attention control_blair razza.pdf

8/13/2019 ATTENTION CONTROL_Blair Razza.pdf

1/17

Relating Effortful Control, Executive Function, and False Belief Understanding

to Emerging Math and Literacy Ability in Kindergarten

Clancy Blair and Rachel Peters Razza

The Pennsylvania State University

This study examined the role of self-regulation in emerging academic ability in one hundred and forty-one 3- to5-year-old children from low-income homes. Measures of effortful control, false belief understanding, and theinhibitory control and attention-shifting aspects of executive function in preschool were related to measures ofmath and literacy ability in kindergarten. Results indicated that the various aspects of child self-regulationaccounted for unique variance in the academic outcomes independent of general intelligence and that the in-hibitory control aspect of executive function was a prominent correlate of both early math and reading ability.Findings suggest that curricula designed to improve self-regulation skills as well as enhance early academicabilities may be most effective in helping children succeed in school.

The emergence of self-regulation provides the basisfor a wide array of developing competencies in earlychildhood ranging from theory of mind (Carlson,Mandell, & Williams, 2004) to child compliance(Kopp, 1989; Stifter, Spinrad, & Braungart-Rieker,1999). Increasingly, accurate description and mea-surement of self-regulation is seen as central to un-derstanding the processes through which childrenadapt to and learn in formal school settings (Blair,2002; Howse, Calkins, Anastopoulos, Keane, &Shelton, 2003). To date, however, only limited workhas examined the relation of various aspects of self-regulation to early school success, the topic of this

article. This is particularly true for children at in-creased risk for early school failure such as thosefrom low-income homes.

Temperament-Based and Neural Systems Approaches toSelf-Regulation

One way to characterize the emergence of self-regulation in children is to consider it as the devel-opmental integration of emotion and cognition inearly childhood. At a behavioral level, this integra-tion is represented in a temperament-based

approach to self-regulation that defines temperamentas emotional reactivity and the regulation of this re-activity (e.g., Posner & Rothbart, 2000; Rothbart &Ahadi, 1994). At a neural systems level, this integra-tion is represented in an approach that examines therole of interconnected brain structures of the pre-frontal and anterior cingulate cortices in emotionalreactivity, emotion regulation, attention, and cogni-tive control (Bush, Luu, & Posner, 2000; Drevets &Raichle, 1998; Groenewegen & Uylings, 2000).

A prominent aspect of the temperament-basedapproach to the study of self-regulation in children isthe construct known as effortful control (Rothbart &

Ahadi, 1994). Effortful control refers to the ability toinhibit a dominant or prepotent response in favor ofa subdominant or less salient response. As such, ef-fortful control allows for the regulation of approachand withdrawal behavioral tendencies in the face ofimmediate cues for reward or punishment. In researchon the influence of temperament on child develop-ment, researchers have been interested in the relationof effortful control to emotion and emotion-relatedregulation of behavior (Raver, 2004), to developingsocial competence and behavior problems (Eisenberget al., 2003; Valiente et al., 2003) and to the parent

child relationship and compliance with caregiversrequests for behavior regulation (Kochanska, Mur-ray, Jacques, Koenig, & Vandegeest, 1996).

A prominent aspect of the neural systems ap-proach to the study of self-regulation has been a fo-cus on cognitive processes referred to collectively asexecutive function. Executive function refers toattention shifting, working memory, and inhibitory

r 2007 by the Society for Research in Child Development, Inc.All rights reserved. 0009-3920/2007/7802-0018

The work reported here was supported, in part, by NationalInstitute of Child Health and Human Development Grants R03HD39750, P01 HD39667, and R01 HD51502. Thanks are due toCathy Jantzer, Rebecca Holland, Tanja Rothrauff, and KatherineWagner for their assistance with data collection. Rachel PetersRazza is now at the National Center for Children and Families,Teachers College, Columbia University, New York, NY.

Correspondence concerning this article should be addressed toClancy Blair, Human Development and Family Studies, Pennsyl-vania State University, 110 Henderson South, University Park, PA16802-6504. Electronic mail may be sent to [email protected].

Child Development, March/April 2007, Volume 78, Number 2, Pages 647 663


2/17

control cognitive processes that are utilized in plan-ning, problem solving, and goal-directed activity(Miyake, Friedman, Emerson, Witzki, & Howerter,2000). Executive function is similar to effortful controlin that it also refers in part to the ability to inhibit aprepotent or dominant response in favor of a lesssalient response. Executive function, however, focusesprimarily on volitional control of cognitive self-regu-latory processes, whereas effortful control includes tosome extent, although not exclusively by any means,a focus on automatic or nonconscious aspects ofemotional reactivity and regulation. That is, effortfulcontrol considers the appetitive or aversive nature ofthe conditions under which control is required whilethe primary focus of work on executive functionconcerns the deployment of cognitive processes un-der conditions that are essentially affectively neutral.

Recent work suggests that certain of the cognitive

processes related to executive function may beinfluenced by the conscious or nonconscious repre-sentation of the motivational or affective significanceof stimuli (Bechara, 2004; Gray, Braver, & Raichle,2002). Data indicating a role for emotional arousal inexecutive function raise the possibility of a distinc-tion between hot or affectively sensitive and cool oraffectively neutral aspects of executive function(Kerr & Zelazo, 2004; Zelazo & Muller, 2002). Fur-thermore, a role for emotional arousal and reactivityin executive function is indicated by work in bothchildren and adults as well as nonhuman animal

models demonstrating that level of stress arousal asindicated by physiological markers of hypothalam-ic pituitary adrenal axis function is associatedwith executive function ability (Arnsten & Shansky,2004; Blair, Granger, & Razza, 2005; Liu, Diorio,Francis, & Meaney, 2000; Lupien, Gillin, & Hauger,1999). Despite this potential overlap, however, workin preschool children from middle-income homessuggests that executive function and effortful controlare only moderately correlated (Carlson & Moses,2001; Davis, Bruce, & Gunnar, 2002; Gerardi-Caul-ton, 2000; Rothbart, Ellis, Rueda, & Posner, 2003).Therefore, one aim of this study was to examine the

relations among measures of effortful control andexecutive function in young children attending HeadStart, the U.S. federal preschool educational programfor children in poverty, and to determine the extentto which each is uniquely related to emerging mathand literacy ability measured in kindergarten.

False Belief Understanding

An additional aspect of development in earlychildhood shown in various ways to be highly rele-

vant to both neural systems and temperament-basedapproaches to self-regulation in young children isfalse belief understanding. False belief understand-ing concerns the emergence in the preschool periodof the awareness that one may hold and act on beliefsthat are false. As such, it is an aspect of a larger,comprehensive construct referred to as theory ofmind. False belief understanding is moderately tohighly correlated with the inhibitory control aspectof executive function (Carlson & Moses, 2001) butlargely unrelated to working memory and attention-shifting aspects of executive function when adjustingfor inhibitory control (Carlson, Moses, & Claxton,2004; Hughes, 1998). False belief understanding isalso associated with higher levels of behavior regu-lation in preschoolers (Hughes, Dunn, & White,1998) and theory of mind is generally understood tobe a central aspect of developing social-emotional

self-regulation (Bachevalier & Loveland, 2006). Again,given that limited work has examined false beliefunderstanding and its correlates in preschool childrenfrom low-income homes (Hughes & Ensor, 2005;Peters & Blair, 2003; Razza, 2005), a second aim ofthis study was to examine the relation of false beliefunderstanding to effortful control and executivefunction. Examination of false belief understandingin relation to other aspects of self-regulation in chil-dren facing socioeconomic disadvantage can help todetermine the extent to which it may be an inde-pendent contributor to developing competence in

school settings.

Relation of Effortful Control, Executive Function, andFalse Belief Understanding to School Readiness

Although effortful control, false belief under-standing, and executive function overlap in severalrespects, there are multiple routes through whicheach might affect early academic ability in school.The characteristics that comprise effortful control arethought to be important for motivation and en-gagement in school settings (Rothbart & Jones, 1998)and prior studies indicate that temperamental char-

acteristics such as persistence, activity level, andabsence of distractibility as reported by teachers arerelated to academic success in elementary school(Martin, Drew, Gaddis, & Moseley, 1988; Normand-eau & Guay, 1998). Similarly, the emergence of falsebelief understanding is accompanied by advances ina number of domains, including language develop-ment, that are important for interpersonal function-ing and thereby to the relationships with teachersand peers that are known to be important contrib-utors to early school achievement (Mashburn & Pi-

648 Blair and Razza


3/17

anta, 2006). Also, the cognitive processes that com-prise executive function are likely to play a role inknowledge acquisition associated with emergingmath and reading ability. Prior studies with youngchildren have established relations between execu-tive function and early math ability (Bull & Scerif,2001; Espy et al., 2004) and deficits in executivefunction have been noted in learning disability inboth reading and math (McLean & Hitch, 1999;Swanson, 1999). However, it is important to considerthe extent to which all aspects of self-regulationmight overlap with general intelligence (Blair, 2006)and to control for general intellectual ability whenexamining relations of executive function to aca-demic progress in school.

