Intelligence

What Is Intelligence?

• Intelligence is difficult to define.

• Intelligence can be legitimately described at three levels of analysis:– As consisting of one thing

– As consisting of a few things

– As consisting of many things

Intelligence as a Single Trait

• Is intelligence a single entity that influences all aspects of cognitive functioning?– Maybe each of us possess a certain amount of g, or general

intelligence, that influences our ability on all intellectual tasks.

– Support: Overall scores on intelligence tests correlate positively with school grades and achievement test performance and with speed of information processing.

Intelligence as a Few Basic Abilities

• Are there two types of intelligence (Cattell)?

• Crystallized intelligence:

• Factual knowledge about the world, word meanings, arithmetic, etc.

• Fluid intelligence:

• The ability to think on the spot by drawing inferences and understanding relations between concepts not previously encountered.

• Measured on IQ tests by object assembly, analogies, and identification tasks.

Intelligence as a Few Basic Abilities

• Thurstone’s seven primary mental abilities:

• Word fluency, verbal meaning, reasoning, spatial visualization, numbering, rote memory, and perceptual speed.

• Support for Thurstone’s view:

• Performance on various tests of a single ability tend to be more similar than performance on tests that are dissimilar.

• Difference from Cattell’s view

• The seven-primary-ability view is more precise and complex than Cattell’s crystallized/fluid distinction.

Intelligence as Multiple Processes

• Others see intelligence as comprising many information processing tasks, such as:

• Attending

• Perceiving

• Encoding

• Associating

• Planning

• Reasoning

• Problem solving

• Generating strategies

• Language production and comprehension.

Resolving the Competing Perspectives

• John Carroll proposed the “three-stratum theory of intelligence,” a hierarchical integration consisting of

• g

• eight generalized abilities

• many specific processes

Carroll’s Three-Stratum Model of Intelligence

Measuring Intelligence: Intelligence Tests

• Intelligence means different things at different ages.

• E.g., you cannot measure an infant’s language ability.

• Wechsler Intelligence Test for Children (WISC) is the most widely used instrument for children 6 years and older.

• The WISC is divided into two main sections:

• Verbal: general knowledge and language skills (crystallized intelligence)

• Performance: spatial and perceptual abilities (fluid intelligence)

Intelligence Quotient (IQ)

• Intelligence tests like the WISC and the Stanford-Binet provide overall quantitative measures of a child’s intelligence relative to that of other children of same age, producing the Intelligence Quotient, or IQ.

Intelligence Quotient (IQ)

• IQ computation is based on a normal distribution of scores, a pattern of data in which scores fall symmetrically around a mean value, with most scores falling close to the mean and fewer scores at the high and low ends.

• The “mean” is the average of all scores normed.

• The mean has been arbitrarily set at 100.

• The normal distribution signifies that most scores are at the mean.

• IQ scores also reflect a “standard deviation,” a measure of variability of scores within a distribution. By definition, 68% of scores must be 1 SD below the mean and 1 SD above the mean.

• On most IQ tests, the SD is 15 points.

A Normal Distribution in IQ Scores

Stability of IQ Scores

• Scores are stable from the age of 5 onward.

• Children’s IQs at 5 have high correlation with their IQs at age 15.

• Scores, however, do show an average change, up or down a few points.

• Changes in IQ scores over time may be influenced by characteristics of children and their parents other than intelligence, such as interest in learning and academic success.

What Do IQ Scores Predict?

• IQ is a strong predictor of academic, economic, and occupational success.

• People with lower IQs are more likely to

• permanently drop out of high school

• be unemployed

• earn an income that fails to surpass the poverty line

• have children without being married

• smoke during pregnancy

• have low birth weight babies

• be on welfare

• be involved in crime

• divorce

Academic Achievement & Income

$6250

$10000

$13750

$17500

$21250

$2500020

0-24

925

0-29

930

0-34

935

0-39

940

0-49

945

0-49

950

0-54

955

0-59

960

0-64

965

0-69

970

0-74

975

0-80

0

• Correlation between IQ and academic achievement scores is about .85

• $25,000 in 1974 is equal to $93,300 in 2003

IQ adds to schooling

Quality of IQ as Predictor

• Generally, three properties seem key in determining the importance of a dimension of individual differences.

• Association with a broad range of qualities measured at the same time

• Stability over time

• Predictiveness of later outcomes in other areas

Intelligence and Individual Differences

Development of Intelligence

• Development of intelligence may be a product of

• Qualities of the child (inherited features)

• Qualities of the environment (shared and non-shared environment)

• home

• school

• economic-based environments (e.g., high-poverty neighborhoods with lots of public housing and high crime)

Genetic similarity and IQ

• About 50% of the variation in IQs is attributable to genetic variation.

• Genetic contribution relative to the environmental contribution is greater in older children than in younger children.

• IQs of adopted children increasingly correlate with their biological parents’ as the children get older.

SES and influence of genetic similarity

• Turkheimer et al. (2003) calculated heritability for twins who differed in SES

• Sample median family income $22,000 (in 1997 dollars); 1997 US median $53,000

PSYCHOLOGICAL SCIENCE

E. Turkheimer et al.

VOL. 14, NO. 6, NOVEMBER 2003

627

correlation was .51 and the MZ twin correlation was .87, consistentwith

h

2

of .72 and

c

2

of .15.

DISCUSSION

These findings suggest that a model in which variability in intelli-gence among children is partitioned into independent componentsattributable to genes and environments is too simple for the dynamicinteraction of genes and real-world environments during development.The relative importance of environmental differences in causing dif-ferences in observed intelligence appears to vary with the SES of thehomes in which children were raised. SES is a complex variable, how-ever, and the substantive interpretation to be placed on our results de-pends on an interpretation of what SES actually measures.

The most obvious interpretation of SES in this study is that it mea-sured the quality of the environment in which the children were bornand raised. Indeed, this is the function for which SES was intended. Un-der this interpretation, the observed interaction between SES and the bi-ometric components of IQ could be indicative of precisely the kind ofnonlinear relationship between rearing environment and intelligencethat has been suggested by Scarr (1981) and Jensen (1981), with differ-ences among poor environments contributing more to differences inphenotypic outcome than differences among middle-class or better envi-ronments contribute.

It would be naive, however, to interpret SES strictly as an environ-mental variable. Most variables traditionally thought of as markersof environmental quality also reflect genetic variability (Plomin &Bergeman, 1991). Children reared in low-SES households, therefore,may differ from more affluent children both environmentally and ge-netically (Gottesman, 1968), and the models we employed in thisstudy do not allow us to determine which aspect of SES is responsiblefor the interactions we observed. Indeed, it will be difficult to separate

the genetic and environmental aspects of SES or other measures of thefamily environment in research designs of this kind, because childrenraised in the same home necessarily have the same SES.

Genetic variability in SES might also introduce a complication tothe models themselves. Phenotypic SES and IQ are correlated, and thatcorrelation is potentially mediated both genetically and environmen-tally. Therefore, the models are attempting to detect an interaction be-tween genotype and environment in the presence of a correlation betweengenotype and environment, raising the concern that the presence of thecorrelation might introduce bias into the estimation of the interaction.However, Purcell (2003) has conducted an exhaustive series of simula-tions that suggest no bias is introduced, as long as the main effect ofthe moderating variable is included in the model, as we have donehere. The presence in the model of the main effect of SES means thatthe biometric model fitting is actually being conducted on the portionof IQ that is independent of both the genetic and environmental com-ponents of SES. (We note, however, that omitting the main effect fromthe model did not change the results to a significant degree.)

