoussystem. Brain, 129(Pt 7), 1659–1673.

Coq, J. O., & Xerri, C. (2000). Age-related alteration of the forepawrepresentation in the rat primary somatosensory cortex. Neuroscience,99(3), 403–411.

Coq, J. O., & Xerri, C. (2001). Sensorimotor experience modulates age-dependent alterations of the forepaw representation in the rat primarysomatosensory cortex. Neuroscience, 104(3), 705–715.

Corballis, M. C. (2007). The uniqueness of human recursive thinking.American Scientist, 95, 240–248.

Corballis, M. C., & Lea, S. E. G. (Eds.). (1999). The descent of mind:Psychological perspectives on hominid evolution. New York: OxfordUniversity Press.

Covey, E. (2005). Neurobiological specializations in echolocating bats.Anatomical Record Part A: Discoveries in Molecular, Cellular, andEvolutionary Biology, 287(1), 1103–1116.

Craik, F. I. M., & Bialystok, E. (2006). Cognition through the lifespan:Mechanisms of change. Trends in Cognitive Sciences, 10, 131–138.

Crone, E. A., Donohue, S. E., Honomichl, R., Wendelken, C., & Bunge,S. A. (2006). Brain regions mediating flexible rule use during develop-ment. Journal of Neuroscience, 26(43), 11239–11247.

Crone, E. A., Wendelken, C., Donohue, S. E., & Bunge, S. A. (2006).Neural evidence for dissociable components of task-switching. CerebralCortex, 16(4), 475–486.

Crowley, K., & Siegler, R. S. (1999). Explanation and generalization inyoung children’s strategy learning. Child Development, 70(2), 304–316.

Custance, D. M., Whiten, A., & Bard, K. A. (1995). Can young chimpan-zees imitate arbitrary actions? Hayes and Hayes revisited. Behaviour,132, 839–858.

Darwin, C. (1871). The descent of man, and selection in relation to sex.London: John Murray.

Dawkins, R. (1976). The selfish gene. Oxford, England: Oxford UniversityPress.

Deary, I. J. (1994). Sensory discrimination and intelligence: Postmortem orresurrection? American Journal of Psychiatry, 107, 95–115.

Deary, I. J. (1995). Auditory inspection time and intelligence: What is thedirection of causation? Developmental Psychology, 31, 237–250.

Deary, I. J. (2000). Looking down on intelligence: From psychometrics tothe brain. Oxford, England: Oxford University Press.

Deary, I. J., Bell, P. J., Bell, A. J., Campbell, M. L., & Fazal, N. D. (2004).Sensory discrimination and intelligence: Testing Spearman’s other hy-pothesis. American Journal of Psychology, 117(1), 1–18.

Deary, I. J., & Caryl, P. G. (1997). Neuroscience and human intelligencedifferences. Trends in Neurosciences, 20(8), 365–371.

Deary, I. J., & Stough, C. (1996). Intelligence and inspection time:Achievements, prospects, and problems. American Psychologist, 51,599–608.

deCharms, R. C., Christoff, K., Glover, G. H., Pauly, J. M., Whitfield, S.,& Gabrieli, J. D. (2004). Learned regulation of spatially localized brainactivation using real-time fMRI. NeuroImage, 21(1), 436–443.

deCharms, R. C., Maeda, F., Glover, G. H., Ludlow, D., Pauly, J. M.,Soneji, D., et al. (2005). Control over brain activation and pain learnedby using real-time functional MRI. Proceedings of the National Acad-emy of Sciences of the United States of America, 102(51), 18626–18631.

de Magalhaes, J. P., & Sandberg, A. (2005). Cognitive aging as anextension of brain development: A model linking learning, brain plas-ticity, and neurodegeneration. Mechanisms of Ageing and Development,126(10), 1026–1033.

Dennett, D. C. (1995). Darwin’s dangerous idea: Evolution and the mean-ings of life. New York: Simon and Schuster.

Dennett, D. C. (1997). Kinds of minds: Toward an understanding ofconsciousness. New York: Basic Books.

Descartes, R. (1999). Discourse on method and meditations on first phi-losophy (D. A. Cress, Trans.). Cambridge, England: Hackett Pub Co.(Original work published in 1637)

Doeller, C. F., Opitz, B., Krick, C. M., Mecklinger, A., & Reith, W. (2006).Differential hippocampal and prefrontal-striatal contributions toinstance-based and rule-based learning. NeuroImage, 31(4), 1802–1816.

Doty, B. A., Jones, C. N., & Doty, L. A. (1967, March 24). Learning-setformation by mink, ferrets, skunks, and cats. Science, 155(769), 1579–1580.

Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size inprimates. Journal of Human Evolution, 22, 469–493.

Dunbar, R. I. M. (2003). The social brain: Mind, language, and society inevolutionary perspective. Annual Review of Anthropology, 32, 163–181.

Duncan, J. (2001). An adaptive coding model of neural function in pre-frontal cortex. Nature Reviews Neuroscience, 2(11), 820–829.

Duncan, J., Seitz, R. J., Kolodny, J., Bor, D., Herzog, H., Ahmed, A., et al.(2000, July 21). A neural basis for general intelligence. Science,289(5478), 457–460.

Duncan, R. O., & Boynton, G. M. (2003). Cortical magnification withinhuman primary visual cortex correlates with acuity thresholds. Neuron,38(4), 659–671.

Ebbesson, S. O. E. (1980). The parcellation theory and its relation tointerspecific variability in brain organization, evolutionary and ontoge-netic development, and neuronal plasticity. Cell and Tissue Research,213(2), 179–212.

Ebbesson, S. O. E. (1984). Evolution and ontogeny of neural circuits.Behavioral and Brain Sciences, 7, 321–366.

Embretson, S. E. (1992). Measuring and validating cognitive modifiabilityas an ability: A study in the spatial domain. Journal of EducationalMeasurement, 29, 25–50.

Embretson, S. E., & Prenovost, L. K. (2000). Dynamic cognitive testing:What kind of information is gained by measuring response time andmodifiability? Educational and Psychological Measurement, 60, 837–863.

Engineer, N. D., Percaccio, C. R., Pandya, P. K., Moucha, R., Rathbun,D. L., & Kilgard, M. P. (2004). Environmental enrichment improvesresponse strength, threshold, selectivity, and latency of auditory cortexneurons. Journal of Neurophysiology, 92(1), 73–82.

