A Neuropsychological Examination of the Underlying … are thought to play a major role in...

Developmental Psychology1998, Vol. 34, No. 5, 956-%9

Copyright 1998 hy the American Psychological Association, Inc.0012-1649/98/$3.00

A Neuropsychological Examination of the Underlying Deficit inAttention Deficit Hyperactivity Disorder: Frontal Lobe Versus

Right Parietal Lobe Theories

Christine J. Aman, Ralph J. Roberts, Jr., and Bruce F. PenningtonUniversity of Denver

A neuropsychological approach was used to examine the frontal lobe and right parietal lobe theoriesof attention deficit hyperactivity disorder (ADHD). Considerable attempts were made to select aspure a group of ADHD boys as possible. The performance of 10-14-year-old ADHD boys (n =22), both on and off stimulant medication, was compared with the performance of non-ADHD controlboys (n = 22) on tasks purported to assess frontal lobe functioning (Stopping Task, Antisaccade Task,Tower of Hanoi) and right parietal lobe functioning (Visual-Spatial Cuing Task, Turning Task, SpatialRelations). Three important findings emerged: (a) unmedicated ADHD boys exhibited performancedeficits on tasks in both frontal and parietal domains compared with control boys, (b) unmedicatedADHD boys appeared to be more severely impaired on the frontal tasks than on the parietal tasks,and (c) medicated ADHD boys performed better in both task domains compared with unmedicatedADHD boys. Several alternative interpretations of the results are discussed.

Attention deficit hyperactivity disorder (ADHD) is one ofthe most commonly diagnosed childhood disorders as well asone of the most extensively studied (see Barkley, 1990, for areview). Despite the consideration this disorder has received,fundamental questions remain unanswered, such as the nature ofthe underlying cognitive and neurobiological deficits in ADHD.Recent neurobiological and neuropsychological evidence sug-gests that two brain regions, the frontal lobes and the rightparietal lobe, may be implicated in ADHD. In the present study,a neuropsychological approach was used to examine the frontallobe and right parietal lobe theories of ADHD in a carefullyselected group of children with the disorder.

Frontal Lobe and Right Parietal LobeTheories of ADHD

The frontal lobes of the brain, particularly the prefrontal re-gions, are thought to play a major role in performing executive

Christine J. Aman, Ralph J. Roberts, Jr., and Bruce F. Pennington,Department of Psychology, University of Denver.

Portions of this work were presented at the annual meeting of theAmerican Psychological Society, San Diego, California, 1992, and atthe biennial meeting of the Society for Research in Child Development,New Orleans. Louisiana, 1993.

This research was supported by grants from the University of Colora-do's Developmental Psychobiology Endowment Fund, the MacArthurFoundation, and the National Institute of Child Health and Human Devel-opment Center (HD27802) and by an American Psychological Associa-tion Dissertation Research Award, a National Institute of Mental HealthMERIT Award (MH38820), and a Research Scientist DevelopmentAward (MH00419).

We thank A. Olmos-Gallo and S. Wiebke for their technical assistanceregarding the project.

Correspondence concerning this article should be addressed to Chris-tine J. Aman, Department of Psychology, University of Denver, Denver,Colorado 80208. Electronic mail may be sent to [email protected].

functions, such as planning and implementing goal-orientedstrategies, controlling impulses, inhibiting prepotent responses,shifting and maintaining strategy sets, and organizing and imple-menting search strategies (e.g., Fuster, 1985; Stuss & Benson,1984) in addition to supporting working memory functions(Goldman-Rakic, 1987). The parietal lobes are thought to bespecialized for integrating sensory input from visual regions ofthe brain as well as from other sensory areas. The posteriorparietal cortex has been found to be essential for accurate visu-ally guided motor activity and for spatial perception and spatialattention (Andersen, 1988). The right parietal lobe appears tobe dominant for these functions, as right-hemispheric lesions(in right-handed individuals) are associated with more frequentand more severe deficits (e.g., Piercy, Hecaen, & Ajuriaguerra,1960), including disturbances involving both hemifields (Heil-man & Van Den Abell, 1980).