Furthermore, to the extent that the inhibitorycontrol aspect of executive function is a centralcomponent of both effortful control and false belief

understanding, it can serve as a focus for the studyof self-regulation and school readiness. Through theuse of task-switching measures that require themaintenance in working memory of a rule for correctresponding while inhibiting a prepotent responsetendency that interferes with successful task com-pletion, a growing body of work has come to identifydevelopmental change in inhibitory control ability atapproximately 4 years of age as a prominent aspectof developing executive function (Diamond, 2002;Diamond, Carlson, & Beck, 2005; Kirkham, Cruess, &Diamond, 2003) and perhaps self-regulation more

generally. It is necessary, however, to juxtapose theinhibitory control explanation for the developmentof executive function with work by others suggestingthat the development of the ability to cognitivelyrepresent hierarchical rule structures provides thebasis for advances in executive function and self-regulation in early childhood (Happaney & Zelazo,2003; Zelazo, Muller, Frye, & Marcovitch, 2003).Specifically, the Cognitive Complexity and Controltheoryrevised (CCCr) states that the emergenceof the cognitive ability to represent embedded if-thenrule structures allows for reflection on rule pairs andconsequently, increasing competency on executive

function and self-regulation tasks.

Study Aims

Given indication of overlap among various as-pects of self-regulation, the overarching aim of thisstudy was to examine interrelations among executivefunction, effortful control, and false belief under-standing in children from low-socio economic statusbackgrounds and to determine the unique contribu-tion of each to measures of emerging math and lit-

eracy ability in kindergarten. Given evidence for apredominant role for the inhibitory control aspect ofexecutive function in self-regulation in preschoolchildren (e.g., Carlson & Moses, 2001; Diamond etal., 2005), we were particularly interested in the re-lation of this aspect of cognitive ability to emergingcompetence in math and literacy. As age 4 yearsseems to be a transitional period for the developmentof inhibitory control, we were interested in deter-mining whether it would continue to be an impor-tant contributor to early math and literacy abilityover and above other aspects of child self-regulationwith which it is associated. However, although weexpected based on prior work that executive func-tion, effortful control, and false belief understandingwould be moderately interrelated, we expected thateach would account for unique variance in mathe-matics and literacy ability in kindergarten even

when controlling for general intelligence. This ex-pectation is in contrast to one in which a single as-pect of self-regulation, such as inhibitory control,would be considered as foundational and to underlierelations of other aspects of self-regulation to indi-cators of school readiness. In light of much priorwork demonstrating an independent relation of so-cial development to early school outcomes (Ladd,Herald, & Kochel, 2006), we expected that mea-sures of false belief and effortful control, to the extentthat they indicate aspects of social-emotional devel-opment associated with interpersonal functioning,

would account for unique variance in measuresof early school progress over and above that associ-ated with measures of the inhibitory control andattention-shifting aspects of executive function.

Method

Participants

The sample included 170 children who attendedHead Start programs predominantly serving Whitefamilies living in rural and nonurban locations.

Children were recruited through a letter sent hometo all children enrolled in classrooms of two HeadStart programs. The mean age of children at the timeof testing in preschool was 5 years, 1 month (range 3years 9 months to 5 years 8 months). The mean age atthe time of testing in kindergarten was 6 years 2months (range 5 years 7 months to 6 years 11months). Eighty of the children were female and 90were male. Among participants reporting ethnicity,80% reported White, 12% reported more thanone ethnicity, 6% reported Asian, and 2% reported

Self-Regulation and School Readiness 649


4/17

African American. All children were either fromhouseholds in which family income was below thepoverty line as defined by federal poverty thresh-olds, or were eligible for the Head Start program dueto the fact that in the absence of child care, familyincome would be below the poverty line. Incomerelative to need in the sample (annual income di-vided by the poverty threshold as determined bynumber of persons in the household) had a mean of1.04 and a median of .86, indicating that over half ofthe sample were living below the poverty level. Thisrepresents an average monthly income adjusted forfamily size of $381 per month. Eighty-five percent ofthe sample were below 150% of poverty ($572 permonth) and 95% of the sample were below 200% ofpoverty ($762 per month). Fourteen percent of thesample reported the highest level of education as lessthan high school, 59% reported a high school di-

ploma or equivalent, and 23% reported some edu-cation beyond high school. The majority of children,71%, lived in a home with two adults and 88% of thesample had one or more siblings in the household.

Procedure

During the preschool year, children were seenindividually in two approximately 45-min sessionsbetween 9:00 a.m. and 3:30 p.m. in a quiet testingarea. During one of the sessions, the first session forapproximately two thirds of the children participat-

ing in the study, children were administered ameasure of receptive vocabulary and an attention-shifting measure of executive function. Also duringthis session, heart rate was recorded and salivasamples were collected as part of a larger study. Inthe second session, a peg-tapping measure of ex-ecutive function was administered along with twomeasures of false belief understanding. Parents andteachers reported on child temperament and teach-ers also reported on child behavior in the classroom.Parents and teachers received packets of question-naires with a self-addressed stamped envelope andwere asked to complete the questionnaires and re-

turn them as soon as possible.All children were seen in the spring of the kin-

dergarten year in a single session, usually in a quiettesting room at the Head Start center at which thechild had previously been enrolled. For some chil-dren, this session occurred either in a testing room atthe Penn State University campus or in a quiet test-ing room of a local library. Two measures from thepreschool testing, peg tapping and item selection,were repeated because the construct they assess,executive function, is central to the studys aims and

we were interested in determining the extent ofchange in performance over one years time. In ad-dition, early academic measures were administeredalong with Ravens Colored Progressive Matricestest. Ravens matrices test, a measure of fluid, non-verbal intelligence, was used for the kindergartenassessment, as we had assessed the verbal aspect ofintelligence during the preschool year

Measures

Table 1 provides information on the study designand lists the measures and scoring for each constructat each time point.

Executive function. We assessed executive functionwith a peg-tapping measure of inhibitory control(Diamond & Taylor, 1996) and an item-selectionmeasure of attention shifting (Jacques & Zelazo,

2001). For the peg-tapping measure, children wereinstructed to tap twice with a wooden dowel (15 cmlong 1 cm in diameter) when the experimentertaps once, and once when the experimenter tapstwice. The task requires children to inhibit a naturaltendency to mimic the action of the experimenterwhile remembering the rule for the correct response.After practice trials, the child was administered aseries of 16 trials in a counterbalanced sequence (8one-tap and 8 two-tap trials) (Diamond & Taylor,1996). Diamond, Prevor, Callendar, and Druin (1997)provide information on the neuropsychological basis

for the task and provide clinically relevant evidencefor impaired performance among children with earlyand continuously treated phenylketonuria, a condi-tion associated with impaired dopaminergic functionin the prefrontal cortex. Coefficienta for the 16 itemsin these data was .82 in preschool and .75 in kinder-garten. A proportion scoreFthe number of correctresponses divided by the total number of trialsFwasused as a measure of performance on the task.

For the item-selection task, children were pre-sented with pictures of three items that vary alongsome combination of two of three dimensions, in-cluding size, shape, and color. Following a pretest,

children are presented with 15 trials in which theyare instructed to point to two objects that go togetherin one way. Children are then instructed to point totwo objects that go together in another way. The taskrequires children to identify two of the three objectsthat are similar along one dimension (i.e., shape) butthen to shift cognitive set and identify two of thethree objects that are similar along a second dimen-sion (i.e., size). Jacques and Zelazo (2001) have ex-tensively investigated the task in a cross-sectionalsample of children between the ages of 2 and 5 years

650 Blair and Razza


5/17

and found that the results converge well with thoseof similar dimensional shift measures of executivefunction. Coefficient a for the 15 items in these data

was .73 in preschool and .77 in kindergarten. Again aproportion scoreFthe number of correct responsesdivided by the total number of trialsFwas used as ameasure of performance on the task.