Engle, R. W., Tuholski, S. W., Laughlin, J. E., & Conway, A. R. (1999).Working memory, short-term memory, and general fluid intelligence: Alatent-variable approach. Journal of Experimental Psychology: General,128(3), 309–331.

Esposito, G., Kirkby, B. S., Van Horn, J. D., Ellmore, T. M., & Berman,K. F. (1999). Context-dependent, neural system-specific neurophysio-logical concomitants of ageing: Mapping PET correlates during cogni-tive activation. Brain, 122(Pt 5), 963–979.

Ewert, J. P. (1984). Behavioral selectivity based on thalamotectal interac-tions: Ontogenetic and phylogenetic aspects in amphibians. Behavioraland Brain Sciences, 7, 337–338.

Eysenck, H. J. (1982). A model for intelligence. New York: CambridgeUniversity Press.

Favorov, O. V., & Diamond, M. E. (1990). Demonstration of discrete

131NEURAL AND COGNITIVE PLASTICITY

place-defined columns—segregates—in the cat SI. Journal of Compar-ative Neurology, 298(1), 97–112.

Fay, R. R. (1974). Auditory frequency discrimination in vertebrates. Jour-nal of the Acoustical Society of America, 56, 206–209.

Feldman, D. E., & Brecht, M. (2005, November 4). Map plasticity insomatosensory cortex. Science, 310(5749), 810–815.

Feldman, D. H. (1993). Child prodigies: A distinctive form of giftedness.National Association for Gifted Children, 37, 188–193.

Fibiger, H. C. (1991). Cholinergic mechanisms in learning, memory anddementia: A review of recent evidence. Trends in Neurosciences, 14(6),220–223.

Fink, A., & Neubauer, A. C. (2005). Individual differences in time esti-mation related to cognitive ability, speed of information processing andworking memory. Intelligence, 33, 5–26.

Fisk, A. D., Cooper, B. P., Hertzog, C., & Anderson-Garlach, M. (1995).Age-related retention of skilled memory search: Examination of asso-ciative learning, interference, and task-specific skills. Journals of Ger-ontology, Series B: Psychological Sciences and Social Sciences, 50(3),P150–161.

Fisk, A. D., & Rogers, W. A. (1991). Toward an understanding of age-related memory and visual search effects. Journal of ExperimentalPsychology: General, 120(2), 131–149.

Fitts, P. M. (1964). Perceptual–motor skill learning. In A. W. Melton (Ed.),Categories of human learning (pp. 243–285). New York: AcademicPress.

Flashman, L. A., Andreasen, N. C., Flaum, M., & Swayze, V. W. (1997).Intelligence and regional brain volumes in normal controls. Intelligence,25, 149–160.

Fodor, J. A. (1975). The language of thought. New York: Thomas Y.Crowell Co.

Fodor, J. A. (1983). The modularity of mind. Cambridge, MA: MIT Press.Fry, A. F., & Hale, S. (1996). Processing speed, working memory, and

fluid intelligence: Evidence for a developmental cascade. PsychologicalScience, 7, 237–241.

Fry, A. F., & Hale, S. (2000). Relationships among processing speed,working memory, and fluid intelligence in children. Biological Psychol-ogy, 54(1–3), 1–34.

Fuster, J. M. (2001). The prefrontal cortex—an update: Time is of theessence. Neuron, 30(2), 319–333.

Galton, F. (1883). Inquiries into human faculty. London: Dent.Gardner, H. (1983). Frames of mind. The theory of multiple intelligences.

New York: BasicBooks.Garlick, D. (2002). Understanding the nature of the general factor of

intelligence: The role of individual differences in neural plasticity as anexplanatory mechanism. Psychological Review, 109(1), 116–136.

Garlick, D. (2003). Integrating brain science research with intelligenceresearch. Current Directions in Psychological Science, 12, 185–189.

Gasser, T., Jennen-Steinmetz, C., & Verleger, R. (1987). EEG coherence atrest and during a visual task in two groups of children. Electroencepha-lography and Clinical Neurophysiology, 67(2), 151–158.

Genovesio, A., Brasted, P. J., Mitz, A. R., & Wise, S. P. (2005). Prefrontalcortex activity related to abstract response strategies. Neuron, 47(2),307–320.

Gevins, A., & Smith, M. E. (2000). Neurophysiological measures ofworking memory and individual differences in cognitive ability andcognitive style. Cerebral Cortex, 10(9), 829–839.

Gibson, K. R. (2002). Evolution of human intelligence: The roles of brainsize and mental construction. Brain Behavior and Evolution, 59(1–2),10–20.

Gignac, G., Vernon, P. A., & Wickett, J. C. (2003). Factors influencing therelationship between brain size and intelligence. In H. Nyborg (Ed.), Thescientific study of general intelligence: Tribute to A. R. Jensen (pp.93–106). New York: Pergamon.

Grady, C. L., Springer, M. V., Hongwanishkul, D., McIntosh, A. R., &

Winocur, G. (2006). Age-related changes in brain activity across theadult lifespan. Journal of Cognitive Neuroscience, 18(2), 227–241.

Gray, J. R., Chabris, C. F., & Braver, T. S. (2003). Neural mechanisms ofgeneral fluid intelligence. Nature Neuroscience, 6(3), 316–322.

Gray, J. R., & Thompson, P. M. (2004). Neurobiology of intelligence:Science and ethics. Nature Reviews Neuroscience, 5(6), 471–482.

Grigorenko, E. L., & Sternberg, R. J. (1998). Dynamic testing. Psycholog-ical Bulletin, 124, 75–111.

Grodzinsky, Y. (2006). The language faculty, Broca’s region, and themirror system. Cortex, 42(4), 464–468.

Haier, R. J., Jung, R. E., Yeo, R. A., Head, K., & Alkire, M. T. (2004).Structural brain variation and general intelligence. NeuroImage, 23(1),425–433.

Haier, R. J., Jung, R. E., Yeo, R. A., Head, K., & Alkire, M. T. (2005). Theneuroanatomy of general intelligence: Sex matters. NeuroImage, 25(1),320–327.

Haier, R. J., Siegel, B. V., Jr., MacLachlan, A., Soderling, E., Lottenberg,S., & Buchsbaum, M. S. (1992). Regional glucose metabolic changesafter learning a complex visuospatial/motor task: A positron emissiontomographic study. Brain Research, 570(1–2), 134–143.

Haier, R. J., Siegel, B. V., Jr., Nuechterlein, K. H., Hazlett, E., Wu, J. C.,Paek, J., et al. (1988). Cortical glucose metabolic rate correlates ofabstract reasoning and attention studied with positron emission tomog-raphy. Intelligence, 12, 199–217.