The frontal lobe theory of ADHD was introduced in the 1930swhen practitioners and investigators noted the behavioral simi-larities between patients with frontal lobe lesions and childrenwith ADHD symptoms (Levin, 1938). Both groups, it was ob-served, exhibit impaired response inhibition, excessive restless-ness and distractibility, and inattention. The right parietal lobetheory of ADHD was introduced more recently by investigatorswho have observed that the attentional deficits and hypoarousaloften observed in ADHD children are similar to behavioralsymptoms observed in individuals with right parietal lobe dam-age (e.g., Mesulam, 1981; Vocller, 1986, 1991; Voeller & Heil-man, 1988). Unlike the frontal theory of ADHD, which is asso-ciated with an overwhelming degree of face validity, the rightparietal theory of ADHD has received little attention. Tn anycase, evidence exists to support both theories.

Neurobiological Evidence

Although previous studies using traditional brain imagingtechniques, such as computed tomography (CT) scans, have

956

UNDERLYING DEFICIT IN ADHD 957

failed to find structural anomalies associated with the brains ofADHD children (e.g., Harcherik et al., 1985; Shaywitz, Shay-witz, Byrne, Cohen, & Rothman, 1983), the results of recentstudies with high-resolution magnetic resonance imaging (MR1)techniques suggest that such aberrations may exist. Hynd, Sem-rud-Clikeman, Lorys, Novey, and Eliopulos (1990) found thatunlike non-ADHD children whose frontal regions are asymmet-rical with the left anterior region being smaller than the right,ADHD children appear to have symmetrical anterior regions.Similarly, Hynd et al. (1993) found reversed patterns of asym-metry of the head of the caudate nucleus, a portion of the basalganglia that is heavily connected to the prefrontal cortex; the leftregion was smaller than the right in ADHD children, whereas itwas larger than the right in normal control children. In addition,Hynd et al. (1991) have shown that areas of the corpus callosumcontaining fibers connecting homologous anterior (frontal) andposterior (parietal) cortical regions in the left and right hemi-spheres are smaller in ADHD children than in non-ADHDchildren.

Results of functional imaging studies generally corroboratethese structural findings regarding ADHD. Specifically, Lou andhis colleagues (Lou, Henriksen, & Bruhn, 1984; Lou, Henrik-sen, Bruhn, Borner, & Nielsen, 1989), using regional cerebralblood flow procedures, found hypoperfusion in the frontal lobesand caudate-striatal regions of ADHD children at rest. The hypo-perfusion of the striatal regions was ameliorated by the adminis-tration of stimulant medication (methylphenidate). More re-cently, Zametkin et al. (1990) found that adults who had ADHDsymptoms persisting since childhood, but who had never beentreated with stimulant medication, showed reduced cerebral glu-cose metabolism relative to non-ADHD adults in premotor andsuperior prefrontal regions as well as in right posterior parietalregions.

Neuropsychological Evidence

Children with ADHD have been found to perform poorly onseveral neuropsychological tasks purported to assess frontal lobedysfunction. These tasks include the Wisconsin Card SortingTest (e.g., Boucugnani & Jones, 1989; Chelune, Ferguson,Koon, & Dickey, 1986), the Stroop Test (e.g., Gorenstein, Mam-mato, & Sandy, 1989; Grodzinsky & Diamond, 1992), the Towerof Hanoi (Pennington, Groisser, & Welsh, 1993), the HandMovements Test (e.g., Breen, 1989), and other motor controltasks, such as the Go/No-Go Test (e.g., Shue & Douglas, 1992).Deficits on these tasks involve response preservation, motordisinhibition, planning difficulties, and inattention.

Neuropsychological evidence implicating the right parietallobe in ADHD comes from the Letter Cancellation Task, onwhich ADHD children have been found to make significantlymore errors of omission and more left-sided errors than non-ADHD children (Voeller & Heilman, 1988). In addition, theperformance of ADHD children has been found to be poor onmental rotation tasks (Snow, 1990), resembling that of patientswho have sustained right parietal lobe damage (e.g., Ditunno &Mann, 1990).

The Present Study

The present study was conducted in light of the foregoingevidence suggesting frontal lobe and right parietal lobe involve-

ment in ADHD. The purpose of this research was to examinethe performance of a well-defined sample of ADHD boys onfrontal as well as parietal tasks in a single study and to examinethe effects of medication on this performance. Specifically, aneuropsychological approach was used to examine the perfor-mance of 10- to 14-year-old ADHD boys, both on and off stimu-lant medication, on several tasks that are purported to assessfrontal lobe and right parietal lobe functioning compared to theperformance of age- and IQ-matched non-ADHD control boys.Comparisons between the performance of the ADHD boys onversus off stimulant medication were also made to assess medi-cation effects on performance.