False belief understanding. For the measurement offalse belief understanding, two assessments that arestandard in the literature with preschoolers wereused: an unexpected contents task and a changedlocations task. For the unexpected contents task, anegg carton was placed in front of the child and thechild was asked, What do you think is inside thiscarton? All children correctly answered this controlquestion. The egg carton was then opened and the

child was shown that the carton contained crayonsinstead of eggs. The carton was shut and the childwas asked the test question, What did you thinkwas inside the carton, before we opened it? fol-lowed by the control question, What is in the cartonreally? If the child did not respond, a forced-choiceprompt was provided (e.g., eggs or crayons).Children were successful if they responded correctlyto both the test and control questions (Hughes, 1998;Hughes & Dunn, 1998). This task was scored as pass(1), fail(0).

For the changed locations task, a scenario wasenacted concerning the placement of an object, a barof chocolate, by one character that was then relocated

by another character while the first character was outof sight (Astington & Jenkins, 1995; Holmes, Black, &Miller, 1996; Wimmer & Perner, 1983). Children wereasked to predict where the character that did notwitness the change would look for the chocolateupon his return. Figurines of two familiar characters,Big Bird and Elmo, were used to act out the scenario.In order to pass this task, children had to correctlyanswer a control question, Do you remember wherethe chocolate is now? and a test question, Wherewill Elmo look for the chocolate? Children weresuccessful if they responded correctly to both the testand control questions. This task was also scored as

pass (1), fail (0). Performance across the two taskswas summed to create a single indicator of falsebelief understanding.

Effortful control. Information on child effortfulcontrol was obtained from teachers and parents (themother for over 90% of participants) using a shortform of the Childrens Behavior Questionnaire(Putnam & Rothbart, 2006). Participants respondedregarding how true or not a particular behavior is ofthe child on a 7-point Likert scale ranging fromextremely true to extremely untrue of the child.

Table1

List of Constructs, Measures, and Time Points of Measurement for the Study

Construct Measure Time points

Executive function Peg-tapping measure of inhibitory control

Proportion of correct responses on 16 trials, range 0 1

Preschool and

kindergarten

Item selection measure of attention shifting

Proportion of correct responses on 15 trials, range 0 1

Preschool and

kindergarten

Effortful control Childrens Behavior Questionnaire composite of inhibitory

control, attention, approach, and anger subscales

Factor scores of teacher and parent ratings, mean50, SD51

Preschool only

False belief

understanding

Unexpected contents and changed locations tasks

Sum of test questions, pass51 and fail50, range 0 2

Preschool only

Intelligence Peabody Picture Vocabulary Test 3

Normed standardized score, mean5100, SD515

Preschool only

Ravens Progressive Colored Matrices

Raw score sum of correct responses, range 0 35

Kindergarten only

Academic ability Math battery adapted from the Early Childhood Longitudinal

Study Kindergarten


Kindergarten only

Elision subtest of the Preschool Comprehensive Test of

Phonological and Print Processing


Kindergarten only

Letter knowledge assessment from the Head Start National

Reporting System Direct Child Assessment


Kindergarten only



6/17

The Anger, Approach, Attention, and InhibitoryControl subscales were utilized for this analysis.Ratings for items on each subscale were reversescored if needed and summed. Cronbachs as forthese scales as reported by Putnam and Rothbartrange from .65 to .80. Sample items for the Angersubscale include gets angry when s/he cant findsomething s/he wants to play with, gets angrywhen called in from play before s/he is ready toquit, and has temper tantrums when s/he doesntget what s/he wants. Sample items for the Ap-proach subscale include gets so worked up beforean exciting event that s/he has trouble sitting still,remains pretty calm about upcoming treats (re-verse scored), and becomes very excited before anouting. Sample items for the Attention subscaleinclude when practicing an activity, has a hard timekeeping her/his mind on it (reverse scored), when

drawing or coloring in a book, shows strong con-centration, and sometimes becomes absorbed in apicture book and looks at it for a long time. Sampleitems for the Inhibitory Control subscale includecan wait before entering into new activities if s/heis asked to, is good at following instructions, andcan easily stop an activity when s/he is told no.Two items on the Anger subscale concerned chil-drens behavior when going to bed; therefore, theseitems were deleted from the teacher scale. Missingvalues for items were replaced with the mean of thesubscale to which that item belonged. No item had

greater than 5% missing. Examination of Cronbachsafor the parent and teacher report subscales in thissample indicated that the teacher ratings were moreinternally consistent on all subscales than were par-ent ratings. Internal consistency estimates on theAttention, Approach, Anger, and Inhibitory Controlsubscales for teachers were .91, .74, .86, and .84. In-ternal consistency estimates for parents on thesesame subscales were .71, .57, .51, and .60.

Owing to a high intercorrelation among theteacher report inhibitory control, attention, anger,and approach subscales, we used factor analysis toextract the first principal component, the only com-

ponent with an eigenvalue 41.0, to create a sum-mary indicator of effortful control. This analysisyielded a single factor accounting for 71% of thevariance in the subscales, with observed loadings of.93 and .87 for the inhibitory control and attentionsubscales and .87 and .70 for the anger andapproach subscales. Higher scores on this factor in-dicate a higher level of regulation and a lower levelof anger and approach reactivity. The same analysisfor parent report also yielded a single principalcomponent with an eigenvalue 41.0. However, this

factor accounted for only 44% of the total variancewith loadings of .81 and .77 for the inhibitory controland attention subscales and .61 and .40 for theanger and approach subscales.

In addition to the Anger, Approach, Attention, andInhibitory Control subscales, we collected only oneother temperament subscale: Discomfort. This sub-scale concerns sensitivity to hot or cold and to painand minor cuts and bruises. Upon analysis, this sub-scale was not found to load with the other subscaleson the effortful control factor and was unrelated tothe academic outcome measures. Therefore, we didnot include this subscale in any further analyses.

Academic outcomes. Academic outcomes in kin-dergarten were assessed in three areas: mathematicsknowledge, phonemic awareness, and letter knowl-edge. Mathematics knowledge was assessed with amath battery adapted from the Early Childhood

Longitudinal StudyF

Kindergarten, created for usein the Head Start National Reporting System DirectChild Assessment. The battery includes 24 itemsdesigned to assess basic numeracy, knowledge ofshapes, quantity, relative size, addition, subtraction,and simple graphic relations. Children receive 1point for each item answered correctly. Cronbachs a,calculated for a sample of 14,380 children partici-pating in the Early Childhood Longitudinal StudyFKindergarten, is .92.

To assess reading readiness, phonemic awarenesswas measured using the Elision subtest of the Pre-

school Comprehensive Test of Phonological andPrint Processing (Pre-CTOPPP) (Lonigan, Wagner,Torgessen, & Rashotte, 2002). This task requireschildren to segment words into phonemic compo-nents and does so by first providing visual aids.Children are shown a plate with four pictures, two ofwhich represent items to be used in a compoundword that the child is asked to segment. For example,children are presented with four pictures, one ofwhich is a picture of a bat and another a picture of aman. The examiner shows the child the pictures andnames each in turn. The examiner then states thecompound word batman and asks the child to

point to the picture of batman without bat. Twopractice items are administered in which the exam-iner provides the child with instructive feedback ifthe child answers incorrectly or fails to respond.Nine items are then administered, with the first 3items presenting pictorially assisted compoundword deletions and the last 6 presenting pictoriallyassisted syllable deletions (point to tease withoutz). Three additional compound word deletionitems are then presented without pictorial assistance.Children receive a score of 1 for each item answered

652 Blair and Razza


7/17

correctly. Cronbachs a calculated for a Head Startsample of 2,629 children participating in the HeadStart impact study is .84.

To further assess reading readiness, letter knowl-edge was measured with the test of letter knowledgedeveloped for use in the Head Start National Re-porting System Direct Child Assessment. Childrenwere presented with the letters of the alphabet inthree grids containing eight, nine, and nine letterseach. All letters were upper case and in a standardfont. The grid of eight letters containing A, O, S, B, E,C, D, and X was presented first, followed by a gridcontaining the letters P, K, T, G, N, F, L, R, and Z. Thefinal grid contained the letters U, Y, J, H, V, Q, M, W,and I. Before the presentation of each grid, childrenwere instructed to point to all the letters that theyknow on the page and to tell the experimenter thename of each one. Children received 1 point for each

letter named correctly. Cronbachs a, calculated for aHead Start sample of 878 children participating in theHead Start Quality Research Consortium study, is .96.