Halford, G. S., Bain, J. D., Maybery, M. T., & Andrews, G. (1998).Induction of relational schemas: Common processes in reasoning andcomplex learning. Cognitive Psychology, 35(3), 201–245.

Harlow, H. F. (1949). The formation of learning sets. PsychologicalReview, 56, 51–65.

Harlow, H. F. (1959). Learning set and error factor theory. In S. Koch(Ed.), Psychology: A study of a science (Vol. 2; pp. 492–537). NewYork: McGraw-Hill.

Harris, K. D. (2005). Neural signatures of cell assembly organization.Nature Reviews Neuroscience, 6, 399–407.

Harris-Warrick, R. M., & Marder, E. (1991). Modulation of neural net-works for behavior. Annual Review of Neuroscience, 14, 39–57.

Hayes, K. J., & Hayes, C. (1952). Imitation in a home-reared chimpanzee.Journal of Comparative and Physiological Psychology, 45, 450–459.

Healy, S. D., & Rowe, C. (2007). A critique of comparative studies of brainsize. Proceedings of the Royal Society of London, Series B: BiologicalSciences, 274, 453–464.

Hebb, D. O. (1949). The organization of behavior: A neuropsychologicaltheory. New York: Wiley.

Hebb, D. O. (1980). Essay on mind. Hillsdale, NJ: Lawrence ErlbaumAssociates.

Hedden, T., & Gabrieli, J. D. (2006). The ebb and flow of attention in thehuman brain. Nature Neuroscience, 9(7), 863–865.

Heiligenberg, W. (1990). Electrosensory systems in fish. Synapse, 6(2),196–206.

Helmbold, N., Troche, S., & Rammsayer, T. (2006). Temporal informationprocessing and pitch discrimination as predictors of general intelligence.Canadian Journal of Experimental Psychology, 60, 294–306.

Herculano-Houzel, S., Mota, B., & Lent, R. (2006). Cellular scaling rulesfor rodent brains. Proceedings of the National Academy of Sciences ofthe United States of America, 103(32), 12138–12143.

Herman, L. M. (2002). Vocal, social, and self-imitation by bottlenoseddolphins. In C. Nehaniv & K. Dautenhahn (Eds.), Imitation in animalsand artifacts (pp. 63–108). Cambridge, MA: MIT Press.

Herman, L. M., & Arbeit, W. R. (1973). Stimulus control and auditorydiscrimination learning sets in the bottlenose dolphin. Journal of theExperimental Analysis of Behavior, 19(3), 379–394.

Herman, L. M., Beach, F. A., III, Pepper, R. L., & Stalling, R. B. (1969).Learning-set formation in the bottlenose dolphin. Psychonomic Science,14, 98–99.

132 MERCADO

Heron, W. T. (1935). The inheritance of maze learning ability in rats.Journal of Comparative Psychology, 19, 77–89.

Hertzog, C. (1991). Aging, information processing speed, and intelligence.In K. W. Schaie (Ed.), Annual review of geronotology and geriatrics(Vol. 11, pp. 55–79). New York: Springer.

Hertzog, C., Cooper, B. P., & Fisk, A. D. (1996). Aging and individualdifferences in the development of skilled memory search performance.Psychology and Aging, 11(3), 497–520.

Heyes, C., & Huber, L. (Eds.). (2000). The evolution of cognition. Cam-bridge, MA: MIT Press.

Hodos, W., & Campbell, C. B. G. (1969). Scala naturae: Why there is notheory in comparative psychology. Psychological Review, 76, 337–350.

Hofman, M. A. (2003). Of brains and minds: A neurobiological treatise onthe nature of intelligence. Evolution and Cognition, 9, 178–188.

Holekamp, K. E. (2007). Questioning the social intelligence hypothesis.Trends in Cognitive Sciences, 11(2), 65–69.

Hubel, D. H., & Wiesel, T. N. (1963). Shape and arrangement of columnsin the cat’s striate cortex. Journal of Physiology, 165, 559–568.

Itoh, K., Izumi, A., & Kojima, S. (2001). Object discrimination learning inaged Japanese monkeys. Behavioral Neuroscience, 115(2), 259–270.

James, W. (1890). The principles of psychology. New York: H. Holt andCo.

Jausovec, N., & Jausovec, K. (2000). Differences in event-related andinduced brain oscillations in the theta and alpha frequency bands relatedto human intelligence. Neuroscience Letters, 293(3), 191–194.

Jensen, A. R. (1989). The relationship between learning and intelligence.Learning and Individual Differences, 1, 37–62.

Jensen, A. R., & Sinha, S. N. (1993). Physical correlates of intelligence. InP. A. Vernon (Ed.), The biological basis of intelligence (pp. 139–242).Norwood, NJ: Ablex.

Jerison, H. J. (1973). Evolution of the brain and intelligence. New York:Academic Press.

Jerison, H. J. (2002). On theory in comparative psychology. In R. J.Sternberg & J. C. Kaufman (Eds.), The evolution of intelligence (pp.251–288). Mahwah, NJ: Lawrence Erlbaum Associates.

Johnson, M. H., & Vecera, S. P. (1996). Cortical differentiation andneurocognitive development: The parcellation conjecture. BehaviouralProcesses, 36, 195–212.

Johnson-Laird, P. N. (1982). Thinking as a skill. Quarterly Journal ofExperimental Psychology: Human Experimental Psychology, 34(A),1–29.

Kaas, J. H. (1984). Duplication of brain parts in evolution. Behavioral andBrain Sciences, 7, 342–343.

Kaas, J. H. (1991). Plasticity of sensory and motor maps in adult mammals.Annual Review of Neuroscience, 14, 137–167.

Kaas, J. H., & Collins, C. E. (2001, May 10). Evolving ideas of brainevolution. Nature, 411, 141–142.

Kahn, D. M., & Krubitzer, L. (2002). Massive cross-modal cortical plas-ticity and the emergence of a new cortical area in developmentally blindmammals. Proceedings of the National Academy of Sciences of theUnited States of America, 99(17), 11429–11434.

Katz, J. S., & Wright, A. A. (2006). Same/different abstract conceptlearning by pigeons. Journal of Experimental Psychology: Animal Be-havior Processes, 32, 80–86.

Kausler, D. H. (1994). Learning and memory in normal aging. San Diego,CA: Academic Press.

Kilgard, M., & Merzenich, M. (1998, March 13). Cortical map reorgani-zation enabled by nucleus basalis activity. Science, 279, 1714–1718.