The developmental mechanisms underlying the effect remain un-clear. Although the models indicate that the A

!

, C

!

, and E

!

interactionsjointly contributed significant variance to differences in FSIQ andPIQ, the models were less able to distinguish which of the individualinteractions with A, C, and E was most important in the effect. The in-teraction could be mediated primarily along genetic pathways, mean-ing that genetic differences among individuals are accentuated in favorableenvironments, as has been theorized by Bronfenbrenner (Bronfenbrenner& Ceci, 1994). It could also be that the slope of the IQ

"

Environmentfunction is steeper at low levels of environment, as suggested by Scarr(1981) and Jensen (1981). Or maybe outcome simply becomes less pre-dictable in poor environments, leading to an increase in E variability,as we have suggested (Turkheimer & Waldron, 2000) based on otherevidence. To resolve this issue, it will be most important to study large

Fig. 3. Proportion of total Full-Scale IQ variance accounted for by A, C, and E plotted as a function of observed socioeconomic status (SES).Shading indicates 95% confidence intervals.

Sex and IQ

• Boys and girls have almost the same IQ scores.

• There are small differences in average performance between boys and girls in specific areas:

• Girls are more fluent in writing, perceptual speed, and verbal fluency

• Boys as a group are stronger in visual–spatial processing, science, and math problem solving.

• Pattern of sex differences in academic achievement is similar in many countries

HOME and IQ

• How much does the home environment matter to IQ scores?

• Home Observation for Measurement of the Environment (HOME) is a scale used to measure factors in the home environment that might affect children’s intellectual and environmental well-being

HOME and IQ

• Scores on HOME correlate with Child’s IQ

• Why?

Parents’IQ

Child’sIQ

HOMEscore

Parents’IQ

Child’sIQHOME

score

Some effect of HOME: No effect of HOME:

HOME and IQ

• Scores on HOME correlate with child IQ because

• Home environment is affected by parents’ genes

• Almost all studies using HOME have focused on biological parents

• With non-biological parents, scores on HOME have very low correlations with child IQ

Parents’IQ

Child’sIQ

HOMEscore

No effect of HOME:

Schooling and IQ

• School attendance makes children smarter.

• Average IQ scores rise during the school year and drop during the summer.

• Jumps between grade levels indicate schooling affects IQ in addition to age.

Relations of Age and Grade to IQ Scores

Poverty and IQ

• Poverty affects intelligence in several ways:

• Inadequate diet can disrupt brain development.

• Reduced access to health service, poor parenting, and insufficient stimulation and emotional support can impair intellectual growth.

• In all countries studied, children from wealthier homes scored higher on IQ than did children from poor homes.

• In countries where there is the greatest economic diversity, the diversity in IQ is the greatest.

Poverty and IQ

• The effect of poverty on IQ may be conceptualized in terms of ‘risk factors’

• Mother did not complete high school

• No father or stepfather in home

• Large number of stressful life events

• Maternal anxiety

Poverty and IQ

Programs for Helping Poor Children

• Programs that may work:

• Home-based programs: Focus on the improving the parenting skills of mothers.

• Center-based programs: Nursery schools with emphasis in teaching reading and arithmetic skills, reinforcement of learning, and providing stimulating environment.

Effectiveness of Programs

• Gains in IQ scores from participation in early intervention programs are short-lived

• There are other long-term effects:– Fewer children needed special education classes

– Fewer children were held back in school

– More program participants graduated from high school.

Academic Achievement

Academic achievement

• Research in cognitive development has led to understanding of basic cognitive processes involved in academic achievement

• These basic processes provide a natural target for intervention and assessment of progress

Reading

• Chall (1979): Description of the typical chronological progression

• Another stage theory...

Acquisition of Reading Skills

• Five stages of reading development

• Stage 0, birth through first grade:

• Acquiring skills for reading, including the letters of alphabet and phonemic awareness (identification of sounds within spoken word)..

• Stage 1, first and second grades:

• Acquisition of phonological recoding skills, the ability to translate letters into sounds and to blend the sounds into words ( (“sounding out”).

• Stage 2, second and third grades:

• Gaining fluency in reading simple material.

Acquisition of Reading Skills

• Five stages of reading development (continued)

• Stage 3, fourth through eighth grades:

• Developing the ability to acquire new information from print—“reading to learn, rather than learning to read” (as in earlier grades).

• Stage 4, eighth through twelfth grades:

• Obtaining information from reading and acquiring the ability to appreciate multiple perspectives and viewpoints.

Information-Processing Analysis

• Rapid, effortless identification of words is central to reading and the enjoyment of reading.

• Words can be identified by

• Phonological recoding: Converting the visual form of a word into a verbal, speechlike form

• Visually based (Orthographic) retrieval: Proceeding directly from the visual form of a word to its meaning.

Basic Processes as “Strategies”

• Children choose between these two word identification approaches through a “strategy-choice process,” in which they choose the fastest approach that is likely to be correct.

• On hard words, they go with the surer strategy.

• On easier words, they go on the fastest approach.

Strategy Choices in Reading

Dyslexics

A C Q U I S ITI O N O F A C A D E MI C S K IL L S : R E A D I N G , WRITI N G , A N D M AT H E M ATI C S 3 1

accurate word identification correlates positively with reading comprehension atall points from the first grade through adulthood (Saarnio, Oka, & Paris, 1990;Vellutino, 1991).

Development of reading comprehension also is aided by acquisition of strate-gies. For example, good readers proceed slowly when they need to master writtenmaterial thoroughly and speed up when they need only a rough sense of it(Pressley et al., 1992). Proficiency in making such adjustments develops surpris-ingly late, however. Even when 10-year-olds are told that some material is crucial

DTSI Graphics, Worth/Siegler Child Development Galleys

individual differences 8.3D y s l e xi aSome children who are of normal intelli-gence and who grow up with parents who encourage them to read neverthelessread very poorly. This inability to read welldespite normal intelligence, referred to asdyslexia, affects roughly 3% to 5% ofchildren in the United States (Rayner &Pollatsek, 1989).

Most children with dyslexia are poor atreading primarily because of a generalweakness at phonological processing.This weakness is evident in the children’spoor ability to discriminate betweenphonemes, their poor short-term memoryfor verbal material (as indicated, for ex-ample, by poor ability to recall an arbi-trary list of words), and their slow recallof the names of objects (Vellutino,Scanlon, & Spearing, 1995). Determiningthe sounds that go with vowels is espe-cially difficult for children with dyslexia,at least in English, where a single vowelcan be pronounced in many ways (con-sider the sounds that accompany a in“hate, ” “hat, ” “hall, ” and “hard”).Because of this poor phonological pro-cessing, dyslexic children have great diffi-culty mastering the letter–soundcorrespondences used in phonological re-coding (Shankweiler et al., 1995;Stanovich & Siegel, 1994). For example,as shown in the figure, when asked toread pseudowords such as parding,dyslexic 13- and 14-year-olds perform atthe same level as typical 7- and 8-year-olds (Siegel, 1993). As would be ex-pected from the strategy-choice modeldescribed earlier, this difficulty withphonological processing causes mostdyslexic children to be poor at visuallybased retrieval as well as at sounding outwords (Manis et al., 1996). The problemcan be a lasting one: individuals whohave poor phonological processing skillsin early elementary school usually are

poor readers as adults as well (Wagner etal., 1997).