Intelligence. Fluid and verbal aspects of intelli-gence were measured using Ravens Colored Pro-gressive Matrices Test (Raven, 1956) and the PeabodyPicture Vocabulary Test 3 (PPVT 3) (Dunn & Dunn,1997), respectively. The Colored Progressive MatricesTest is a measure of fluid intelligence in which chil-dren are presented with a series of plates containingpatterns from which a single uniformly shaped pieceis missing. At the bottom of each plate, six versions

of the missing piece are presented, only one of whichcorrectly fits the pattern. Children were presentedwith a demonstration trial and then instructed onsuccessive trials to point to the piece that best com-pletes the pattern. Children were presented withthree sets of 12 plates of increasing difficulty. Totalnumber correct summed across the three sets wasutilized for analyses.

The PPVT 3 is a norm-referenced measure of re-ceptive vocabulary for spoken English. Children arepresented with a series of plates, each of whichcontains four line drawings of objects, places, or ac-tions. For each plate, children are asked to indicate

the picture that corresponds to a word spoken by theexperimenter. Age-based standard scores are calcu-lated relative to a mean of 100 and a standard devi-ation of 15. The internal consistency, test retestreliability, and criterion-related validity of the in-strument are high (Williams & Wang, 1997). Somechildren received a short form of the PPVT3 inwhich no basal or ceiling scoring is required. In thisversion of the measure, children were administered24 consecutive items of increasing difficulty as de-termined by item response theory analysis of PPVT

3 normative data. A scoring algorithm was utilizedto derive standard scores from the raw score data. Toadjust for a significantly lower mean standard scoreamong children receiving the short form of themeasure, we standardized scores for the short andlong forms separately before analysis. This providesan individual indicator of receptive vocabularyrelative to the mean score for the group receivingthat version of the measure and is appropriate for thepurposes of controlling for individual differences inreceptive verbal ability.

Missing Data

Of the sample of 170 children, data on academicmeasures in kindergarten were available for 141.Children for whom academic data were missing didnot differ in any way from those for whom these data

were present. Of the 141 children for whom datawere available in kindergarten, 2 were missing datafor peg tapping in kindergarten, 2 were missing datafor item selection in kindergarten, 9 were missingdata for the PPVT, 9 were missing data for teacher-reported regulation, 13 were missing data for falsebelief understanding, 18 were missing data for pegtapping measured in Head Start, and 23 were miss-ing data for flexible item selection measured in HeadStart. The reasons for missing data for all measuresprimarily related to child refusal to participate inthat aspect of the assessment and reflect the diffi-

culty of collecting data with young children. Exam-ination of variables with missing values indicatedsome systematic relations. Higher scores for letterknowledge were observed among children missingdata for either the PPVT, teacher-reported effortfulcontrol, or parent-reported effortful control. Indi-viduals missing data for item selection measured inpreschool tended to have lower scores for the PPVT.

To address missing data and to avoid the samplesize reduction that would result from listwise dele-tion of cases with missing values, full informationmaximum likelihood (FIML) estimation was utilizedto derive estimates of relations among variables in all

analyses. FIML directly estimates model parametersand standard errors using all available data and hasbeen shown to provide more accurate estimates ofregression coefficients and variance accounted forthan do listwise deletion or mean replacementmissing data strategies (Enders, 2001).

Results

We first examined zero-order correlations amongall independent variables and then used multiple



8/17

regression to examine the unique relation of themeasures of executive function (the peg-tappingmeasure of inhibitory control and the item selectionmeasure of attention shifting) to false belief under-standing and effortful control. Given prior indica-tions of a role for executive function in both effortfulcontrol and false belief understanding, we were in-terested in the extent to which the inhibitory controland attention-shifting aspects of executive functionwould make unique or overlapping contributions tothese constructs. For the regression equations weentered all variables simultaneously and interpretedcoefficients associated with Type III sums of squaresto estimate the unique effect of each independentvariable on the dependent variable.

We then examined zero-order relations among theaspects of child self-regulation and the indicatorsof early academic ability. Following this analysis, we

again used multiple regression to examine the uniquerelation of each aspect of child self-regulation to eachof the academic outcomes, controlling for verbal andfluid intelligence. In these regression equations, allpredictors were again entered simultaneously andcoefficients associated with Type III sums of squareswere interpreted to estimate the unique effect of eachindependent variable on the academic ability meas-ures. Relations reported throughout the manuscriptare significant at po.05 or below.

Table 2 presents the means and standard devi-ations for independent variables in the analysis as

well as correlations among variables. The table in-dicates that the inhibitory control and attentionshifting aspects of executive function as measuredin preschool and kindergarten are moderately stableand are also moderately correlated at both time

points. With the exception of attention-shiftingmeasured in kindergarten, the measures of executivefunction are also related to teacher-reported effortfulcontrol in preschool. False belief understanding asmeasured in preschool demonstrates relations withboth the attention-shifting and inhibitory controlaspects of executive function and with teacher-re-ported effortful control. Of some interest is thefinding that parent-reported effortful control is re-lated only to the inhibitory control aspect of execu-tive function in preschool and to teacher-reportedeffortful control. In contrast, child verbal and fluidintelligence are moderately related to all indicatorsof child self-regulation except parent-reported ef-fortful control. Examination of the relation of childage to each of the variables indicated only minimalrelations of age with false belief understanding,r5 .19, and inhibitory control measured in preschool,

r5 .18. Given the wide age range of the sample,however, we elected to control for age in all subse-quent analyses.

Relation of aspects of executive function to effortfulcontrol and false belief understanding

Having observed a moderate correlation amongvarious indicators of child self-regulation in pre-school, we next examined the unique relation of theattention-shifting and inhibitory control aspects ofexecutive function to the teacher report measure of

temperamental effortful control in preschool. Thisanalysis addressed the extent to which the two as-pects of executive function account for unique oroverlapping variance in effortful control. The resultsfrom the regression analysis interpreting Type III

Table2

Correlation Among Indicators of Child Self-Regulation

False belief

in HS

Inhibitory

control

in HS

Attention

shifting

in HS

Inhibitory

control

in K

Attention

shifting

in K

Teacher

report EC

in HS

Parent

report EC

in HS

PPVT

in HS Mean SD

False belief in HS F

1.04 0.82IC in HS .41 0.56 0.37

AS in HS .29 .34 0.68 0.25

IC in K .18 .39 .28 0.89 0.15

AS in K .29 .27 .32 .41 0.84 0.14

Teacher EC in HS .18 .38 .27 .26 .13 0.01 0.99

Parent EC in HS .01 .21 .12 .11 .01 .31 0.01 1.00

PPVT in HS .24 .27 .29 .26 .25 .37 .03 92.9 15.9

Raven in K .19 .21 .26 .20 .37 .21 .06 .30 17.26 4.1

Note. AS5attention shifting; EC5 effortful control; HS5Head Start; IC5 inhibitory control; K5kindergarten; PPVT5Peabody PictureVocabulary Test.po0.01, po0.05.

654 Blair and Razza


9/17

sums of squares controlling for age and including allvariables in the model simultaneously indicated thatthe inhibitory control aspect of executive functionwas uniquely related to teacher-reported effortfulcontrol, b5 .23. Child-receptive vocabulary, b5 .26,and parent-reported effortful control, b5 .24, werealso uniquely related to teacher report of effortfulcontrol. Attention shifting in preschool was unre-lated to the teacher report variable when in theequation as were false belief understanding andfluid intelligence. However, as noted in Table 2, thesevariables demonstrated significant zero-order asso-ciations with teacher-reported effortful control.

We then examined the relation of the attention-shifting and inhibitory control aspects of executivefunction to false belief understanding in preschool.Results from a regression analysis interpreting TypeIII sums of squares and including measures of effort-

ful control, intelligence, and executive function in themodel simultaneously indicated that both the inhi-bitory control,b5 .30, and attention shifting, b5 .19,aspects of executive function made unique contri-butions to performance on the composite measure offalse belief understanding. Child age, b5 .14, wasalso uniquely related to the outcome while effortfulcontrol and fluid and verbal intelligence were not.

Relation of executive function, effortful control, andfalse belief understanding to academic ability in kinder-garten. Having examined the relations of the inhibi-tory control and attention-shifting aspects of

executive function to effortful control and false beliefunderstanding in preschool, we then examined therelation of all aspects of child self-regulation toemerging math and literacy ability in kindergarten.Table 3 presents zero-order correlations of measuresof executive function, effortful control, and false be-lief understanding with mathematics ability, phon-emic awareness, and letter knowledge. Measures ofverbal and fluid intelligence are also included in thetable. Relations among the measures were significant

for all but the relation between letter knowledge andattention shifting measured in preschool and therelations of phonemic awareness with teacher-re-ported effortful control and receptive verbal ability.Parent-reported effortful control was unrelated to allof the academic ability measures.