The tasks were selected because they have been shown to besensitive to frontal lobe or right parietal lobe functioning orbecause they are similar to tasks that have been shown to besensitive to the functioning of one of these brain regions andare experimentally more rigorous than the original tasks. Threetasks that are purported to assess frontal lobe functioning (Stop-ping Task, Antisaccade Task, Tower of Hanoi) and three tasksthat are thought to be sensitive to right parietal lobe functioning(Visual-Spatial Cuing Task, Turning Task, Spatial Relations)were included.

Research Questions

There were three main questions addressed in this study. Theprimary question of interest concerned which theory, the frontallobe theory of ADHD or the right parietal lobe theory of ADHD,would be supported. Given the existing neuropsychological andneurobiological findings regarding ADHD, we expected theADHD children to exhibit deficits on aspects of tasks in boththe frontal and the parietal domains. A related question waswhether the ADHD boys would exhibit more severe deficits onone set of tasks compared to the other. Given the lack of studiesthat have taken a neuropsychological approach to test both fron-tal lobe and right parietal lobe deficits in the same group ofADHD children, no specific predictions were made regardingthe relative severity of these putative deficits. The third, andfinal, question of interest involved whether stimulant medicationwould differentially affect performance on the frontal and pari-etal tasks. Specifically, we were interested in whether stimulantswould reduce or even eliminate any deficits exhibited by theADHD boys off medication. In general, we expected that stimu-lant medication would be associated with better overall perfor-mance in the ADHD children compared to their off medicationperformance.

Method

Participants

The participants were 22 ADHD boys (21 Caucasian, 1 Hispanic)and 22 non-ADHD control boys (21 Caucasian, 1 Hispanic) 10 to 14years of age, who were paid for their participation. All of the boys wereright-handed, as demonstrated on the Edinburgh Handedness Inventory(Oldfield, 1971), and most were from middle- to upper-middle-classfamilies, as measured by the Two Factor Index of Social Position (Hol-lingshead, 1957). The boys in the two groups were matched pairwiseon age (within 7 months) and on an estimate of IQ (within 6 points).Descriptive characteristics of the sample are presented in Table 1.

958 AMAN, ROBERTS, AND PENNINGTON

Table 1Descriptive Characteristics and Criterion and Related Measures: Mean Scores, Standard Deviations, and Group Differences

Measure

Group

ADHD

Descriptive characteristics

Age (years)MSD

WISC-R IQ estimateMSD

WISC-R VocabularyMSD

WISC-R Block DesignMSD

Holhngshead IndexMSD

Mother's education3

MSD

Father's education"MSD

12.141.21

110.508.18

12.002.27

11.552.22

28.6415.31

2.451.14

2.321.21

ADHD criterion measures

CBCL: T score Hyperactive scaleMSD

DICA: number of ADD symptomsMSD

HSQ: % of situationsMSD

80.416.71

17.002.90

75.1417.73

Control

12.131.21

110.007.34

11.411.82

12.002.07

24.146.66

2.090.75

1.860.64

55.000.00

1.501.50

12.369.38

P

ns

ns

ns

ns

ns

ns

ns

<.001

<.001

<.001

Measure

Group

ADHD

Reading measures

Reading quotientMSD

Child Reading QuestionnaireMSD

Gray Oral Reading ageMSD

WRAT-R Spelling ageMSD

1.040.09

1.951.33

12.662.00

13.092.00

Conduct disorder measures

CBCL AggressionMSD

CBCL DelinquencyMSD

RBPC Socialized AggressionMSD

SRED IllegalMSD

SRED Norm ViolatingMSD

SRED TotalMSD

70.648.88

67.868.58

0.951.40

2.862.68

4.273.83

7.146.16

Control

1.100.16

0.550.60

13.952.73

13.071.92

55.952.80

55.550.91

0.320.72

2.142.88

3.143.48

5.276.18

P

ns

<.001

ns

ns

<.001

<.001

ns

ns

ns

ns

Note. ADHD = attention deficit hyperactivity disorder; WISC-R = Wechsler Intelligence Scale for Children—Revised; CBCL - AchenbachChild Behavior Checklist; DICA = Diagnostic Interview for Children and Adolescents; ADD = attention deficit disorder; HSQ = Home SituationsQuestionnaire; WRAT-R = Wide-Range Achievement Test—Revised; RBPC = Revised Behavior Problem Checklist; SRED = Self-Reported EarlyDelinquency Instrument.i Education level: 1 = graduate/professional degree; 2 = 4-year college degree; 3 = partial college training (including 2-year degree}; 4 = highschool degree; 5 = less than a high school degree.