Studies of brain functioning supportthe view that poor phonological process-ing is at the heart of dyslexia. Whendyslexic adults read, two areas of theirbrains are less active than the correspon-ding areas in typical adults reading thesame words (Shaywitz et al., 1998). Onesuch area, located toward the back of thebrain, is directly involved in phonologicalprocessing; the other area, more towardthe middle of the brain, is involved in in-tegrating visual and auditory data (in thiscase, integrating the letters on the pagewith accompanying sounds).

How can dyslexic children behelped? One tempting inference isto conclude that because these chil-dren have difficulty learning phon-ics, they would learn better throughan approach that de-emphasizesletter–sound relations and insteademphasizes either visually based re-trieval or reliance on context. Thesealternative methods work poorly,however (Lyon, 1995). There is sim-ply no substitute for being able tosound out unfamiliar words.

Instead, what seems to work best is toteach children with dyslexia to use strate-gies that enhance their phonological re-coding (Lovett et al., 1994). Effectivestrategies include drawing analogies toknown words with similar spellings; gen-erating alternative pronunciations of vow-els when the first attempt at sounding outdoes not yield a plausible word; and, withlong words, “peeling off” prefixes andsuffixes and then trying to identify therest of the word. Using such strategieshelps children with dyslexia to improvetheir reading-achievement scores andspelling (Lovett et al., 1994).

Age

7– 8 9–10 11–12 13–140

10

20

30

40

50

60

Num

ber

of p

seud

owor

ds id

entif

ied

corr

ectly

Children with reading disabilities

Typical children

Number of pseudowords identified cor-rectly by 7- to 14-year-olds with andwithout reading disabilities. Note that13- and 14-year-olds with reading dis-abilities correctly identified no morewords than typical 7- and 8-year-olds.The poor phonological recoding skillsof children with learning disabilitiesleads them to have special difficultywith pseudowords that, because theyare totally unfamiliar, can be pro-nounced only by using phonologicalrecoding. (Data from Siegel, 1993)

HCD_dummy_CH08_Caa 10/3/02 4:47 PM Page 31

• Dyslexics are much less likely to use phonological-recoding than non-dyslexics -- even when it is necessary to do so

Siegel, 1993

Proc. Natl. Acad. Sci. USAVol. 95, pp. 2636–2641, March 1998Neurobiology

Functional disruption in the organization of the brain for readingin dyslexiaSALLY E. SHAYWITZ*†, BENNETT A. SHAYWITZ*‡, KENNETH R. PUGH*§, ROBERT K. FULBRIGHT¶,R. TODD CONSTABLE¶, W. EINAR MENCL*§, DONALD P. SHANKWEILER§, ALVIN M. LIBERMAN§,PAWEL SKUDLARSKI¶, JACK M. FLETCHER!, LEONARD KATZ§, KAREN E. MARCHIONE*, CHERYL LACADIE¶,CHRISTOPHER GATENBY¶, AND JOHN C. GORE¶***Department of Pediatrics, ‡Department of Neurology, §Haskins Laboratories, ¶Department of Diagnostic Radiology, **Department of Applied Physics,Yale University School of Medicine, New Haven, CT 06520; and !Department of Pediatrics, University of Texas Medical School, Houston, TX 77030

Contributed by Alvin M. Liberman, January 9, 1998

ABSTRACT Learning to read requires an awareness thatspoken words can be decomposed into the phonologic con-stituents that the alphabetic characters represent. Such pho-nologic awareness is characteristically lacking in dyslexicreaders who, therefore, have difficulty mapping the alphabeticcharacters onto the spoken word. To find the location andextent of the functional disruption in neural systems thatunderlies this impairment, we used functional magnetic res-onance imaging to compare brain activation patterns indyslexic and nonimpaired subjects as they performed tasksthat made progressively greater demands on phonologic anal-ysis. Brain activation patterns differed significantly betweenthe groups with dyslexic readers showing relative underacti-vation in posterior regions (Wernicke’s area, the angulargyrus, and striate cortex) and relative overactivation in ananterior region (inferior frontal gyrus). These results supporta conclusion that the impairment in dyslexia is phonologic innature and that these brain activation patterns may providea neural signature for this impairment.

Speech enables its users to create an indefinitely large numberof words by combining and permuting a small number ofphonologic segments, the consonants and vowels that serve asthe natural constituents of the biologic specialization forlanguage. An alphabetic transcription brings this same abilityto readers but only as they connect its arbitrary characters(letters) to the phonologic segments they represent. Makingthat connection requires an awareness that all words, in fact,can be decomposed into phonologic segments. Thus, it is thisawareness that allows the reader to connect the letter strings(the orthography) to the corresponding units of speech (pho-nologic constituents) that they represent. As numerous studieshave shown, however, such awareness is largely missing indyslexic children and adults (1–4). Not surprisingly, then,perhaps the most sensitive measure of the reading problem indyslexia is inability to read phonologically legal nonsensewords (5–7). As for why dyslexic readers should have excep-tional difficulty developing phonologic awareness, there issupport for the notion that the difficulty resides in the pho-nologic component of the larger specialization for language(8–10). If that component is imperfect, its representations willbe less than ideally distinct and, therefore, harder to bring toconscious awareness.

Previous efforts using functional imaging methods to exam-ine brain organization in dyslexia have been inconclusive(11–17) largely, we think, because the experimental taskstapped the several aspects of the reading process in somewhat

unsystematic ways. Our aim therefore was to develop a set ofhierarchically structured tasks that control the kind of lan-guage-relevant coding required, including especially the de-mand on phonologic analysis, and then to compare the per-formance and brain activation patterns (as measured by func-tional MRI) of dyslexic (DYS) and nonimpaired (NI) readers.Thus, proceeding from the base of the hierarchy to the top, thetasks made demands on visual–spatial processing, ortho-graphic processing, simple phonologic analysis, complex pho-nologic analysis, and lexical–semantic judgment. We hypoth-esized that differences in brain activation patterns wouldemerge as DYS and NI readers were asked to perform tasksthat make progressively greater demands on phonologic anal-ysis.

METHODSTasks. Both the decision and response components of the

tasks were comparable; in each instance the subject viewed twosimultaneously presented stimulus displays, one above theother, and was asked to make a same"different judgment bypressing a response button if the displays matched on a givencognitive dimension, such as line orientation judgment, lettercase judgment, single-letter rhyme, nonword rhyme, and cat-egory judgment. As noted above, the five tasks were orderedhierarchically. (i) At the lowest level, the line orientation (L)judgment task (e.g., Do [\\\/] and [\\\/] match?) taps visual–spatial processing but makes no orthographic demands. (ii)Next, the letter-case (C) judgment task (e.g., Do [bbBb] and[bbBb] match in the pattern of upper- and lowercase letters?)adds an orthographic processing demand but makes no pho-nologic demands, because the stimulus items that consistentirely of consonant strings are, therefore, phonotacticallyimpermissible. (iii) The third task, single letter rhyme (SLR;e.g., Do the letters [T] and [V] rhyme?), although orthograph-ically more simple than C, adds a phonologic processingdemand, requiring the transcoding of the letters (orthography)into phonologic structures and then requiring a phonologicanalysis of those structures sufficient to determine that they door do not rhyme. (iv) The fourth task, nonword rhyme (NWR;e.g., Do [leat] and [jete] rhyme?), requires analysis of morecomplex structures. (v) The final task, semantic category (SC)judgment (e.g., Are [corn] and [rice] in the same category?),also makes substantial demands on transcoding from print tophonology (18, 19) but requires in addition that the printed

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked ‘‘advertisement’’ inaccordance with 18 U.S.C. §1734 solely to indicate this fact.