Table 4 presents the results of separate regressionequations predicting the academic outcomes con-trolling for age and with all predictors entered intoeach equation simultaneously. The equations presentestimates of the unique effect of each variable on theoutcome adjusted for all other terms in the model.Results indicated that the inhibitory control aspect ofexecutive function measured both in preschool andkindergarten, teacher-reported effortful controlmeasured in preschool, and fluid intelligence asmeasured by Ravens progressive matrices test madeindependent contributions to mathematics ability in

kindergarten. False belief understanding measuredin preschool and attention shifting measured inkindergarten were also moderately related to themeasure of math ability. Although the various as-pects of self-regulation examined in the equationwere interrelated, many made independent contri-butions to early mathematics knowledge. Of par-ticular interest is the finding that inhibitory controlboth in preschool and in kindergarten made inde-pendent contributions to the outcome variable.

Examination of relations of executive function,effortful control, and false belief understanding to

phonemic awareness, also presented in Table 4, in-dicated that only the inhibitory control aspect ofexecutive function measured in kindergarten andfluid intelligence as measured by Ravens progres-sive matrices test were related to this aspect ofemergent literacy ability in kindergarten. Thus,unlike the findings for the regression equationpredicting math knowledge, no aspect of self-regu-lation measured in preschool was independentlyrelated to phonemic awareness in kindergarten.

Table3

Correlation Among Measures of Child Self-Regulation in Head Start and Kindergarten and Emerging Math and Literacy Ability Measured inKindergarten

IC

in HS

AS

in HS

False belief

in HS

Teacher EC

in HS

Parent EC

in HS

PPVT

in HS

IC

in K

AS

in K

Raven

in K Mean SD

Math .47 .26 .38 .39 .13 .39 .44 .42 .38 13.1 2.6

Phonemic awareness .21 .21 .27 .14 .07 .15 .35 .32 .32 9.7 1.8

Letter knowledge .18 .10 .22 .37 .17 .36 .25 .18 .19 23.6 5.0

Note. AS5attention shifting; EC5effortful control; HS5Head Start; IC5 inhibitory control; K5kindergarten; PPVT5Peabody PictureVocabulary Test.po0.01, po0.05.



10/17

Examination of relations of executive function,effortful control, and false belief understanding toletter knowledge, also in Table 4, indicated thatteacher-reported effortful control, false belief un-derstanding, and receptive vocabulary in preschooland the inhibitory control aspect of executive func-tion measured in kindergarten were related to letterknowledge. Here, the results are somewhat similar tothose obtained for mathematics ability, indicatingrelations of various aspects of child self-regulation to

letter knowledge and again indicating the inhibitorycontrol aspect of executive function to be a consistentcorrelate of early academic ability.

Relations among aspects of self-regulation as pre-dictors of academic outcomes. To assess the extent ofoverlapping variance in the academic outcomesassociated with executive function, false belief un-derstanding, and effortful control, we reran the re-gression equations including specific subsets of the

predictors. This analysis allowed us to examine theextent to which standardized regression coefficientschanged across models with different sets of pre-dictors. Standardized regression coefficients fromthe various models predicting mathematics ability inkindergarten are presented in Table 5. The secondcolumn of the table presents standardized coeffi-cients from the full model and the third columnpresents standardized coefficients in a model fromwhich the measures of executive function in kin-

dergarten have been removed. As expected, due tothe correlation between the inhibitory control aspectof executive function measured in preschool and inkindergarten, r5 .39, the effect for inhibitory controlin preschool increased by 47%. As shown in the ta-ble, this reflects an increase in the standardized co-efficient for this variable from b5 .17 to .25. Varianceaccounted for by this model, R25 .41, represents asmall reduction from the full model. The effects for

Table4

Regression Equation Predicting Emerging Math and Literacy Ability in Kindergarten from Child Self-Regulation Measured in Head Start and

Kindergarten

Kindergarten outcome measures

Math Phonemic awareness Letter knowledge

b SE b b SE b b SE b

PPVT in HS 0.33 0.22 .12 0.13 0.18 .08 1.27 0.47 .25

Teacher EC in HS 0.45 0.20 .18 0.04 0.17 .01 1.45 0.46 .28

False belief in HS 0.51 0.24 .16w 0.36 0.19 .16w 1.10 0.56 .18

Inhibitory control in HS 1.15 0.60 .17 0.12 0.50 .01 1.33 1.37 .12

Attention shifting in HS 0.55 0.80 .01 0.37 0.68 .06 2.23 1.87 .10

Inhibitory control in K 3.55 1.36 .20 3.21 1.13 .27 5.86 2.95 .17

Attention shifting in K 2.77 1.46 .15w 1.25 1.22 .09 0.43 3.35 .01

Raven in K 0.10 0.05 .16 0.10 0.04 .23 0.04 0.10 .03

R2 .45 .23 .30

Note. EC5 effortful control; HS5Head Start; K5kindergarten.

po0.01,

po0.05,w

po0.10.

Table5

Changes in Standardized Regression Coefficients Associated with the Prediction of Math Ability Measured in Kindergarten

Full model

Model with EF

measures in K

removed

Model with inhibitory

control measures

removed

Model with false

belief and effortful

control removed

Inhibitory control in HS .17 .25 F .28

Attention shifting in HS .01 .01 .01 .01

Inhibitory control in K .20 F F .20

Attention shifting in K .15 F .23 .15

False belief understanding in HS .16 .18 .22 F

Teacher report effortful control in HS .18 .18 .24 F

Note. HS5Head Start; K5kindergarten.

656 Blair and Razza


11/17

general fluid intelligence and receptive vocabularyremained unchanged in this and subsequent modelsand are not shown in the table. For the equationspredicting phonemic awareness and letter knowl-edge, also not shown, no changes were observedwith the removal of the kindergarten time point forthe measures of executive function.

The fourth column of Table 5 presents standard-ized coefficients in a model from which the pre-school and kindergarten measures of the inhibitorycontrol aspect of executive function have been re-moved. Here, the effect for false belief understandingincreased by approximately 37% from b5 .16 to .22,the effect for attention shifting measured in kinder-garten increased by 53% fromb5 .15 to .23, and theeffect for teacher-reported effortful control increasedby 33% from b5 .18 to .24. Again, variance ac-counted for by this model was very similar to the

full model at R2

5 .41. For the equation predictingphonemic awareness (not shown), removal of themeasures of inhibitory control resulted in a signifi-cant effect for attention shifting measured in kin-dergarten, b5 .19. For letter knowledge, no changesin predictors were associated with the removal of themeasures of inhibitory control.

Finally, the last column of Table 5 presents stan-dardized coefficients in a model from which falsebelief understanding and teacher-reported effortfulcontrol have been removed. Change was observedonly in the coefficient relating inhibitory control

measured in preschool to math ability, which in-creased by approximately 60% from b5 .17 to .28.No effects were observed for the prediction ofphonemic awareness or letter knowledge. Again,variance accounted for by this model was highlysimilar to all models at R25 .43.

Discussion

The primary aim of this study was to determine theextent to which distinct but overlapping aspects ofdeveloping self-regulation in young children wereassociated with emerging academic ability in kin-

dergarten. The findings indicated that although thevarious aspects of child self-regulation, includingexecutive function, effortful control, and false beliefunderstanding, were moderately correlated in thissample of children from low-income homes, eachtended to account for unique variance in earlymathematics or literacy ability. For mathematicsknowledge and letter knowledge, self-regulationmeasured both in preschool and in kindergartenaccounted for significant variation. Teacher-reportedeffortful control and the inhibitory control aspect of

executive function were positively associated withmathematics ability and with the letter knowledgeaspect of emerging literacy. False belief understand-ing was significantly related to letter knowledge andexhibited a marginally significant relation withmathematics knowledge and phonemic awareness.Unlike mathematics and letter knowledge, no pre-school measures of self-regulation were related tophonemic awareness. As with mathematics knowl-edge, however, fluid intelligence was uniquelypositively associated with phonemic awareness. Incontrast, the measure of receptive verbal ability wasuniquely associated only with letter knowledge.Overall, the results indicated some overlap in theprediction of early academic ability from measuresof self-regulation but also indicated some uniquerelations of predictors to outcomes. We first consideroverlap in prediction and then consider the reasons

for unique relations among predictors and outcomes.