Koechlin, E., Basso, G., Pietrini, P., Panzer, S., & Grafman, J. (1999, May13). The role of anterior prefrontal cortex in human cognition. Nature,399, 148–151.

Koechlin, E., Ody, C., & Kouneiher, F. (2003, November 14). The archi-tecture of cognitive control in the human prefrontal cortex. Science,302(5648), 1181–1185.

Kramer, A. F., Bherer, L., Colcombe, S. J., Dong, W., & Greenough, W. T.(2004). Environmental influences on cognitive and brain plasticity dur-ing aging. Journals of Gerontology, Series A: Biological Sciences andMedical Sciences, 59(9), M940–957.

Krubitzer, L., & Kaas, J. (2005). The evolution of the neocortex inmammals: How is phenotypic diversity generated? Current Opinion inNeurobiology, 15(4), 444–453.

Laland, K. N., & Janik, V. M. (2006). The animal cultures debate. Trendsin Ecology and Evolution, 21(10), 542–547.

Lashley, K. S. (1929). Brain mechanisms and intelligence: A quantitativestudy of injuries to the brain. Chicago: University of Chicago Press.

Lashley, K. S. (1949). Persistent problems in the evolution of mind.Quarterly Review of Biology, 24, 28–42.

Leblanc, P., Weyers, M. H., & Soffie, M. (1996). Age-related differencesfor duration discrimination in rats. Physiology and Behavior, 60(2),555–558.

Lee, K. H., Choi, Y. Y., Gray, J. R., Cho, S. H., Chae, J.-H., Lee, S., &Kim, K. (2006). Neural correlates of superior intelligence: Strongerrecruitment of posterior parietal cortex. NeuroImage, 29, 578–586.

Lefebvre, L. (1996). Ecological correlates of social learning: Problems andsolutions for the comparative method. Behavioural Processes, 35, 163–171.

Lefebvre, L., Whittle, P., Lascaris, E., & Finkelstein, A. (1997). Feedinginnovations and forebrain size in birds. Animal Behaviour, 53, 549–560.

Leise, E. M. (1990). Modular construction of nervous systems: A basicprinciple of design for invertebrates and vertebrates. Brain ResearchBrain Research Reviews, 15(1), 1–23.

Levy, R., & Goldman-Rakic, P. S. (2000). Segregation of working memoryfunctions within the dorsolateral prefrontal cortex. Experimental BrainResearch, 133(1), 23–32.

Li, S. C. (2003). Biocultural orchestration of developmental plasticityacross levels: The interplay of biology and culture in shaping the mindand behavior across the life span. Psychological Bulletin, 129(2), 171–194.

Li, S. C., Brehmer, Y., Shing, Y. L., Werkle-Bergner, M., & Lindenberger,U. (2006). Neuromodulation of associative and organizational plasticityacross the life span: Empirical evidence and neurocomputational mod-eling. Neuroscience and Biobehavioral Reviews, 30, 775–790.

Li, S. C., Lindenberger, U., Hommel, B., Aschersleben, G., Prinz, W., &Baltes, P. B. (2004). Transformations in the couplings among intellec-tual abilities and constituent cognitive processes across the life span.Psychological Science, 15, 155–163.

Lieberman, P. (2000). Human language and our reptilian brain. Cam-bridge, MA: Harvard University Press.

Lindenberger, U., & Baltes, P. B. (1994). Sensory functioning and intel-ligence in old age: A strong connection. Psychology and Aging, 9(3),339–355.

Lindenberger, U., & Baltes, P. B. (1997). Intellectual functioning in oldand very old age: Cross-sectional results from the Berlin Aging Study.Psychology and Aging, 12(3), 410–432.

Lindenberger, U., Scherer, H., & Baltes, P. B. (2001). The strong connec-tion between sensory and cognitive performance in old age: Not due tosensory acuity reductions operating during cognitive assessment. Psy-chology and Aging, 16(2), 196–205.

Lindenfors, P. (2005). Neocortex evolution in primates: The ‘social brain’is for females. Biology Letters, 1, 407–410.

Mace, G. M., Harvey, P. H., & Clutton-Brock, T. H. (1980). Is brain sizean ecological variable? Trends in Neurosciences, 3, 193–196.

Mackintosh, N. J. (1975). A theory of attention: Variations in the associa-bility of stimuli with reinforcement. Psychological Review, 82, 276–298.

Mackintosh, N. J. (1986). The biology of intelligence? British Journal ofPsychology, 77(Pt 1), 1–18.

133NEURAL AND COGNITIVE PLASTICITY

Mackintosh, N. J. (1998). IQ and human intelligence. Oxford, England:Oxford University Press.

Macphail, E. M. (1982). Brain and intelligence in vertebrates. Oxford,England: Clarendon Press.

Marino, L. (2002). Convergence of complex cognitive abilities in cetaceansand primates. Brain Behavior and Evolution, 59(1–2), 21–32.

Marino, L. (2006). Absolute brain size: Did we throw the baby out with thebathwater? Proceedings of the National Academy of Sciences of theUnited States of America, 103(37), 13563–13564.

Marino, L., Sudheimer, K. D., Murphy, T. L., Davis, K. K., Pabst, D. A.,McLellan, W. A., et al. (2001). Anatomy and three-dimensional recon-structions of the brain of a bottlenose dolphin (Tursiops truncatus) frommagnetic resonance images. Anatomical Record, 264(4), 397–414.

McDaniel, M. A. (2005). Big-brained people are smarter: A meta-analysisof the relationship between in vivo brain volume and intelligence.Intelligence, 33, 337–346.

McGarry-Roberts, P. A., Stelmack, R. M., & Campbell, K. B. (1992).Intelligence, reaction-time, and event-related potentials. Intelligence, 16,289–313.

Mercado, E., III, Bao, S., Orduna, I., Gluck, M. A., & Merzenich, M. M.(2001). Basal forebrain stimulation changes cortical sensitivities to com-plex sound. NeuroReport, 12(10), 2283–2287.

Mercado, E., III, Killebrew, D. A., Pack, A. A., Macha, I. I., & Herman,L. M. (2000). Generalization of ‘same–different’ classification abilitiesin bottlenosed dolphins. Behavioural Processes, 50(2–3), 79–94.

Mercado, E., III, Murray, S. O., Uyeyama, R. K., Pack, A. A., & Herman,L. M. (1998). Memory for recent actions in the bottlenosed dolphin(Tursiops truncatus): Repetition of arbitrary behaviors using an abstractrule. Animal Learning and Behavior, 26, 210–218.