© 1998 by The National Academy of Sciences 0027-8424"98"952636-6$2.00"0PNAS is available online at http:""www.pnas.org.

Abbreviations: fMRI, functional MRI; DYS, dyslexic; NI, nonim-paired; L, line; C, case; SLR, single letter rhyme; NWR, nonwordrhyme; SC, semantic category; ROI, regions of interest; BA, Brod-mann’s area; IFG, inferior frontal gyrus; STG, superior temporalgyrus; ILES, inferior lateral extrastriate.†To whom reprint requests should be addressed at: Department ofPediatrics, Yale University School of Medicine, P.O. Box 3333, NewHaven, CT 06520-8064. e-mail: Sally.Shaywitz@Yale.edu.

2636

for reading. In addition, an anterior region, the IFG, demon-strates significant differences in the pattern of activationbetween NI and DYS readers (Figs. 1 Lower and 2). However,in this case, in contrast to findings in the posterior system, DYScompared with NI readers demonstrate greater activation inresponse to increasing phonologic decoding demands. Anadditional anterior–frontal brain region, BA 46!47!11, dem-onstrates a pattern of activation across tasks comparable to thecontiguous IFG, although the group difference was marginal.

Hemispheric differences between NI and DYS readers havelong been suspected (13, 14, 31, 32), and these were found intwo regions: the angular gyrus and BA 37. According to thelogic of the statistical analytic strategy, we first looked for andfound an overall significant reading group–hemisphere–ROIinteraction. We then looked for reading group–hemisphereinteractions in each ROI. First an overall significant readinggroup–hemisphere–ROI interaction was obtained [F(16, 912)! 2.14; P " 0.05] and then significant reading group–hemisphere interactions were found in two regions: the angu-lar gyrus (BA 39) [F(1, 57) ! 5.04; P " 0.05] and BA 37 [F(1,57) ! 7.88; P " 0.01]. The task–reading group interaction inthe angular gyrus described above (showing anomalous activ-ity across tasks for DYS readers) was not further qualified byhemispheric differences; hence, these two different readinggroup effects in this ROI appear to be orthogonal to oneanother. BA 37 encompasses the posterior aspect of theinferior and middle temporal gyri and anterior aspect of the

lateral occipital gyrus (Talairach coordinates 44, #66, 21). Ineach case, activations in NI readers were greater in lefthemisphere and, in contrast, in DYS readers activations inthese regions were greater in the right hemisphere. Thispattern was observed across all tasks. On the basis of ourearlier work (33), we examined for hemispheric differencesbetween males and females. In the IFG, a significant sexdifference was found [sex–hemisphere–task interaction: F(3,171) ! 3.37; P " 0.025]. During NWR, men showed signifi-cantly greater activation in the left hemisphere compared withright, and women showed relatively greater right hemisphereactivation, consistent with previous observations.

DISCUSSIONIn this study we found significant differences in brain activationpatterns between DYS and NI readers, differences thatemerge during tasks that make progressive demands on pho-nologic analysis. These findings relate the cognitive!behavioral deficit characterizing DYS readers to anomalousactivation patterns in both posterior and anterior brain regions(Fig. 3).Thus, within a large posterior cortical system includingWernicke’s area, the angular gyrus, the extrastriate and striatecortex, DYS readers fail to systematically increase activation asthe difficulty of mapping print onto phonologic structuresincreases. In contrast, in anterior regions including the IFGand BA 46!47!11, dyslexic readers show a pattern of overac-

FIG. 1. Number of activated pixels for brain regions where activation patterns across tasks differ significantly between NI and DYS readers.Activations (mean $ SEM) are shown on ordinate and tasks are on abscissa. We performed an overall ANOVA and followed up those interactionsthat were significant (minimizing type I error). Data are also shown for regions with marginal P values (minimizing type II error). Significance levelsof the task by group effect (Huynh–Feldt corrected P values): STG, F(3, 171) ! 4.3 and P ! 0.009; BA 17, F(3, 171) ! 4.0 and P ! 0.012; IFG,F(3, 171) ! 3.8 and P ! 0.012; angular gyrus, F(3, 171) ! 2.7 and P ! 0.054; BA 46!47!11, F(3, 171) ! 2.4 and P ! 0.071; ILES, F(3, 171) !2.2 and P ! 0.094. The six anatomic regions (with center or ROI given in x, y, and z coordinates of Talairach) are (i) posterior STG, BA 22 (53,#43, 11); (ii) angular gyrus, BA 39, angular gyrus of the inferior parietal lobule (47, #45, 33); (iii) ILES, BA 18, 19, inferior occipital gyrus, inferioraspect of lateral occipital gyrus (36, #80, #5); (iv) BA 17, striate cortex (8, #89, 3); (v) IFG, BA 44 posterior aspect (pars operculum) of IFGand BA 45 middle aspect (pars triangularis) of IFG (47, 18, 18); (vi) BA 47, 11, 46, anterior inferior aspect of IFG, lateral and medial orbital gyri,and superior aspect of IFG and inferior aspect of middle frontal gyrus (33, 36, 0). Coordinates are shown for right hemisphere where x is positive(x is negative for left hemisphere).

2638 Neurobiology: Shaywitz et al. Proc. Natl. Acad. Sci. USA 95 (1998)

tivation in response to even the simplest phonologic task (SLR;Fig. 1). For NI readers, these data provide functional evidenceof a widely distributed computational system for readingcharacterized by specialization and reciprocity: within thesystem, task-specific responses vary from region to region. Forexample, in the IFG only the complex phonologic task (NWR)produced a significant increase in activation relative to theorthographic (C) task, suggesting that this region is engaged inletter to sound transcoding; in Wernicke’s area both simple(SLR) and more complex (NWR) phonologic tasks producedsignificant increases in activation relative to the orthographictask, implying that this region processes information in a moreabstract phonological form (Fig. 1).

These data help to reconcile the seemingly contradictoryfindings of previous imaging studies of dyslexia, some of whichinvolved anomalous findings in the visual system (15) andothers indicated abnormal activation within components of thelanguage system (11–14, 16, 17). Our data indicate that dyslexicreaders demonstrate a functional disruption in an extensivesystem in posterior cortex encompassing both traditional visualand traditional language regions and a portion of associationcortex. The involvement of this latter region centered aboutthe angular gyrus is of particular interest because this portionof association cortex is considered pivotal in carrying out thosecross-modal integrations necessary for reading [i.e., mappingthe visual percept of the print onto the phonologic structures

FIG. 2. Composite activation maps in DYS and NI readers for the C and NWR judgment tasks. As shown, DYS and NI readers differ in thedegree of activation produced in different brain regions during phonologic (NWR) compared with orthographic (C) coding; DYS readersdemonstrate a pattern of relative overactivation anteriorly in IFG in contrast to relative underactivation posteriorly, in STG and the angular gyrus.Composite maps (with z-axis Talairach position) are shown for the left anterior region (IFG, z ! 33) and two regions in the left posterior system[post STG (STG, z ! 12) and the angular gyrus (ANG, z ! 23)]. Composite maps are based on brain activations representing C and NWR. Themedian t value was obtained for each pixel in each of the Talairach-transformed images of the 29 DYS and 32 NI readers, respectively. Those tvalues greater than 0.2 were cluster-filtered (cluster size ! 3) and overlaid on composite anatomic images that were obtained by addingTalairach-transformed anatomical images from the two groups. The cluster criterion used in this composite differs from that used in the statisticalanalysis; when combining multiple activation maps from different subjects, it is necessary to change the threshold and cluster criterion to compensatefor imprecise overlap of activation regions between subjects.