The ADHD children were recruited through a local support group forparents of ADHD children and through two suburban school districtsin Denver, Colorado. The control children were recruited through aparticipant pool in the Psychology Department at the University ofDenver.

At the time of the study, all of the ADHD children were being treatedwith stimulant medication and were responding favorably to this medica-tion on the basis of parental report. Of the 22 ADHD boys, 20 weretaking Ritalin or methylphenidate (14 regular release; 6 sustained re-lease) and 1 each was taking sustained-release Dexedrine or dextroam-phetamine and sustained-release Cylert or pemoline. For the boys onRitalin or methylphenidate, both the mean and median daily dosageswere about 30 mg, with individual daily dosages ranging from 10 to100 mg. None of the children were taking concurrent psychotropic medi-cations (e.g., antidepressants) at the time of the study. Parents of theADHD children reported that the mean age of onset of their boys'ADHD symptoms was 2 years 6 months, but ranged from less than 1year 0 months to 6 years 11 months of age.

Research and Diagnostic Criteria

For this study, considerable attempts were made to reduce the method-ological problems associated with previous ADHD research by selectingas homogenous a group of ADHD boys as possible. Specifically, theboys were screened for comorbid psychiatric disorders, such as conductdisorder and reading disability, which have contributed to a conflictingpattern of findings across the ADHD literature (Barkley, 1990), as wellas for serious problems or conditions that otherwise could confound theresults (e.g., gross neurological damage, sensory-motor handicaps). Aninitial phone contact with parents of potential participants and a sessionconducted at the University of Denver in which information was obtainedregarding IQ, reading ability, family environment, ADHD symptomatol-ogy, and current stimulant medication usage facilitated the screening forsuch confounding factors. None of the boys who participated in thescreening session were excluded from participation because of low IQ(i.e., IQ < 80 on the Wechsler Intelligence Scale for Children—Revised[WISC-R]; Wechsler, 1974) or because of a severely chaotic, abusive,


or deprived family environment based on information provided on asupplemental parent questionnaire. Twenty-two potential ADHD partici-pants and 12 potential control participants were excluded for exhibitinga reading disability based on the calculation of a reading quotient (de-scribed below), 1 potential ADHD participant and 5 potential controlparticipants were excluded for not meeting the diagnostic criteria fortheir respective group (described below), and 1 ADHD boy and 25control boys who otherwise met the research criteria were excludedbecause they could not be matched with a boy in the alternate participantgroup.

The primary inclusionary or diagnostic criteria for the ADHD groupwere based on the Diagnostic and Statistical Manual of Mental Disor-ders (3rd ed., rev.; DSM-III-R; American Psychiatric Association[APA], 1987) although supplemental criteria were added to further re-strict our ADHD sample. These criteria are described in Table 2. Foreach of the criterion measures, parents rated their child's off-medicationbehavior. Restrictive cutoff scores were used for each measure to mini-mize the chances of including nonhyperactive boys in our ADHD group.All of the ADHD boys met or surpassed the criterion on each of thethree ADHD measures, with one exception; 1 boy was just below thecriterion on the Child Behavior Checklist (Achenbach & Edelbrock,1983) Hyperactivity scale (T score == 66) but was included in the ADHDgroup because he received a score of 15 on the Attention Deficit Disorderscale of the Diagnostic Interview for Children and Adolescents (Her-janic, 1983), which is very high.

In light of the recently published Diagnostic and Statistical Manualof Mental Disorders (4th ed.; DSM-IV; APA, 1994) criteria for ADHD,it was deemed important to attempt to describe our ADHD sample interms of the current classification system. Review of the responses onthe criterion measures suggests that 16 of our ADHD boys would meetthe DSM-IV criteria for both the inattentive and hyperactive-impulsivesubtypes, whereas 4 boys would meet the criteria for the inattentivesubtype only and 2 boys would meet the criteria for the hyperactive-impulsive subtype only.