Executive Function and Developing Academic Ability

Among the various aspects of self-regulationexamined, the inhibitory control aspect of executivefunction was the only one to be independently re-lated to all three measures of academic ability. Thisrole for inhibitory control in the study of self-regu-lation and early academic ability is of some interest,given work suggesting inhibitory control to be acentral feature of the development of executive

function in early childhood (Diamond, 2002; Dia-mond et al., 2005; Kirkham et al., 2003). Evidence infavor of the inhibitory control account for the de-velopment of executive function is provided by dataindicating that childrens performance on a classicmeasure of executive function in young children, thedimensional change card sort task, is improvedwhen the conditions of task administration reducethe inhibitory control requirement of the task but notthe relational complexity among the items to besorted (Brooks et al., 2003; Diamond et al., 2005; Kloo& Perner, 2005). Although by no means the onlyworking hypothesis concerning the development of

executive function, the inhibitory account provides acomprehensive explanation for the emergence of acentral aspect of self-regulation in early childhood.

In contrast to Diamonds inhibitory account, Zel-azo and colleagues (Zelazo et al., 2003; Zelazo, Frye,& Rapus, 1996; Zelazo & Muller, 2002) suggest thatdevelopmental increases in executive function gen-erally, including inhibitory control, reflect advancesin an underlying cognitive ability to represent com-plex rule structures. Although data from this studycannot directly address the inhibitory versus



12/17

complexity distinction in the development of ex-ecutive function, the findings are consistent with theidea that executive function is a central aspect ofcognitive development during the preschool period.Furthermore, the inhibitory and cognitive complex-ity accounts for the development of executive func-tion suggest somewhat different interpretations ofthe relation of executive function to academic ability.An inhibitory account suggests that the ability toinhibit prepotent response tendencies in the face ofirrelevant or distracting information, such as in thecontext of a math problem or when discriminatingletters or phonemes, is a unique contributor to de-veloping academic ability over and above that ofspecific knowledge of problem elements or solutions.In contrast, the cognitive complexity account placesemphasis on the role of knowledge of problemelements and relations among these elements as

central to executive function and its relation to aca-demic ability. In the cognitive complexity account,the ability to maintain knowledge of embedded rulestructures enables reflection, which in turn allows forthe inhibition of prepotent response tendencies, theretention of information in working memory, and theappropriate shifting of attention among aspects of agiven problem or task. At the risk of oversimplifyingdistinctions between the two approaches, an inhibi-tory account emphasizes doing somewhat more thanknowing, while the cognitive complexity accountemphasizes knowing somewhat more than doing.

That is, the inhibitory account suggests that an in-dividual can have knowledge of correct relationsamong problem elements but still experience diffi-culty inhibiting a prepotent response tendency thatinterferes with the correct solution of the problembased on knowledge alone. In contrast, the com-plexity account suggests that knowledge of thehigher order rule structures that define relationsamong problem elements allows for the executivefunctioning required for problem solution.

Executive function and ability in mathematics. Al-though both the inhibitory and cognitive complexityaccounts for the development of executive function

have merit, the finding that inhibitory control both inpreschool and kindergarten accounted for uniquevariance in mathematics ability provides further in-dication of the salience of this aspect of executivefunction for the development of early academicability. The relation between inhibitory control andmathematics ability reported here with preschoolersfrom low-income homes is consistent with priorwork in preschoolers from typical income homes inwhich the inhibitory control but not the attention-shifting aspect of executive function was related to

mathematics ability (Espy et al., 2004). However, in astudy examining mathematics ability in children inthe first grade, at age 7 years, independent effectswere reported for inhibitory control, attention shift-ing, and working memory aspects of executivefunction, with the most robust relation being that forworking memory (Bull & Scerif, 2001).

Findings of moderate to strong relations betweenexecutive function and mathematics ability in thedevelopmental literature are consistent with neuro-scientific work demonstrating an overlap in theneural substrates supporting executive function andnumerical ability and quantitative reasoning. Usingfunctional brain imaging, parietal frontal corticalcircuitry known to be important for executive func-tion abilities (Klingberg, 2006) has been shown tosupport the representation of numerosity and thesolution of simple calculation problems (Dehaene,

Molko, Cohen, & Wilson, 2004; Prabhakaran, Rypma,& Gabrieli, 2001; Simon et al., 2004). Similarly, usingsingle-unit physiology with nonhuman primates,neurons tuned to specific numerosities have beenidentified in prefrontal and parietal cortical areasimportant for both numerical ability and executivefunction (Neider, 2005; Neider & Miller, 2004).

Also, at the sociological level, relations betweenexecutive function and mathematics ability are con-sistent with historical evidence indicating a shift inthe math curriculum for children in the early elem-entary grades toward instructional strategies and

problem types designed to exercise executive func-tion abilities directly. Historical review of mathe-matics textbooks in use in U.S. public schools overthe last 100 years indicates an increasing pedagogicalemphasis on the promotion of thinking skills andexecutive function abilities as central to the devel-opment of early competence in math (Blair, Gamson,Thorne, & Baker, 2005). Primarily this emphasis isseen in the use of pattern completion and visualspatial working memory problems in the earlyelementary grades that require children to use ex-ecutive function to reason about problem elements.An important feature of pattern completion and

working memory types of problems is that no au-tomatized knowledge or formal problem-solving al-gorithm can be provided to facilitate their solution(Blair, Knipe, Cummings, Baker, Gamson, Eslinger, &Thorne, in press).

The relation of executive function to mathematicsability is particularly apparent when consideredfrom the standpoint of what it is that children areexpected to do, cognitively speaking, when solvingmathematics problems. Proficiency in mathematicsat all levels requires the individual to reason actively

658 Blair and Razza


13/17


14/17

in this study is consistent with results obtained inother samples of preschool children (Carlson &Moses, 2001; Davis et al., 2002; Rothbart et al., 2003).Unlike prior work (Carlson & Mandell et al., 2004),however, we also identified an independent relationbetween attention shifting and false belief under-standing even when covarying inhibitory control.This may be due to the preschool sample in thisstudy being somewhat larger and older (n5 170,mean age 5 years 1 month) than preschool samples inother studies (n5 81, mean age 4 years 0 month inCarlson & Mandell et al., 2004).

Conclusions and Limitations

Although the results provide relatively straight-forward evidence for relations among aspects of self-regulation and early academic ability, it is necessary

to consider the extent to which the method ofmeasurement may have influenced the findings. It isalso necessary to consider the extent to which tem-peramental effortful control as an aspect of self-regulation reported on by teachers may reflect gen-eral perceptions of child ability, particularly aca-demic ability. Method of measurement for executivefunction and false belief understanding was throughdirect assessment while method of measurement foreffortful control was through questionnaires com-pleted by parents and teachers. It may be that if dataon executive function and false belief understanding

had been gathered through questionnaires or if in-formation on effortful control had been obtainedusing behavioral assessments, results would havebeen different. Certainly the way in which psycho-logical constructs are measured can influence rela-tions among variables.

In the interpretation of the effect for teacher reportof effortful control on academic outcomes, it may bethat teachers ratings of child temperament areunique to the classroom and school context and in-fluenced by perceptions of child academic ability.Effects for teacher-reported effortful control on aca-demic outcomes, however, were observed when

controlling for aspects of verbal and fluid intelli-gence as well as for measures of executive functionand false belief understanding. This indicates thatalthough teachers reports may have been influencedby academic ability, the relation of teacher-reportedeffortful control to child academic ability persistseven when taking into account objective predictorsof child academic competence. Whether effortfulcontrol as reported by teachers should be consideredas temperament per se, or perhaps an aspect oftemperament-like behavior unique to the classroom

setting and reflecting engagement in and liking ofschooling, is an important direction for future re-search. Similarly, although the null findings forparent-reported effortful control are puzzling andnot interpretable, our results suggest that furtherdifferentiation of the relation of temperament tochildrens early progress in school is needed. Parentsobserve children in a much wider range of contextsthan do teachers and no doubt have an accurate al-though distinct view of the effortful control aspect oftemperament. The relatively poor internal consist-ency estimates of the parent report subscales in thislow-income sample, however, may reflect misun-derstanding or misreading of items by parents andmay perhaps limit the validity of this measure in thisanalysis.