FIG. 3. Relative increase in activation during phonologic compared with orthographic coding in different brain regions in NI and DYS readers.As shown in the key, the shadings represent the relative magnitude of the increase in activation (mean pixel counts) for a given ROI calculatedas: (NWR " C!C) ! R. In posterior regions [e.g., posterior BA 22 (STG) and BA 39 (angular gyrus)], the relative change in activation is large(#2, shown in black) in NI readers but very small in DYS readers ($0.5, shown as lightest gray). A contrasting pattern is shown in anterior regions,for example, in BA 44 and 45 (IFG), where NI readers demonstrate an increase in activation (0.5–1) and DYS readers demonstrate an even greaterincrease (#2). There are regions where NI and DYS readers show similar increases in activation, for example, BA 6 and anterior STG (BA 41,BA 42, anterior BA 22). Brain regions shown in white were not part of the 17 ROIs examined; numbers represent BAs.

Neurobiology: Shaywitz et al. Proc. Natl. Acad. Sci. USA 95 (1998) 2639

Proc. Natl. Acad. Sci. USAVol. 95, pp. 2636–2641, March 1998Neurobiology

Functional disruption in the organization of the brain for readingin dyslexiaSALLY E. SHAYWITZ*†, BENNETT A. SHAYWITZ*‡, KENNETH R. PUGH*§, ROBERT K. FULBRIGHT¶,R. TODD CONSTABLE¶, W. EINAR MENCL*§, DONALD P. SHANKWEILER§, ALVIN M. LIBERMAN§,PAWEL SKUDLARSKI¶, JACK M. FLETCHER!, LEONARD KATZ§, KAREN E. MARCHIONE*, CHERYL LACADIE¶,CHRISTOPHER GATENBY¶, AND JOHN C. GORE¶***Department of Pediatrics, ‡Department of Neurology, §Haskins Laboratories, ¶Department of Diagnostic Radiology, **Department of Applied Physics,Yale University School of Medicine, New Haven, CT 06520; and !Department of Pediatrics, University of Texas Medical School, Houston, TX 77030

Contributed by Alvin M. Liberman, January 9, 1998

ABSTRACT Learning to read requires an awareness thatspoken words can be decomposed into the phonologic con-stituents that the alphabetic characters represent. Such pho-nologic awareness is characteristically lacking in dyslexicreaders who, therefore, have difficulty mapping the alphabeticcharacters onto the spoken word. To find the location andextent of the functional disruption in neural systems thatunderlies this impairment, we used functional magnetic res-onance imaging to compare brain activation patterns indyslexic and nonimpaired subjects as they performed tasksthat made progressively greater demands on phonologic anal-ysis. Brain activation patterns differed significantly betweenthe groups with dyslexic readers showing relative underacti-vation in posterior regions (Wernicke’s area, the angulargyrus, and striate cortex) and relative overactivation in ananterior region (inferior frontal gyrus). These results supporta conclusion that the impairment in dyslexia is phonologic innature and that these brain activation patterns may providea neural signature for this impairment.

Speech enables its users to create an indefinitely large numberof words by combining and permuting a small number ofphonologic segments, the consonants and vowels that serve asthe natural constituents of the biologic specialization forlanguage. An alphabetic transcription brings this same abilityto readers but only as they connect its arbitrary characters(letters) to the phonologic segments they represent. Makingthat connection requires an awareness that all words, in fact,can be decomposed into phonologic segments. Thus, it is thisawareness that allows the reader to connect the letter strings(the orthography) to the corresponding units of speech (pho-nologic constituents) that they represent. As numerous studieshave shown, however, such awareness is largely missing indyslexic children and adults (1–4). Not surprisingly, then,perhaps the most sensitive measure of the reading problem indyslexia is inability to read phonologically legal nonsensewords (5–7). As for why dyslexic readers should have excep-tional difficulty developing phonologic awareness, there issupport for the notion that the difficulty resides in the pho-nologic component of the larger specialization for language(8–10). If that component is imperfect, its representations willbe less than ideally distinct and, therefore, harder to bring toconscious awareness.

Previous efforts using functional imaging methods to exam-ine brain organization in dyslexia have been inconclusive(11–17) largely, we think, because the experimental taskstapped the several aspects of the reading process in somewhat

unsystematic ways. Our aim therefore was to develop a set ofhierarchically structured tasks that control the kind of lan-guage-relevant coding required, including especially the de-mand on phonologic analysis, and then to compare the per-formance and brain activation patterns (as measured by func-tional MRI) of dyslexic (DYS) and nonimpaired (NI) readers.Thus, proceeding from the base of the hierarchy to the top, thetasks made demands on visual–spatial processing, ortho-graphic processing, simple phonologic analysis, complex pho-nologic analysis, and lexical–semantic judgment. We hypoth-esized that differences in brain activation patterns wouldemerge as DYS and NI readers were asked to perform tasksthat make progressively greater demands on phonologic anal-ysis.

METHODSTasks. Both the decision and response components of the

tasks were comparable; in each instance the subject viewed twosimultaneously presented stimulus displays, one above theother, and was asked to make a same"different judgment bypressing a response button if the displays matched on a givencognitive dimension, such as line orientation judgment, lettercase judgment, single-letter rhyme, nonword rhyme, and cat-egory judgment. As noted above, the five tasks were orderedhierarchically. (i) At the lowest level, the line orientation (L)judgment task (e.g., Do [\\\/] and [\\\/] match?) taps visual–spatial processing but makes no orthographic demands. (ii)Next, the letter-case (C) judgment task (e.g., Do [bbBb] and[bbBb] match in the pattern of upper- and lowercase letters?)adds an orthographic processing demand but makes no pho-nologic demands, because the stimulus items that consistentirely of consonant strings are, therefore, phonotacticallyimpermissible. (iii) The third task, single letter rhyme (SLR;e.g., Do the letters [T] and [V] rhyme?), although orthograph-ically more simple than C, adds a phonologic processingdemand, requiring the transcoding of the letters (orthography)into phonologic structures and then requiring a phonologicanalysis of those structures sufficient to determine that they door do not rhyme. (iv) The fourth task, nonword rhyme (NWR;e.g., Do [leat] and [jete] rhyme?), requires analysis of morecomplex structures. (v) The final task, semantic category (SC)judgment (e.g., Are [corn] and [rice] in the same category?),also makes substantial demands on transcoding from print tophonology (18, 19) but requires in addition that the printed

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked ‘‘advertisement’’ inaccordance with 18 U.S.C. §1734 solely to indicate this fact.

© 1998 by The National Academy of Sciences 0027-8424"98"952636-6$2.00"0PNAS is available online at http:""www.pnas.org.