Consistent with the DSM-II1-R criteria for ADHD, our sample wasdefined primarily by information provided from the boys' parents. Al-though information was not obtained from the boys' current classroomteachers for this study, the investigators did request information con-tained in their cumulative school records. The information that wasprovided, although not standard across children, was used to verify thatthe boys indeed experienced ADHD symptomatology at school duringkindergarten or first grade. Thus, together with the information providedby the parents, the information provided by the schools suggests thatthis sample of ADHD boys exhibits behavioral difficulties across situa-tions (i.e., both at home and at school).

Table 2Criteria for Inclusion in the Attention Deficit HyperactivityDisorder (ADHD) and Control Groups

Criterion measure ADHD Control

Child Behavior Checklist1

Hyperactive T ScoreDiagnostic Interview for Children

and Adolescents^Number of ADD symptoms

Home Situations Questionnaire0

% of situations

^70

^50

^60

<6

<33

Note. ADD = attention deficit disorder.a Achenbach and Edelbrock, 1983. b Herjanic, 1983. c Barkley andEdelbrock, 1987.

Additional Measures and Indexes

Several measures were included to assess the presence of readingdifficulties and the degree of conduct-disordered symptomatology in ourADHD sample. Two conduct-disorder measures based on parental reportwere administered: (a) the Socialized Aggression (SA) scale from theRevised Behavior Problem Checklist (RBPC; Quay & Peterson, 1987),which assesses rejection of authority and societal norms and (b) theCBCL Delinquency and Aggression scales, which assess more generalconduct problems. In addition, each boy was administered the Self-Report Early Delinquency Instrument (SRED; Moffitt & Silva, 1988)to assess "norm-violating" behaviors such as cheating and more seriousillegal behaviors such as carrying a weapon. Also, a two-subtest, short-form of the WISC-R (Wechsler, 1974), consisting of the Vocabularyand Block Design subtests, served as a screening device for identifyingboys with below-average IQs as well as a means for matching boys inthe two groups on general intellectual functioning.

Each of the measures included in the study has been found to haveadequate reliability and validity for research purposes (CBCL: Achen-bach & Edelbrock, 1983; DICA: Edelbrock & Costello, 1984; HomeSituations Questionnaire [HSQ]: Barkley, Fischer, Newby, & Breen,1988; RBPC: Quay & Peterson, 1987; SRED: Moffitt & Silva, 1988;WISC-R estimate: Wechsler, 1974).

A reading quotient (RQ) was calculated for each of the participatingboys to determine reading ability in this study. The RQ, as describedby Finucci, Isaacs, Whitehouse, and Childs (1982), represents the ratioof observed reading level to expected reading level. In the RQ formula,the observed reading level reflects the average of the age-equivalentscore obtained on the Spelling subtest from the Wide Range AchievementTest—Revised (WRAT-R; Jastak & Wilkinson, 1984) and the readingage score obtained on the Gray Oral Reading Test (Gray, 1963). Theexpected reading level reflects the average of the child's chronologicalage, mental age (IQ/chronological age), and age equivalent of the child'sgrade level. Only children with RQs greater than or equal to .90, whichis considered normal (Finucci et al., 1982), were included.

The Child Reading Questionnaire (CRQ; DeFries, Olson, Pen-nington, & Smith, 1991) was also included to facilitate the identificationof boys with reading difficulties. The CRQ assesses parental perceptionsof their child's past and current reading problems as well as behaviorsoften associated with dyslexia. Children with borderline RQs (i.e., inthe low .90s) and a history of reading problems, as reported on theCRQ, were excluded to reduce the chances of including remediateddyslexic children in our study.

Experimental Tasks

The Stopping Task, Antisaccade Task, Visual-Spatial Cuing Task, andTurning Task were programmed on an IBM-compatible computer withthe Micro Experimental Laboratory software (MEL; Schneider, 1988).Brief descriptions of the experimental tasks are presented here; moredetailed information may be obtained from Christine J. Aman, Ralph J.Roberts, Jr., and Bruce F. Pennington.