A further issue related to measurement is the ab-sence in the full-regression models of a unique re-

lation of attention shifting as measured by the itemselection task to any of the measures of academicability. The series of regression equations in Table 5indicated that the relation of attention shifting tomathematics ability and phonemic awareness wasaccounted for by inhibitory control. This suggeststhat both the peg tapping and item selection mea-sures provide useful assessments of executive func-tion but, as with most if not all measures of theconstruct, they overlap in certain respects. The rela-tion of abstraction as an aspect of attention shifting toemerging academic ability, however, may be some-

what later in developing in this sample and even inkindergarten may be subsumed by inhibitory con-trol. It is important to acknowledge, however, thepossibility that error in measurement at the pre-school assessment may have influenced the results,as the item selection measure of attention shifting issomewhat more complex to administer than the peg-tapping measure of inhibitory control.

Also, in studies with young children, participantfatigue can sometimes affect the validity of assess-ments. At 45 min, our assessment sessions were wellwithin the range of what most children were able tobear. For some children, however, fatigue may have

influenced performance on the various measures. Byskillfully engaging children in our assessment pro-cedures, however, we believe that we effectivelyminimized any effects of child fatigue on our data.

In conclusion, although issues associated withmeasurement may have influenced the findings tosome degree, studies demonstrating relations ofvarious aspects of self-regulation including effortfulcontrol, false belief understanding, and executivefunction to indicators of school readiness may be ofparticular value to efforts to ensure the early school

660 Blair and Razza


15/17

success of children from low-income homes. In theattempt to improve educational achievement anddecrease inequities in educational progress associ-ated with socioeconomic status, it is important tounderstand the nature of multiple influences onearly progress in school. Of course, the results fromthis study, as with the vast majority of studiesexamining self-regulation in young children, arecorrelational and cannot establish a definitive causalrelation between self-regulation and academic abil-ity. Given the central role of self-regulation in childdevelopment, examination of the construct fromboth temperament-based and neuropsychologicalapproaches can improve understanding of behaviorsthat underlie early success in school and serve astargets for systematic efforts to enhance the qualityof childrens early educational experiences. The de-velopment and experimental evaluation of well-

specified preschool and early school-age curriculadesigned to promote academic achievement by fos-tering self-regulation is an important direction forresearch on school readiness. As a first step in thisdirection, two promising studies, one with a pre-school sample (Rueda, Rothbart, McCandliss, Sac-comanno, & Posner, 2005) and the other with aschool-age sample with attention deficit hyperactiv-ity disorder (Klingberg et al., 2005), have demon-strated the effects of intensive computer-basedtraining on certain of the aspects of executive func-tion. An important direction for future work will be

the replication with diverse populations of effectsseen in these studies and assessment of the gener-alizability of training effects to academic ability.

References

Arnsten, A., & Shansky, R. (2004). Adolescence: Vulnerableperiod for stress-induced cortical function? Annals of theNew York Academy of Sciences,1021, 143147.

Astington, J. W., & Jenkins, J. M. (1995). Theory of minddevelopment and social understanding. Cognition andEmotion,9, 151165.

Bachevalier, J., & Loveland, K. A. (2006). The orbitofrontal-amygdala circuit and self-regulation of social-emotional

behavior in autism. Neuroscience and Biobehavioral Re-views,30, 97117.

Bechara, A. (2004). The role of emotion in decision-making:Evidence from neurological patients with orbitofrontaldamage.Brain and Cognition,55, 3040.

Blair, C. (2002). School readiness: Integrating cognition andemotion in a neurobiological conceptualization of childfunctioning at school entry. American Psychologist, 57,111127.

Blair, C. (2006). How similar are fluid cognition and gen-eral intelligence? A developmental neuroscience per-

spective on fluid cognition as an aspect of humancognitive ability. Behavioral and Brain Sciences, 29, 109125.

Blair, C., Gamson, D., Thorne, S., & Baker, D. (2005). Risingmean IQ: Changing cognitive demand of mathematicseducation for young children, population exposure to

formal schooling, and the neurobiology of the prefrontalcortex.Intelligence,33, 93 106.

Blair, C., Granger, D., & Razza, R. P. (2005). Cortisol re-activity is positively related to executive function inpreschool children attending Head Start. Child Develop-ment,76, 554567.

Blair, C., Knipe, H., Cummings, E., Baker, D., Gamson, D.,Eslinger, P., et al. (in press) A developmental neurosci-ence approach to the study of school readiness. In R.Pianta, M. Cox, & K. Snow (Eds.), School readiness and thetransition to kindergarten in the era of accountability. Balti-more: Brooks.

Brooks, P. J., Hanauer, J. B., Padowska, B., & Rosman, H.(2003). The role of selective attention in preschoolersrule use in a novel dimensional card sort. Cognitive De-velopment,18, 195215.

Bull, R., & Scerif, G. (2001). Executive functioning as apredictor of childrens mathematics ability: Inhibition,switching, and working memory. Developmental Neuro-psychology,19, 273293.

Bush, G., Luu, P., & Posner, M. I. (2000). Cognitive andemotional influences in the anterior cingulate cortex.Trends in Cognitive Sciences,4, 215222.

Carlson, S., Mandell, D. J., & Williams, L. (2004). Executivefunction and theory of mind: Stability and predictionfrom ages 2 to 3. Developmental Psychology, 40, 11051122.

Carlson, S., & Moses, L. (2001). Individual differences ininhibitory control and childrens theory of mind. ChildDevelopment,72, 1032 1053.

Carlson, S. M., Moses, L. J., & Claxton, L. J. (2004). Indi-vidual differences in executive functioning and theoryof mind: An investigation of inhibitory control andplanning ability.Journal of Experimental Child Psychology,87, 299319.

Davis, E. P., Bruce, J., & Gunnar, M. R. (2002). The anteriorattention network: Associations with temperament andneuroendocrine activity in 6-year-old children. Develop-mental Psychobiology,40, 4356.

Dehaene, S., Molko, N., Cohen, L., & Wilson, A. J. (2004).Arithmetic and the brain. Current Opinion in Neurobiol-ogy,14, 218224.

Diamond, A. (2002). Normal development of prefrontalcortex from birth to young adulthood: Cognitive func-tions, anatomy, and biochemistry. In D. Stuss & R.Knight (Eds.),Principles of frontal lobe function (pp. 466503). New York: Oxford.

Diamond, A., Carlson, S. M., & Beck, D. (2005). Preschoolchildrens performance in task switching on the Di-mensional Change Card Sorting task: Separating thedimensions aids the ability to switch. DevelopmentalNeuropsychology,28, 689729.



16/17

Diamond, A., Prevor, M. B., Callendar, G., & Druin, D. P.(1997). Prefrontal cognitive deficits in children treatedearly and continuously for PKU. Monographs of the So-ciety for Research in Child Development,62.

Diamond, A., & Taylor, C. (1996). Development of an as-pect of executive control: Development of the abilities to

remember what I said and to do as I say, not as I do.Developmental Psychobiology,29, 315334.

Drevets, W. C., & Raichle, M. E. (1998). Reciprocal sup-pression of regional cerebral blood flow during emo-tional versus higher cognitive processes: Implicationsfor interactions between emotion and cognition. Cogni-tion and Emotion,12, 353385.

Dunn, L. W., & Dunn, L. M. (1997). Peabody picture vo-cabulary testFrevised. Circle Pines, MN: American Gui-dance Service.

Eisenberg, N., Valiente, C., Fabes, R. A., Smith, C. L., Reiser,M., Shepard, S. A., et al. (2003). The relations of ef-fortful control and ego control to childrens resiliencyand social functioning. Developmental Psychology, 39,761776.

Enders, C. K. (2001). The performance of the full infor-mation maximum likelihood estimator in multiple re-gression models with missing data.Journal of Educationaland Psychological Measurement,61, 713740.

Espy, K., McDiarmid, M., Kwik, M., Stalets, M., Hamby, A.,& Senn, T. (2004). The contribution of executive func-tions to emergent mathematics skills in preschool chil-dren.Developmental Neuropsychology,26, 465486.

Gerardi-Caulton, G. (2000). Sensitivity to spatial conflictand the development of self-regulation in children 24 36 months of age. Developmental Science,3, 397404.

Gray, J. R., Braver, T. S., & Raichle, M. E. (2002). Integration

of cognition and emotion in the lateral prefrontal cortex.Proceedings of the National Academy of Sciences,99, 41154120.

Groenewegen, H. J., & Uylings, H. B. M. (2000). The pre-frontal cortex and the integration of sensory, limbic, andautonomic information. In H. Uylings, C. Van Eden, J.De Bruin, M. Feenstra, & C. Pennartz (Eds.), Progress inBrain research, Vol. 126, cognition, emotion and autonomicresponses: The integrative role of the prefrontal cortex andlimbic structures (pp. 3 28). Amsterdam, the Nether-lands: Elsevier.