Abbreviations: fMRI, functional MRI; DYS, dyslexic; NI, nonim-paired; L, line; C, case; SLR, single letter rhyme; NWR, nonwordrhyme; SC, semantic category; ROI, regions of interest; BA, Brod-mann’s area; IFG, inferior frontal gyrus; STG, superior temporalgyrus; ILES, inferior lateral extrastriate.†To whom reprint requests should be addressed at: Department ofPediatrics, Yale University School of Medicine, P.O. Box 3333, NewHaven, CT 06520-8064. e-mail: Sally.Shaywitz@Yale.edu.

2636

(Figure 2), compared with their preintervention images, EIsubjects were activating bilateral inferior frontal gyri, left superiortemporal sulcus, the occipital temporal region involving theposterior aspects of the middle and inferior temporal gyri and theanterior aspect of the middle occipital gyrus, the inferior occipitalgyrus, and the lingual gyrus.

The thresholded voxel-based activation maps in Figures 1 and2 are presented for the purpose of display of the spatialdistribution of these differences. The voxelwise threshold (p !.05) on these maps takes into account all of the processing stepsinvolved in terms of smoothing, registration, interpolation, andcluster filtering. Applying these processing steps to simulatedwhite noise distributions produced no activations through 10,000iterations. This suggests that the results shown are unaffected bythe multiple comparison problem. To validate the significance ofthe effect, we defined regions of interest anatomically, focusingon regions (Talairach coordinates for centers of mass) in theinferior frontal gyrus (49, 11, 12), parietotemporal region (51,"28, 12), and occipitotemporal region (52, "45, "4), whichprevious studies indicate are critical for reading (Shaywitz et al2002). Within these regions, the sum of the percent signal changein the activated voxels was used as a measure of activation. Foreach of the three anatomically defined regions, the activationwas found to be significantly (p ! .01) greater in year 2compared with year 1, and significantly greater (p ! .001)between year 3 and year 1. For these three regions of interest,group differences between year 2 and year 1 were significant inthe inferior frontal gyrus region for CC versus CI (p # .007) andEI versus CI (p # .04), and in the occipitotemporal region for EIversus CI (p # .02). This corroborates the significant differencesobserved between the groups in the voxel-based maps shown inthe figures.

Discussion

These findings indicate that the nature of the remedial edu-cational intervention is critical to successful outcomes in childrenwith reading disabilities and that the use of an evidence-basedphonologic reading intervention facilitates the development ofthose fast-paced neural systems that underlie skilled reading. Inthis study, a phonologically based reading intervention leads tothe development of neural systems both in anterior (inferiorfrontal gyrus) and posterior (middle temporal gyrus) readingsystems. Converging evidence from a number of lines of inves-tigation indicates that a portion of the posterior reading systems,the occipitotemporal area, is critical for the development ofskilled reading and functions as an automatic, instant wordrecognition system, the visual word form area (Cohen et al 2000,2002; Dehaene et al 2002). In this region, brain activationincreases as reading skill increases (Shaywitz et al 2002); thisregion responds preferentially to rapidly presented stimuli (Priceet al 1996), responds within 150 msec after presentation of astimulus (Salmelin et al 1996), and is engaged even when theword has not been consciously perceived (Dehaene et al 2001).It is this occipitotemporal region that continued to develop 1 yearafter the intervention had ended (Figure 2). In addition to thechanges in posterior brain regions, both the CC and EI groupsshowed changes in anterior activation. Such findings are conso-nant with reports that anterior brain systems, especially involvingregions around the inferior frontal gyrus, have been implicated inreading both in individuals with brain lesions (Benson 1977) aswell as in functional brain imaging studies (Brunswick et al 1999;Corina et al 2001; Georgiewa et al 1999; Gross-Glenn et al 1991;Paulesu et al 1996; Rumsey et al 1997; Shaywitz et al 1998). Inaddition, as shown in Figure 2, in the EI group, two regions were

Figure 1. Composite contrast maps demonstrating the interaction of studygroup and intervention on brain activation patterns. Red-yellow indicatesthe differences in brain activation between year 1 and year 2 that were moreactive (p # .05) in the first group compared with the second; blue-purpleindicates the differences in brain activation between year 1 and year 2 thatwere more active (p # .05) in the second group compared with the first. Forexample, the left column (community control group [CC] vs. experimentalintervention group [EI]) indicates how the brain activation differences inyear 1 and 2 in the CC group compared with the brain activation differencesin year 1 and year 2 in the EI group. The slice locations are 12 and "4 inTalairach space. The legend for brain activation (Talairach x, y, z, coordinatesin parentheses) is as follows: 1, inferior frontal gyrus (41, 23, 12); 2, caudatenucleus ("7, 10, "4); and 3, posterior aspect of the middle temporal gyrus(58, "38, "4). CI, community intervention group.

Figure 2. Composite maps indicating the difference in activation betweenyear 3 and year 1 in the EI study group (n # 25). Red-yellow indicates brainregions that were more active (p # .05) in the third year; blue-purple indi-cates brain regions that were more active (p # .05) in the first year. The slicelocations are 12 and "4 in Talairach space. Brain regions (Talairach x, y, z,coordinates in parentheses) more active in the third year compared with thefirst were as follows: 1, bilateral inferior frontal gyri ($41, 23, 12); 2, the leftsuperior temporal sulcus (51, "42, 12); 3, the occipital temporal regioninvolving the posterior aspects of the middle and inferior temporal gyri andthe anterior aspect of the middle occipital gyrus (42, "49, "4); 4, the inferioroccipital gyrus (34, "71, "4); and 5, the lingual gyrus (13, "88, "4). Thebrain regions more active in the first year compared with the third year were6, the right middle temporal gyrus ("35, "69, 12); and 7, the caudatenucleus ("7, 10, "4).

930 BIOL PSYCHIATRY 2004;55:926–933 B.A. Shaywitz et al

www.elsevier.com/locate/biopsych

Development of Left Occipitotemporal Systems forSkilled Reading in Children After a Phonologically-Based InterventionBennett A. Shaywitz, Sally E. Shaywitz, Benita A. Blachman, Kenneth R. Pugh, Robert K. Fulbright,Pawel Skudlarski, W. Einar Mencl, R. Todd Constable, John M. Holahan, Karen E. Marchione,Jack M. Fletcher, G. Reid Lyon, and John C. GoreBackground: A range of neurobiological investigations shows a failure of left hemisphere posterior brain systems to function properlyduring reading in children and adults with reading disabilities. Such evidence of a disruption in the normal reading pathwaysprovides a neurobiological target for reading interventions. In this study, we hypothesized that the provision of an evidence-based,phonologically mediated reading intervention would improve reading fluency and the development of the fast-paced occipitotemporalsystems serving skilled reading.Methods: Functional magnetic resonance imaging was used to study the effects of a phonologically based reading intervention onbrain organization and reading fluency in 77 children aged 6.1–9.4 years (49 with reading disability and 28 control subjects).Children comprised three experimental groups: experimental intervention (n ! 37), community intervention (n ! 12), andcommunity control subjects (n ! 28).Results: Immediately after the year-long intervention, children taught with the experimental intervention had made significant gainsin reading fluency and demonstrated increased activation in left hemisphere regions, including the inferior frontal gyrus and themiddle temporal gyrus; 1 year after the experimental intervention had ended these children were activating bilateral inferior frontalgyri and left superior temporal and occipitotemporal regions.Conclusions: These data indicate that the nature of the remedial educational intervention is critical to successful outcomes inchildren with reading disabilities and that the use of an evidence-based phonologic reading intervention facilitates the developmentof those fast-paced neural systems that underlie skilled reading.