Frontal Tasks

Stopping Task. The Stopping Task is a computerized measure ofinhibitory control that is based on a formal theory of inhibition (Logan &Cowan, 1984; Logan, Cowan, & Davis, 1984). It is similar to the go/no-go tasks on which frontally lesioned adult humans (Drewe, 1975)and nonhuman primates (e.g., Mishkin, 1964) have been found to exhibitdeficits but is experimentally more rigorous (see Logan & Cowan, 1984;Logan et al., 1984). The Stopping Task may also reflect working memorybecause of the association between working memory and the inhibitionof prepotent responses (e.g., Roberts, Hager, & Heron, 1994). Schachar


and Logan (1990) found that ADHD children exhibit inhibitory controldeficits on the Stopping Task.

The Stopping Task consists of two types of trials: primary task trialsand stop-signal trials. On primary task trials, the letters X or O arepresented in the center of the video screen and the child responds bypressing the corresponding key on a keyboard. On stop-signal trials, theX or O is presented along with a tone, or stop signal, which instructsthe child to inhibit his key press to the primary task stimulus. The toneswere presented randomly at 500 ms, 350 ms, 250 ms, and 100 ms beforethe child's mean reaction time (MRT). The child sat in front of a videomonitor with his left and right index fingers on the two randomly labeledX and O keys. For each trial, a cross-shaped fixation point was presentedin the center of the screen (500 ms), followed by the letter for that trial(1,000 ms), and then a blank screen (1,500 ms).

The children completed six blocks of trials for this task. The firstblock, consisting of 32 primary task trials (16 each of X and O), helpedto familiarize the children with the task and yielded an MRT that wasused for determining the tone intervals for the second block of trials.The children were instructed to press the key as soon as they had decidedwhich letter was presented. The second block consisted of 48 practicetrials in which 16 stop-signal trials (4 trials at each of the 4 intervals)were randomly interspersed with 32 primary task trials. The remainingfour blocks of 48 trials constituted the experimental trials for the study.These blocks were equivalent to the second block and to each other withtwo exceptions: (a) the trials within each block were presented in differ-ent random orders and (b) the stop-signal intervals for each block variedaccording to the MRT for the primary task trials in the immediatelypreceding block. A total of 128 primary task trials and 64 stop-signaltrials were included.

The main dependent measure for this task is the probability of inhib-iting a response at each of the stop-signal intervals, with the slope ofthis inhibitory function of particular importance. Because ADHD chil-dren often exhibit long and highly variable reaction time latencies (e.g.,Douglas, 1988), it is important to separate deficiencies in primary taskprocesses from deficiencies in inhibitory processes on the Stopping Task.This can be done by computing the relative finishing time (RFT) of theinhibitory process (i.e., the internal reaction time to the stop signal) andprimary task process and converting this score into a Z score (ZRFT,as described by Logan et al., 1984). If the inhibition functions for thetwo groups cannot be aligned at the two sessions by plotting them as afunction of ZRFT. then the steeper functions are thought to reflect betterinhibitory control.

Antisaccade Task. The Antisaccade Task, which is based on a taskdescribed by Hallett and Adams (1980), is thought to measure inhibitorycontrol and working memory (Roberts et al., 1994) and has been foundto be sensitive to frontal lobe dysfunction in adult humans (Guitton,Buchtel, & Douglas, 1985). This task involves inhibiting a reflexiveresponse (a saccade in the direction of a cue) and the generation andinitiation of an alternative response (a saccade in the direction oppositefrom the cue). For each trial, a small cross-shaped fixation point (FP)was presented in the center of a video screen (500-3000 ms), followedby the cue (a 1° X 1° square) randomly presented 10° to the right or tothe left of the FP (100 ms). Approximately 400 ms after the onset ofthe cue, the larget appeared on the side opposite from the cue (150 ms)and then was masked. The target consisted of a three-sided .]/2° X '4°square surrounded by a square identical to the cue stimulus. Participantswere told not to look in the direction of the cue but rather an equaldistance from the FP in the direction opposite from the cue. Participantswere additionally instructed to indicate the open side of the target squareby pressing the corresponding left, right, or up arrow key on a keyboard.Several practice trials preceded the 42 experimental trials for this task.There were 7 trials in each of the six cells corresponding to the orthogo-nal combinations of the three gap locations (left, right, and top) and thetwo sides on which the cue appeared.

To control for potential confounding factors (e.g., ability to make eyemovements to a target), a Prosaccade Task, on which frontal patientshave not demonstrated deficits (Guitton et al., 1985), was also included.The Prosaccade Task was identical to the Antisaccade Task, with twoexceptions: (a) participants were instructed to look in the direction ofthe cue rather than in the opposite direction and (b) die cue-target onsetasynchrony in the Prosaccade Task was 300 ms rather than 400 ms. The42-trial Prosaccade Task was performed prior to the Antisaccade Taskto help scaffold the instructions and button press responses for themore difficult task. Each group at each of the two sessions had a meanproportion of correct prosaccades of at least .98.