Mercado, E., III, Myers, C. E., & Gluck, M. A. (2001). A computationalmodel of mechanisms controlling experience-dependent reorganizationof representational maps in auditory cortex. Cognitive Affective andBehavioral Neuroscience, 1(1), 37–55.

Merzenich, M. M. (1985). Sources of intraspecies and interspecies corticalmap variability in mammals: Conclusions and hypotheses. In M. J.Cohen & F. Strumwasser (Eds.), Comparative neurobiology: Modes ofcommunication in the nervous system (pp. 105–116). New York: Wiley.

Merzenich, M. M., Jenkins, W. M., Johnston, P., Schreiner, C., Miller,S. L., & Tallal, P. (1996, January 5). Temporal processing deficits oflanguage-learning impaired children ameliorated by training. Science,271(5245), 77–81.

Mesulam, M. M. (1990). Large-scale neurocognitive networks and distrib-uted processing for attention, language, and memory. Annals of Neurol-ogy, 28(5), 597–613.

Mesulam, M. M. (1998). From sensation to cognition. Brain, 121(Pt 6),1013–1052.

Metzner, W. (1999). Why are there so many sensory brain maps? Cellularand Molecular Life Science, 56(1–2), 1–4.

Metzner, W., & Juranek, J. (1997). A sensory brain map for each behavior?Proceedings of the National Academy of Sciences of the United States ofAmerica, 94(26), 14798–14803.

Milgram, N. W. (2003). Cognitive experience and its effect on age-dependent cognitive decline in beagle dogs. Neurochemical Research,28(11), 1677–1682.

Milgram, N. W., Siwak-Tapp, C. T., Araujo, J., & Head, E. (2006).Neuroprotective effects of cognitive enrichment. Ageing Research Re-views, 5(3), 354–369.

Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontalcortex function. Annual Review of Neuroscience, 24, 167–202.

Miller, G. F., & Penke, L. (2007). The evolution of human intelligence andthe coefficient of additive genetic variance in human brain size. Intelli-gence, 35, 97–114.

Miller, N., & Dollard, J. (1941). Social learning and imitation. NewHaven, NJ: Yale University Press.

Milner, B. (1982). Some cognitive effects of frontal-lobe lesions in man.Philosophical Transactions of the Royal Society of London, Series B:Biological Sciences, 298, 211–226.

Mountcastle, V. B. (1957). Modality and topographic properties of singleneurons of cat’s somatic sensory cortex. Journal of Neurophysiology, 20,403–434.

Mountcastle, V. B. (1997). The columnar organization of the neocortex.Brain, 120(Pt 4), 701–722.

Mountcastle, V. B. (1998). Perceptual neuroscience: The cerebral cortex.Cambridge, MA: Harvard University Press.

Naatanen, R., Tervaniemi, M., Sussman, E., Paavilainen, P., & Winkler, I.(2001). ‘Primitive intelligence’ in the auditory cortex. Trends in Neuro-sciences, 24, 283–288.

Neisser, U., Boodoo, G., Bouchard, T. J., Boykin, A. W., Brody, N., Ceci,S. J., et al. (1996). Intelligence: Knowns and unknowns. AmericanPsychologist, 51, 77–101.

Neubauer, A. C., Fink, A., & Schrausser, D. G. (2002). Intelligence andneural efficiency: The influence of task content and sex on the brain–IQrelationship. Intelligence, 30, 515–536.

Neubauer, A. C., Freudenthaler, H. H., & Pfurtscheller, G. (1995). Intel-ligence and spatiotemporal patterns of event-related descynchronization(ERD). Intelligence, 20, 249–266.

Northcutt, R. G., & Kaas, J. H. (1995). The emergence and evolution ofmammalian neocortex. Trends in Neurosciences, 18(9), 373–379.

Ohta, H., Matsumoto, K., & Watanabe, H. (1993). Impairment of acqui-sition but not retention of a simple operant discrimination performancein aged Fischer 344 rats. Physiology and Behavior, 54(3), 443–448.

Orduna, I., Mercado, E., III, Gluck, M. A., & Merzenich, M. M. (2005).Cortical responses predict perceptual sensitivities to complex sounds inrats. Behavioral Neuroscience, 119, 256–264.

Pascual-Leone, A., Amedi, A., Fregni, F., & Merabet, L. B. (2005). Theplastic human brain cortex. Annual Review of Neuroscience, 28, 377–401.

Pasternak, T., & Greenlee, M. W. (2005). Working memory in primatesensory systems. Nature Reviews Neuroscience, 6(2), 97–107.

Pauli, P., Lutzenberger, W., Rau, H., Birbaumer, N., Rickard, T. C.,Yaroush, R. A., et al. (1994). Brain potentials during mental arithmetic:Effects of extensive practice and problem difficulty. Brain ResearchCognitive Brain Research, 2(1), 21–29.

Pavlov, I. P. (1927). Conditioned reflexes: An investigation of the physi-ological activity of the cerebral cortex. London: Oxford UniversityPress.

Petrides, M. (1996). Specialized systems for the processing of mnemonicinformation within the primate frontal cortex. Philosophical Transac-tions of the Royal Society of London, Series B: Biological Sciences,351(1346), 1455–1461.

Piaget, J. (1953). The origins of intelligence in children. London: Rout-ledge and Kegan Paul.

Pinker, S. (1994). The language instinct: How the mind creates language.New York: William Morrow & Co.

Plomin, R. (2001). The genetics of g in human and mouse. Nature ReviewsNeuroscience, 2(2), 136–141.

Poldrack, R. A., Clark, J., Pare-Blagoev, E. J., Shohamy, D., CresoMoyano, J., Myers, C., et al. (2001, November 29). Interactive memorysystems in the human brain. Nature, 414(6863), 546–550.

Poldrack, R. A., Prabhakaran, V., Seger, C. A., & Gabrieli, J. D. (1999).Striatal activation during acquisition of a cognitive skill. Neuropsychol-ogy, 13(4), 564–574.

Polley, D. B., Kvasnak, E., & Frostig, R. D. (2004, May 6). Naturalisticexperience transforms sensory maps in the adult cortex of caged animals.Nature, 429(6987), 67–71.

Prabhakaran, V., Rypma, B., & Gabrieli, J. D. (2001). Neural substrates ofmathematical reasoning: A functional magnetic resonance imaging study

134 MERCADO

of neocortical activation during performance of the necessary arithmeticoperations test. Neuropsychology, 15(1), 115–127.