Key Words: Reading, dyslexia, fluency, functional magnetic reso-nance imaging, plasticity, intervention

Reading disability (developmental dyslexia) is one of themost common neurobehavioral problems affecting chil-dren and adults. There is now a strong consensus that the

central difficulty in reading disability reflects a deficit within thelanguage system, and more particularly, in a lower-level compo-nent, phonology, which has to do with the ability to access theunderlying sound structure of words (Shaywitz 1998; Wagnerand Torgesen 1987). In young school-age children, a deficit inphonology represents the most reliable and specific correlate ofreading disability (Fletcher et al 1994; Morris et al 1998). Suchfindings form the basis for the most successful and evidence-based interventions designed to improve reading (Report of the

National Reading Panel 2000). According to recent findings(Torgesen et al 1999), provision of an evidence-based interven-tion at an early stage of reading instruction leads to the devel-opment of fluent reading (“the ability to read a text quickly,accurately and with proper expression” (Report of the NationalReading Panel 2000), the hallmark of skilled reading.

A range of neurobiological investigations, using postmortembrain specimens (Galaburda et al 1985), brain morphometry(Filipek 1996), diffusion tensor magnetic resonance imaging(MRI) (Klingberg et al 2000), and functional brain imaging inreading-disabled readers (Brunswick et al 1999; Helenius et al1999; Horwitz et al 1998; Paulesu et al 2001; Rumsey et al 1992,1997; Salmelin et al 1996; Shaywitz et al 1998) shows a failure ofleft hemisphere posterior brain systems to function properlyduring reading. This neurobiological evidence of dysfunction inleft hemisphere posterior reading circuits is already present inreading-disabled children and cannot be ascribed simply to alifetime of poor reading (Seki et al 2001; Shaywitz et al 2002;Simos et al 2000; Temple et al 2000).

Such evidence of a disruption in the normal reading pathwaysprovides a neurobiological target for reading interventions. Inthis study, we hypothesized that the provision of an evidence-based, phonologically mediated reading intervention would im-prove reading fluency and the development of the fast-pacedoccipitotemporal systems serving skilled reading. We chose aletter identification task, because we wanted a task that was easyenough for very young disabled readers to perform with highaccuracy—we did not want the issue of effort to be a factor ininterpreting the results. A number of investigations (reviewed inScarborough 1998) indicate that in young children, letter identi-fication is the strongest predictor of reading ability. The experi-mental intervention was adapted from a model used previouslyin a prevention study with first-grade children (Blachman et al

From the Departments of Pediatrics (BAS, SES, KRP, WEM, JMH, KEM), Neu-rology (BAS), and Diagnostic Radiology (RKF, PS, RTC, JCG), Yale Univer-sity School of Medicine; Department of Applied Physics (JCG), Yale Uni-versity; and Haskins Laboratories (KRP, WEM), New Haven, Connecticut;Department of Psychology (BAB) and School of Education (BAB), Syra-cuse University, Syracuse, New York; Department of Pediatrics (JMF),University of Texas-Houston Health Science Center, Houston, Texas;Child Development and Behavior Branch (GRL), National Institute ofChild Health and Human Development, National Institutes of Health,Bethesda, Maryland; and the Institute of Imaging Science (JCG) andDepartment of Radiology and Radiological Sciences (JCG), VanderbiltUniversity, Nashville, Tennessee.

Address reprint requests to Dr. Bennett A. Shaywitz, Yale University Schoolof Medicine, Department of Pediatrics, P.O. Box 3333, New Haven, CT06510-8064.

Received September 18, 2003; revised December 17, 2003; accepted December19, 2003.

BIOL PSYCHIATRY 2004;55:926–9330006-3223/04/$30.00doi:10.1016/j.biopsych.2003.12.019 © 2004 Society of Biological Psychiatry

Development of Left Occipitotemporal Systems forSkilled Reading in Children After a Phonologically-Based InterventionBennett A. Shaywitz, Sally E. Shaywitz, Benita A. Blachman, Kenneth R. Pugh, Robert K. Fulbright,Pawel Skudlarski, W. Einar Mencl, R. Todd Constable, John M. Holahan, Karen E. Marchione,Jack M. Fletcher, G. Reid Lyon, and John C. GoreBackground: A range of neurobiological investigations shows a failure of left hemisphere posterior brain systems to function properlyduring reading in children and adults with reading disabilities. Such evidence of a disruption in the normal reading pathwaysprovides a neurobiological target for reading interventions. In this study, we hypothesized that the provision of an evidence-based,phonologically mediated reading intervention would improve reading fluency and the development of the fast-paced occipitotemporalsystems serving skilled reading.Methods: Functional magnetic resonance imaging was used to study the effects of a phonologically based reading intervention onbrain organization and reading fluency in 77 children aged 6.1–9.4 years (49 with reading disability and 28 control subjects).Children comprised three experimental groups: experimental intervention (n ! 37), community intervention (n ! 12), andcommunity control subjects (n ! 28).Results: Immediately after the year-long intervention, children taught with the experimental intervention had made significant gainsin reading fluency and demonstrated increased activation in left hemisphere regions, including the inferior frontal gyrus and themiddle temporal gyrus; 1 year after the experimental intervention had ended these children were activating bilateral inferior frontalgyri and left superior temporal and occipitotemporal regions.Conclusions: These data indicate that the nature of the remedial educational intervention is critical to successful outcomes inchildren with reading disabilities and that the use of an evidence-based phonologic reading intervention facilitates the developmentof those fast-paced neural systems that underlie skilled reading.

Key Words: Reading, dyslexia, fluency, functional magnetic reso-nance imaging, plasticity, intervention

Reading disability (developmental dyslexia) is one of themost common neurobehavioral problems affecting chil-dren and adults. There is now a strong consensus that the

central difficulty in reading disability reflects a deficit within thelanguage system, and more particularly, in a lower-level compo-nent, phonology, which has to do with the ability to access theunderlying sound structure of words (Shaywitz 1998; Wagnerand Torgesen 1987). In young school-age children, a deficit inphonology represents the most reliable and specific correlate ofreading disability (Fletcher et al 1994; Morris et al 1998). Suchfindings form the basis for the most successful and evidence-based interventions designed to improve reading (Report of the

National Reading Panel 2000). According to recent findings(Torgesen et al 1999), provision of an evidence-based interven-tion at an early stage of reading instruction leads to the devel-opment of fluent reading (“the ability to read a text quickly,accurately and with proper expression” (Report of the NationalReading Panel 2000), the hallmark of skilled reading.

A range of neurobiological investigations, using postmortembrain specimens (Galaburda et al 1985), brain morphometry(Filipek 1996), diffusion tensor magnetic resonance imaging(MRI) (Klingberg et al 2000), and functional brain imaging inreading-disabled readers (Brunswick et al 1999; Helenius et al1999; Horwitz et al 1998; Paulesu et al 2001; Rumsey et al 1992,1997; Salmelin et al 1996; Shaywitz et al 1998) shows a failure ofleft hemisphere posterior brain systems to function properlyduring reading. This neurobiological evidence of dysfunction inleft hemisphere posterior reading circuits is already present inreading-disabled children and cannot be ascribed simply to alifetime of poor reading (Seki et al 2001; Shaywitz et al 2002;Simos et al 2000; Temple et al 2000).