Happaney, K., & Zelazo, P. D. (2003). Inhibition as aproblem in the psychology of behavior. DevelopmentalScience,6, 468470.

Holmes, H. A., Black, C., & Miller, S. A. (1996). A cross-taskcomparison of false belief understanding in a Head Startpopulation.Journal of Experimental Child Psychology, 63,263285.

Howse, R. B., Calkins, S. D., Anastopoulos, A. D., Keane, S.P., & Shelton, T. L. (2003). Regulatory contributors tochildrens kindergarten achievement. Early Educationand Development,14, 101 119.

Hughes, C. (1998). Executive function in preschoolers:Links with theory of mind and verbal ability. British

Journal of Developmental Psychology,16, 233253.

Hughes, C., & Dunn, J. (1998). Understanding mind andemotion: Longitudinal associations with mental-statetalk between young friends. Developmental Psychology,34, 1026 1037.

Hughes, C., Dunn, J., & White, A. (1998). Trick or Treat?Patterns of cognitive performance and executive func-

tion among hard to manage preschoolers. Journal ofChild Psychology and Psychiatry,39, 981994.

Hughes, C., & Ensor, R. (2005). Theory of mind and ex-ecutive function in two-year-olds: A family affair? De-velopmental Neuropsychology,28, 645668.

Jacques, S., & Zelazo, P. D. (2001). The Flexible Item Se-lection Task (FIST): A measure of executive function inpreschoolers. Developmental Neuropsychology, 20, 573591.

Kerr, A., & Zelazo, P. D. (2004). Development of hotexecutive function: The childrens gambling task. Brainand Cognition,55, 148157.

Kirkham, N., Cruess, L., & Diamond, A. (2003). Helpingchildren apply their knowledge to their behavior on adimension-switching task.Developmental Science,6, 449 476.

Klingberg, T. (2006). Development of a superior frontalintraparietal network for visuo-spatial working memory.Neuropsychologia,44, 2171 2177.

Klingberg, T., Fernell, E., Olesen, P., Johnson, M., Gustafs-son, P., Dahlstrom, K., et al. (2005). Computerizedtraining of working memory in children with ADHDFarandomized, controlled trial. Journal of the American

Academy of Child and Adolescent Psychiatry,44, 177186.Kloo, D., & Perner, J. (2005). Disentangling dimensions in

the dimensional card-sort task. Developmental Science,8,4456.

Kochanska, G., Murray, K., Jacques, T., Koenig, A., &Vandegeest, K. (1996). Inhibitory control in young chil-dren and its role in emerging internalization. Child De-velopment,67, 490507.

Kopp, C. (1989). Regulation of distress and negative emo-tions: A developmental view. Developmental Psychology,25, 343354.

Ladd, G., Birch, S., & Buhs, E. (1999). Childrens social andscholastic lives in kindergarten: Related spheres of in-fluence?Child Development,70, 1373 1400.

Ladd, G., Herald, S. L., & Kochel, K. P. (2006). Socialprerequisites of school readiness. Early Education andDevelopment,17, 115 150.

Liu, D., Diorio, J., Francis, D. D., & Meaney, M. J. (2000).Maternal care, hippocampal neurogenesis, andcognitive development in rats. Nature Neuroscience, 3,799806.

Lonigan, C., Wagner, R., Torgessen, J., & Rashotte, C.(2002). Preschool Comprehensive Test of Phonological andPrint Processing.

Lupien, S. J., Gillin, C. J., & Hauger, R. L. (1999). Workingmemory is mores sensitive than declarative memoryto the acute effects of corticosteroids: A doseresponse study in humans. Behavioral Neuroscience, 113,420430.

662 Blair and Razza


17/17

Martin, R. P., Drew, D., Gaddis, L. R., & Moseley, M. (1988).Prediction of elementary school achievement from pre-school temperament: Three studies. School PsychologyReview, 17, 125137.

Mashburn, A. J., & Pianta, R. C. (2006). Social relationshipsand school readiness. Early Education and Development,

17, 151176.McLean, J. F., & Hitch, G. J. (1999). Working memory im-

pairments in children with specific arithmetic learningdifficulties.Journal of Experimental Child Psychology, 74,240260.

Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H.,& Howerter, A. (2000). The unity and diversity of ex-ecutive functions and their contributions to complexfrontal lobe tasks: A latent variable analysis. CognitivePsychology,41, 49 100.

Neider, A. (2005). Counting on neurons: The neurobiologyof numerical competence. Nature Reviews: Neuroscience,6, 177190.

Neider, A., & Miller, E. (2004). A parietal frontal networkfor visual numerical information in the monkey. Pro-ceedings of the National Academy of Sciences USA, 101,74577462.

Normandeau, S., & Guay, F. (1998). Preschool behavior andfirst-grade school achievement: The mediational role ofcognitive self-control. Journal of Educational Psychology,90, 111121.

Peters, R., & Blair, C. (2003). False belief understanding inchildren living in poverty. Early Education and Develop-ment,14, 425439.

Posner, M., & Rothbart, M. (2000). Developing mechanismsof self-regulation. Development and Psychopathology, 12,427441.

Prabhakaran, V., Rypma, B., & Gabrieli, J. D. E. (2001).Neural substrates of mathematical reasoning: A func-tional magnetic resonance imaging study of neocorticalactivation during performance of the Necessary Arith-metic Operations Test. Neuropsychology,15, 115 127.

Putnam, S. P., & Rothbart, M. K. (2006). Development ofshort and very short forms of the Childrens BehaviorQuestionnaire.Journal of Personality Assessment 87, 102 112.

Raven, J. C. (1956). Colored progressive matrices. San Anto-nio, TX: Psychological Corporation.

Raver, C. (2004). Placing emotional self-regulation insociocultural and socioeconomic contexts. Child Devel-opment,75, 346353.

Razza, R. P. (2005). Relations among false belief understanding,executive function, and social competence: A longitudinalstudy. Unpublished doctoral dissertation. PennsylvaniaState University.

Rothbart, M. K., & Ahadi, S. A. (1994). Temperament andthe development of personality. Journal of AbnormalPsychology,103, 5566.

Rothbart, M., Ellis, P., Rueda, R., & Posner, M. (2003). De-veloping mechanisms of temperamental effortful con-trol.Journal of Personality,71, 1113 1143.

Rothbart, M., & Jones, L. (1998). Temperament, self-regu-lation, and education.School Psychology Review,27, 479491.

Rueda, R., Rothbart, M., McCandliss, B., Saccomanno, L., &Posner, M. (2005). Training, maturation, and genetic in-fluences on the development of executive attention.Proceedings of the National Academy of Sciences USA, 102,1493114936.

Simon, O., Kherif, F., Flandin, G., Poline, J-B., Riviere, D.,Mangin, J-F., et al. (2004). Automatized clustering andfunctional geometry of human parietofrontal networksfor language, space, and number. NeuroImage,23, 11921202.

Stifter, C., Spinrad, T., & Braungart-Rieker, J. (1999). To-ward a developmental model of child compliance: Therole of emotion regulation in infancy.Child Development,70, 2132.

Swanson, H. L. (1999). Reading comprehension andworking memory in learning-disabled readers: Is thephonological loop more important that the executivesystem? Journal of Experimental Child Psychology, 72, 1 31.

Valiente, C., Eisenberg, N., Smith, C., Reiser, M., Fabes, R.A., Losoya, S., et al. (2003). The relations of effortfulcontrol and reactive control to childrens externalizingproblems: A longitudinal assessment. Journal of Person-ality,71, 11711196.

Williams, K. T., & Wang, J. (1997). Technical references to thePeabody Picture Vocabulary Test: Third Edition. CirclePines, MN: American Guidance Service Inc.

Wimmer, H., & Perner, J. (1983). Beliefs about beliefs:Representation and constraining function of wrong be-liefs in young childrens understanding of deception.Cognition,13, 103128.

Zelazo, P. D., Frye, D., & Rapus, T. (1996). An age-relateddissociation between knowing rules and using them.Cognitive Development,11, 3763.

Zelazo, P. D., & Muller, U. (2002). Executive function intypical and atypical development. In U. Goswami (Ed.),

Handbook of childhood cognitive development (pp. 445 469). Oxford, UK: Blackwell.

Zelazo, P. D., Muller, U., Frye, D., & Marcovitch, S. (2003).The development of executive function in early child-hood.Monographs of the Society for Research on Child De-velopment,68.


attention control_blair razza.pdf

Documents