Prabhakaran, V., Smith, J. A., Desmond, J. E., Glover, G. H., & Gabrieli,J. D. (1997). Neural substrates of fluid reasoning: An fMRI study ofneocortical activation during performance of the Raven’s ProgressiveMatrices Test. Cognitive Psychology, 33(1), 43–63.

Premack, D. (1983a). Animal cognition. Annual Review of Psychology, 34,351–362.

Premack, D. (1983b). The codes of man and beasts. Behavioral and BrainSciences, 6, 125–137.

Premack, D. (2004, January 16). Is language the key to human intelli-gence? Science, 303(5656), 318–320.

Raichle, M. E. (1998). The neural correlates of consciousness: An analysisof cognitive skill learning. Philosophical Transactions of the RoyalSociety of London, Series B: Biological Sciences, 353(1377), 1889–1901.

Rakic, P. (1988, July 8). Specification of cerebral cortical areas. Science,241(4862), 170–176.

Rakic, P. (1995). A small step for the cell, a giant leap for mankind: Ahypothesis of neocortical expansion during evolution. Trends in Neuro-sciences, 18(9), 383–388.

Ram´on y Cajal, S. (1904). La textura del sistema nerviosa del hombre ylos vertebrados [Texture of the nervous system of man and the verte-brates]. Madrid, Spain: Moya.

Ramsey, N. F., Jansma, J. M., Jager, G., Van Raalten, T., & Kahn, R. S.(2004). Neurophysiological factors in human information processingcapacity. Brain, 127(Pt 3), 517–525.

Ratcliffe, J. M., Fenton, M. B., & Shettleworth, S. J. (2006). Behavioralflexibility positively correlated with relative brain volume in predatorybats. Brain Behavior and Evolution, 67(3), 165–176.

Raz, N., Willerman, L., & Yama, M. (1987). On sense and senses:Intelligence and auditory information processing. Personality and Indi-vidual Differences, 8, 201–210.

Reader, S. M., & Laland, K. N. (2002). Social intelligence, innovation, andenhanced brain size in primates. Proceedings of the National Academyof Sciences of the United States of America, 99(7), 4436–4441.

Recanzone, G. H., Schreiner, C. E., & Merzenich, M. M. (1993). Plasticityin the frequency representation of primary auditory cortex followingdiscrimination training in adult owl monkeys. Journal of Neuroscience,13(1), 87–103.

Reed, T. E. (1993). Effect of enriched (complex) environment on nerveconduction velocity: New data and review of implications for the speedof information processing. Intelligence, 17, 533–540.

Reed, T. E. (2004). Confirmation of correlation between brain nerveconduction velocity and intelligence level in normal adults. Intelligence,32, 563–572.

Reichle, E. D., Carpenter, P. A., & Just, M. A. (2000). The neural bases ofstrategy and skill in sentence-picture verification. Cognitive Psychology,40(4), 261–295.

Rendell, L., & Whitehead, H. (2001). Culture in whales and dolphins.Behavioral and Brain Sciences, 24(2), 309–324.

Rensch, B. (1956). Increase in learning capability with increase in brainsize. American Naturalist, 90, 81–85.

Rescorla, R. A., & Wagner, A. R. (1972). A theory of Pavlovian condi-tioning: Variations in the effectiveness of reinforcement and nonrein-forcement. In A. H. Black & W. F. Prokasy (Eds.), Classical condition-ing II: Current research and theory (pp. 64–99). New York: Appleton-Century-Crofts.

Riddell, W. I., & Corl, K. G. (1977). Comparative investigation of therelationship between cerebral indices and learning abilities. Brain Be-havior and Evolution, 14(6), 385–398.

Rizzolatti, G., & Arbib, M. A. (1998). Language within our grasp. Trendsin Neurosciences, 21(5), 188–194.

Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. AnnualReview of Neuroscience, 27, 169–192.

Rogers, W. A., Fisk, A. D., & Hertzog, C. (1994). Do ability–performancerelationships differentiate age and practice effects in visual search?Journal of Experimental Psychology: Learning, Memory, and Cogni-tion, 20(3), 710–738.

Roitblat, H. L., & von Fersen, L. (1992). Comparative cognition: Repre-sentation and processes in learning and memory. Annual Review ofPsychology, 43, 671–710.

Romo, R., & Salinas, E. (2003). Flutter discrimination: Neural codes,perception, memory and decision making. Nature Reviews Neuro-science, 4(3), 203–218.

Rosenbaum, D. A., Carlson, R. A., & Gilmore, R. O. (2001). Acquisitionof intellectual and perceptual–motor skills. Annual Review of Psychol-ogy, 52, 453–470.

Rosenzweig, M. R. (1966). Environmental complexity, cerebral change,and behavior. American Psychologist, 21, 321–332.

Roth, G., & Dicke, U. (2005). Evolution of the brain and intelligence.Trends in Cognitive Sciences, 9(5), 250–257.

Rumbaugh, D. M. (1970). Learning skills of anthropoids. In L. Rosenblum(Ed.), Primate behavior: Developments in field and laboratory research(pp. 1–70). New York: Academic Press.

Rumbaugh, D. M., & McCormack, D. (1967). The learning skills ofprimates: A comparative study of apes and monkeys. In D. Stark, R.Schneider, & H. J. Kuhn (Eds.), Progress in primatology (pp. 289–306).Stuggart, Germany: Gustav Fischer.

Rumbaugh, D. M., & Washburn, D. A. (2003). Intelligence of apes andother rational beings. New Haven, CT: Yale University Press.

Salthouse, T. A. (1992). Working-memory mediation of adult age differ-ences in integrative reasoning. Memory and Cognition, 20(4), 413–423.

Salthouse, T. A. (1996). The processing-speed theory of adult age differ-ences in cognition. Psychological Review, 103(3), 403–428.

Salthouse, T. A. (2000). Aging and measures of processing speed. Biolog-ical Psychology, 54(1–3), 35–54.

Sarter, M., Bruno, J. P., & Turchi, J. (1999). Basal forebrain afferentprojections modulating cortical acetylcholine, attention, and implica-tions for neuropsychiatric disorders. Annals of the New York Academy ofSciences, 877, 368–382.

Sarter, M., Hasselmo, M. E., Bruno, J. P., & Givens, B. (2005). Unravelingthe attentional functions of cortical cholinergic inputs: Interactions be-tween signal-driven and cognitive modulation of signal detection. BrainResearch Brain Research Reviews, 48(1), 98–111.