Such evidence of a disruption in the normal reading pathwaysprovides a neurobiological target for reading interventions. Inthis study, we hypothesized that the provision of an evidence-based, phonologically mediated reading intervention would im-prove reading fluency and the development of the fast-pacedoccipitotemporal systems serving skilled reading. We chose aletter identification task, because we wanted a task that was easyenough for very young disabled readers to perform with highaccuracy—we did not want the issue of effort to be a factor ininterpreting the results. A number of investigations (reviewed inScarborough 1998) indicate that in young children, letter identi-fication is the strongest predictor of reading ability. The experi-mental intervention was adapted from a model used previouslyin a prevention study with first-grade children (Blachman et al

From the Departments of Pediatrics (BAS, SES, KRP, WEM, JMH, KEM), Neu-rology (BAS), and Diagnostic Radiology (RKF, PS, RTC, JCG), Yale Univer-sity School of Medicine; Department of Applied Physics (JCG), Yale Uni-versity; and Haskins Laboratories (KRP, WEM), New Haven, Connecticut;Department of Psychology (BAB) and School of Education (BAB), Syra-cuse University, Syracuse, New York; Department of Pediatrics (JMF),University of Texas-Houston Health Science Center, Houston, Texas;Child Development and Behavior Branch (GRL), National Institute ofChild Health and Human Development, National Institutes of Health,Bethesda, Maryland; and the Institute of Imaging Science (JCG) andDepartment of Radiology and Radiological Sciences (JCG), VanderbiltUniversity, Nashville, Tennessee.

Address reprint requests to Dr. Bennett A. Shaywitz, Yale University Schoolof Medicine, Department of Pediatrics, P.O. Box 3333, New Haven, CT06510-8064.

Received September 18, 2003; revised December 17, 2003; accepted December19, 2003.

BIOL PSYCHIATRY 2004;55:926–9330006-3223/04/$30.00doi:10.1016/j.biopsych.2003.12.019 © 2004 Society of Biological Psychiatry

Mathematics

• Achievement in mathematics can be studied similarly by addressing core conceptual deficits and poor strategy choices

Basic Mathematical Concepts

Basic Mathematical Concepts

Training Basic Concepts

Strategy-Choice in Mathematics

• Children use different arithmetic strategies, just as they use different word-identification strategies.

• Children use strategies of counting, retrieval, decomposition (dividing a problem into easier problems) to solve math problems.

Individual Differences in Learning Arithmetic

• There are three different types of arithmetic learners, based on individual rate of learning and cognitive style:

• Good students: Answer quickly and accurately retrieve answers.

• Not-so-good students: Answer more slowly and less accurately.

• Perfectionists: Answer quickly and accurately, but only use retrieval when they are sure of the answer. If they are not 100% sure, they use strategies to check their answers.

Cultural Context of Arithmetic Performance

• Some national educational systems are better than others in teaching math.

• Teachers in countries with higher math achievement (Japan, Hong Kong, Hungary, the Netherlands)

• spend more time on math overall than do American teachers,

• spend more time on math concepts than on memorization of procedures.

Exam

Review Questions.

1. In Piaget’s theory, what are the sources of continuity and discontinuity in cognitive development? Provide an example of each. For example, what would be an example of equilibration?

2. How does the object concept develop? According to Piaget, what does this development

indicate about infants’ thinking? What evidence indicates that Piaget’s interpretation of the A-not-B error is mistaken?

3. How did Piaget interpret failure on the conservation tasks? What did he think failure indicated

about children’s thinking? What did he think success indicated about children’s thinking?

4. Why did Piaget give children the pendulum problem?

5. In what ways do the sociocultural, information-processing, and core-knowledge approaches address weaknesses in Piaget’s theory?

6. According to the information processing approach, children improve their ability to solve

problems as a result of improved planning and analogical reasoning. Why do children fail to plan in situations where it would be adaptive? What obstacles do children face in analogical reasoning?

7. What evidence indicates that children’s basic cognitive processes improve with age?

8. What evidence supports the core-knowledge approach to infant cognition?

9. How do the Piagetian, information-processing, and core-knowledge theories offer unique

contributions to the improvement of educating children? Provide an example of each.

10. What evidence indicates that humans possess a language acquisition device?

11. What evidence suggests that children have certain assumptions about language before they learn the meanings of words?

12. What evidence suggests that children use grammatical structure before they actually speak in

fully grammatical sentences? Why is analogical reasoning unlikely to be helpful in acquiring grammar?

13. What environmental support is there for the acquisition of language? Does the evidence suggest

that all, some, or none of this support is necessary for the acquisition of grammar? Justify your answer with experimental findings.

14. How do humans use symbolic reference in ways that other species do not?

15. What does gesture tell us about language?

16. What is cue validity, and how does this concept help to explain perceptual categorization, the

order in which children acquire concepts, different levels of object hierarchies, and why some members of a category are more prototypical than others?

17. How did Krascum & Andrew’s (1998) study about the learning of the categories wugs and gillies

indicate the importance of knowing causal relations? What evidence indicates that children go beyond perceptual similarities in their concepts?

Review Questions.

1. In Piaget’s theory, what are the sources of continuity and discontinuity in cognitive development? Provide an example of each. For example, what would be an example of equilibration?

2. How does the object concept develop? According to Piaget, what does this development

indicate about infants’ thinking? What evidence indicates that Piaget’s interpretation of the A-not-B error is mistaken?

3. How did Piaget interpret failure on the conservation tasks? What did he think failure indicated

about children’s thinking? What did he think success indicated about children’s thinking?

4. Why did Piaget give children the pendulum problem?

5. In what ways do the sociocultural, information-processing, and core-knowledge approaches address weaknesses in Piaget’s theory?

6. According to the information processing approach, children improve their ability to solve

problems as a result of improved planning and analogical reasoning. Why do children fail to plan in situations where it would be adaptive? What obstacles do children face in analogical reasoning?

7. What evidence indicates that children’s basic cognitive processes improve with age?

8. What evidence supports the core-knowledge approach to infant cognition?

9. How do the Piagetian, information-processing, and core-knowledge theories offer unique

contributions to the improvement of educating children? Provide an example of each.

10. What evidence indicates that humans possess a language acquisition device?

11. What evidence suggests that children have certain assumptions about language before they learn the meanings of words?

12. What evidence suggests that children use grammatical structure before they actually speak in

fully grammatical sentences? Why is analogical reasoning unlikely to be helpful in acquiring grammar?

13. What environmental support is there for the acquisition of language? Does the evidence suggest

that all, some, or none of this support is necessary for the acquisition of grammar? Justify your answer with experimental findings.

14. How do humans use symbolic reference in ways that other species do not?

15. What does gesture tell us about language?

16. What is cue validity, and how does this concept help to explain perceptual categorization, the

order in which children acquire concepts, different levels of object hierarchies, and why some members of a category are more prototypical than others?

17. How did Krascum & Andrew’s (1998) study about the learning of the categories wugs and gillies

indicate the importance of knowing causal relations? What evidence indicates that children go beyond perceptual similarities in their concepts?