Schrier, A. M. (1966). Learning-set formation by three species of macaquemonkeys. Journal of Comparative and Physiological Psychology, 61(3),490–492.

Sharma, J., Angelucci, A., & Sur, M. (2000, April 20). Induction of visualorientation modules in auditory cortex. Nature, 404(6780), 841–847.

Shaw, P., Greenstein, D., Lerch, J., Clasen, L., Lenroot, R., Gogtay, N., etal. (2006, March 30). Intellectual ability and cortical development inchildren and adolescents. Nature, 440(7084), 676–679.

Sherry, D. F. (2006). Neuroecology. Annual Review of Psychology, 57,167–197.

Singer, T., Lindenberger, U., & Baltes, P. B. (2003). Plasticity of memoryfor new learning in very old age: A story of major loss? Psychology andAging, 18(2), 306–317.

Slotnick, B., Hanford, L., & Hodos, W. (2000). Can rats acquire anolfactory learning set? Journal of Experimental Psychology: AnimalBehavior Processes, 26(4), 399–415.

Sol, D., Duncan, R. P., Blackburn, T. M., Cassey, P., & Lefebvre, L.(2005). Big brains, enhanced cognition, and response of birds to novelenvironments. Proceedings of the National Academy of Sciences of theUnited States of America, 102(15), 5460–5465.

Spearman, C. (1904). “General intelligence,” Objectively determined andmeasured. The American Journal of Psychology, 15, 201–292.

135NEURAL AND COGNITIVE PLASTICITY

Sternberg, R. J., & Kaufman, J. C. (Eds.). (2002). The evolution ofintelligence. Mahwah, NJ: Lawrence Erlbaum Associates.

Stevens, C. F. (2001, May 10). An evolutionary scaling law for the primatevisual system and its basis in cortical function. Nature, 411(6834),193–195.

Stevens, C. F. (2002). Predicting functional properties of visual cortexfrom an evolutionary scaling law. Neuron, 36(1), 139–142.

Stiles, J. (2000). Neural plasticity and cognitive development. Develop-mental Neuropsychology, 18(2), 237–272.

Striedter, G. F. (2006). Precis of principles of brain evolution. Behavioraland Brain Sciences, 29(1), 1–12.

Suga, N., & Ma, X. (2003). Multiparametric corticofugal modulation andplasticity in the auditory system. Nature Reviews Neuroscience, 4(10),783–794.

Sutherland, N. S., & Mackintosh, N. J. (1971). Mechanisms of animaldiscrimination learning. New York: Academic Press.

Swanson, L. W. (2003). Brain architecture: Understanding the basic plan.New York: Oxford University Press.

Tallal, P., Miller, S. L., Bedi, G., Byma, G., Wang, X., Nagarajan, S. S., etal. (1996, January 5). Language comprehension in language-learningimpaired children improved with acoustically modified speech. Science,271(5245), 81–84.

Tang, Y. P., Shimizu, E., Dube, G. R., Rampon, C., Kerchner, G. A., Zhuo,M., et al. (1999). Genetic enhancement of learning and memory in mice.Nature, 401(6748), 63–69.

Temple, E., Deutsch, G. K., Poldrack, R. A., Miller, S. L., Tallal, P.,Merzenich, M. M., et al. (2003). Neural deficits in children with dyslexiaameliorated by behavioral remediation: Evidence from functional MRI.Proceedings of the National Academy of Sciences of the United States ofAmerica, 100(5), 2860–2865.

Thatcher, R. W., North, D., & Biver, C. (2005). EEG and intelligence:Relations between EEG coherence, EEG phase delay and power. Clin-ical Neurophysiology, 116(9), 2129–2141.

Thomas, R. K. (1996). Investigating cognitive abilities in animals: Unre-alized potential. Brain Research Cognitive Brain Research, 3(3–4),157–166.

Thompson, P. M., Cannon, T. D., Narr, K. L., van Erp, T., Poutanen, V. P.,Huttunen, M., et al. (2001). Genetic influences on brain structure. NatureNeuroscience, 4(12), 1253–1258.

Thompson, R., Crinella, F. M., & Yu, J. (1990). Brain mechanisms inproblem solving and intelligence. New York: Plenum Press.

Thompson, R. F. (1965). The neural basis of stimulus generalization. InD. I. Mostofsky (Ed.), Stimulus generalization (pp. 154–178). Stanford,CA: Stanford University Press.

Thompson, R. K., Oden, D. L., & Boysen, S. T. (1997). Language-naivechimpanzees (Pan troglodytes) judge relations between relations in aconceptual matching-to-sample task. Journal of Experimental Psychol-ogy: Animal Behavior Processes, 23(1), 31–43.

Thorndike, E. L. (1911). Animal intelligence. New York: Macmillan.Thorndike, E. L. (1935). Organization of behavior in the albino rat. Genetic

Psychology Monographs, 17, 1–70.Thorndike, E. L., Lay, W., & Dean, P. R. (1909). The relation of accuracy

in sensory discrimination to general intelligence. American Journal ofPsychology, 20, 364–369.

Turner, J. G., Hughes, L. F., & Caspary, D. M. (2005). Affects of aging onreceptive fields in rat primary auditory cortex layer V neurons. Journalof Neurophysiology, 94(4), 2738–2747.

VanLehn, K. (1996). Cognitive skill acquisition. Annual Review of Psy-chology, 47, 513–539.

van Praag, H., Kempermann, G., & Gage, F. H. (2000). Neural conse-quences of environmental enrichment. Nature Reviews Neuroscience,1(3), 191–198.

Verhaeghen, P., & Marcoen, A. (1996). On the mechanisms of plasticity in

young and older adults after instruction in the method of loci: Evidencefor an amplification model. Psychology and Aging, 11(1), 164–178.

Verhaeghen, P., Marcoen, A., & Goossens, L. (1992). Improving memoryperformance in the aged through mnemonic training: A meta-analyticstudy. Psychology and Aging, 7(2), 242–251.

Vicente, A., & Martinez-Manrique, F. (2005). Semantic underdetermina-tion and the cognitive uses of language. Mind and Language, 20,537–558.

Volkmar, F. R., & Greenough, W. T. (1972, June 30). Rearing complexityaffects branching of dendrites in the visual cortex of the rat. Science,176(42), 1445–1447.

Volman, S. F., Grubb, T. C., & Schuett, K. C. (1997). Relative hippocam-pal volume in relation to food-storing behavior in four species ofwoodpeckers. Brain Behavior and Evolution, 49, 110–120.

Waddell, S., & Quinn, W. G. (2001). Flies, genes, and learning. AnnualReview of Neuroscience, 24, 1283–1309.

Wall, J. T., Xu, J., & Wang, X. (2002). Human brain plasticity: Anemerging view of the multiple substrates and mechanisms that causecortical changes and related sensory dysfunctions after injuries of sen-sory inputs from the body. Brain Research Brain Research Reviews,39(2–3), 181–215.

Wallis, J. D., & Miller, E. K. (2003). From rule to response: Neuronalprocesses in the premotor and prefrontal cortex. Journal of Neurophys-iology, 90(3), 1790–1806.

Warren, J. M. (1965). Primate learning in comparative perspective. InA. M. Schrier, H. F. Harlow, & F. Stollnitz (Eds.), Behavior of nonhu-man primates: Modern research trends (pp. 249–281). New York:Academic Press.

Watson, B. U. (1991). Some relationships between intelligence and audi-tory discrimination. Journal of Speech and Hearing Research, 34, 621–627.

Weinberger, N. M. (2003). The nucleus basalis and memory codes: Audi-tory cortical plasticity and the induction of specific, associative behav-ioral memory. Neurobiology of Learning and Memory, 80(3), 268–284.

Weinberger, N. M. (2004). Specific long-term memory traces in primaryauditory cortex. Nature Reviews Neuroscience, 5(4), 279–290.

Weinberger, N. M., Ashe, J. H., Metherate, R., McKenna, T. M., Diamond,D. M., & Bakin, J. S. (1990). Retuning auditory cortex by learning: Apreliminary model of receptive field plasticity. Concepts in Neuro-science, 1, 91–131.

Wenk, G. L. (1997). The nucleus basalis magnocellularis cholinergicsystem: One hundred years of progress. Neurobiology of Learning andMemory, 67(2), 85–95.

West, R. L. (1996). An application of prefrontal cortex function theory tocognitive aging. Psychological Bulletin, 120(2), 272–292.

Whitehouse, P. J., Price, D. L., Struble, R. G., Clark, A. W., Coyle, J. T.,& Delon, M. R. (1982, March 5). Alzheimer’s disease and seniledementia: Loss of neurons in the basal forebrain. Science, 215(4537),1237–1239.

Whiten, A., Horner, V., Litchfield, C. A., & Marshall-Pescini, S. (2004).How do apes ape? Learning and Behavior, 32(1), 36–52.

Wiesel, T. N., & Hubel, D. H. (1963). Single-cell responses in striatecortex of kittens deprived of vision in one eye. Journal of Neurophysi-ology, 26, 1003–1017.

Wilke, M., Sohn, J. H., Byars, A. W., & Holland, S. K. (2003). Brightspots: Correlations of gray matter volume with IQ in a normal pediatricpopulation. NeuroImage, 20(1), 202–215.

Wilson, M. (2001). Perceiving imitatible stimuli: Consequences of isomor-phism between input and output. Psychological Bulletin, 127(4), 543–553.

Wilson, M., & Knoblich, G. (2005). The case for motor involvement inperceiving conspecifics. Psychological Bulletin, 131(3), 460–473.

Wissler, C. (1901). The correlation of mental and physical tests. Psycho-logical Monographs, 3, 1–62.

136 MERCADO

Witelson, S. F., Beresh, H., & Kigar, D. L. (2006). Intelligence and brainsize in 100 postmortem brains: Sex, lateralization and age factors. Brain,129(Pt 2), 386–398.

Wood, J. N., & Grafman, J. (2003). Human prefrontal cortex: Processingand representational perspectives. Nature Reviews Neuroscience, 4(2),139–147.

Yang, L., Krampe, R. T., & Baltes, P. B. (2006). Basic forms of cognitiveplasticity extended into the oldest-old: Retest learning, age, and cogni-tive functioning. Psychology and Aging, 21(2), 372–378.

Yokoyama, C., Tsukada, H., Watanabe, Y., & Onoe, H. (2005). A dynamicshift of neural network activity before and after learning-set formation.Cerebral Cortex, 15(6), 796–801.

Yoo, S. S., O’Leary, H. M., Fairneny, T., Chen, N. K., Panych, L. P., Park,H., et al. (2006). Increasing cortical activity in auditory areas throughneurofeedback functional magnetic resonance imaging. NeuroReport,17(12), 1273–1278.

Young, E. D. (1997). Parallel processing in the nervous system: Evidencefrom sensory maps. Proceedings of the National Academy of Sciences ofthe United States of America, 94(3), 933–934.

Yu, A. J., & Dayan, P. (2005). Uncertainty, neuromodulation, and atten-tion. Neuron, 46(4), 681–692.

Yuste, R., & Bonhoeffer, T. (2001). Morphological changes in dendriticspines associated with long-term synaptic plasticity. Annual Review ofNeuroscience, 24, 1071–1089.

Zaborszky, L. (2002). The modular organization of brain systems. Basalforebrain: The last frontier. Progress in Brain Research, 136, 359–372.

Zaborszky, L., Pang, K., Somogyi, J., Nadasdy, Z., & Kallo, I. (1999). Thebasal forebrain corticopetal system revisited. Annals of the New YorkAcademy of Science, 877, 339–367.

Zahm, D. S. (2006). The evolving theory of basal forebrain functional–anatomical ‘macrosystems’. Neuroscience and Biobehavioral Reviews,30(2), 148–172.

Zentall, T. R. (2006). Imitation: Definitions, evidence, and mechanisms.Animal Cognition, 9(4), 335–353.

Zentall, T. R., & Akins, C. (2001). Imitation in animals: Evidence, func-tion, and mechanisms. In R. G. Cook (Ed.), Avian visual cognition.Retrieved December 3, 2005, from http://www.pigeon.psy.tufts.edu/avc/zentall/

Zhang, L. I., Bao, S., & Merzenich, M. M. (2001). Persistent and specificinfluences of early acoustic environments on primary auditory cortex.Nature Neuroscience, 4(11), 1123–1130.

Zhang, L. I., Bao, S., & Merzenich, M. M. (2002). Disruption of primaryauditory cortex by synchronous auditory inputs during a critical period.Proceedings of the National Academy of Sciences of the United States ofAmerica, 99(4), 2309–2314.

Received December 18, 2006Revision received August 13, 2007

Accepted August 28, 2007 �

137NEURAL AND COGNITIVE PLASTICITY