During the saccade tasks, the participant sat in a darkened room withhis chin on a molded chin rest and his forehead leaning against a clothrestraint to help minimize head movements. The participant's eyes wereabout 62 cm from the video monitor. Eye movements were recordedwith a pupil-comeal reflection eye tracker (see Roberts. Brown,Wiebke, & Haith, 1991). Looking locations were recorded 60 times persecond and were accurate within %° of visual angle. The eye movementdata were scored by hand with reliable heuristics for determining whensaccades occur (Roberts et al., 1991). A brief calibration task wasadministered before the saccade tasks. For each trial of the saccadetasks, the latency and direction of the saccades were recorded. Theprimary dependent variable of interest was the accuracy of the saccades.

Tower of Hanoi. The Tower of Hanoi is a measure of planning abilityrequiring the formulation and execution of a strategy in accordance witha set of rules to achieve an externally imposed goal (e.g., Simon, 1975).Impaired performance on the Tower of Hanoi or a similar task has beenexhibited by frontally damaged adults (Shallice, 1982) and children(Levin et al., 1994) and, more recently, by ADHD boys compared withnormal and dyslexic controls (Pennington et al., 1993).

The materials for the Tower of Hanoi consist of two identical boards(one for the child and one for the experimenter), each holding threetapered vertical pegs arrayed in a straight line and three (or four) plasticrings of graduated sizes that fit on the pegs. There are three rules: (a)a larger ring may not be placed on top of a smaller ring, (b) only onering may be moved at a time, and (c) a ring has to be moved to one ofthe child's pegs {not placed on the table or on the experimenter's board).The experimenter's board displays the desired goal state, a 3- or 4-ringpyramidal configuration. The initial setup of the rings on the child'sboard (i.e., the start state) differs for each problem.

Five experimental problems were administered in order of increasingdifficulty. The difficulty of a problem is defined by the number of movesassociated with its optimal solution; problems requiring fewer moves tosolution are considered easier than problems requiring more moves. Asolution is correct if the goal state is replicated in the fewest numberof moves. To pass a problem, the child had to correctly solve the problemon two consecutive trials out of a maximum of six trials. Once a childpassed a problem, the next problem was administered. The task wasterminated when a child failed two consecutive problems. Points wereassigned per problem on the basis of the number of trials the childneeded to pass it (correct solutions on Trials 1 and 2 = 6 points; onTrials 2 and 3 = 5 points, etc.) as described by Borys, Spitz, and Dorans(1982). Failed problems received a score of 0. Summing the pointsacross the individual problems yielded a total score (maximum = 30points).

Parietal Tasks

Visual-Spatial Cuing Task. The Visual-Spatial Cuing Task is a mea-sure of visual-spatial attention used by several researchers on whichpatients with parietal lobe damage have demonstrated difficulty in "dis-engaging" their attention from invalid cues presented to the ipsilesionalvisual field (e.g., Baynes, Holtzman, & Volpe, 1986; Morrow & Ratcliff,1987; Posner, Inhoff, Friedrich, & Cohen, 1987). In this version of the


task, two boxes of equal size (1.7 cm X 1.9 cm) and a cross-shaped FP(3 mm X 3 mm) were displayed horizontally across a video screen,with the two peripheral boxes positioned approximately 7V2° from thecentral FP. Participants were instructed to maintain fixation on the FPthroughout the task. Each trial began when the outline of either one orboth of the peripheral boxes turned bright green, presumably elicitinga covert shift of attention to that location. This cue was followed ran-domly at intervals of 100 ms or 500 ms by a target stimulus (an asterisk)presented in the center of one of the two peripheral boxes. Participantsmade a single key press with their right hand as soon as they detectedthe target.

There were three types of trials: valid, invalid, and neutral. A validtrial consisted of the brightening of one of the peripheral boxes followedby the presentation of the target in the same box. An invalid trial wasone in which the target appeared in the box on the side opposite fromthe one that was brightened. On neutral trials, both of the peripheralboxes brightened and the target appeared in only one of the boxes.

The 240 experimental trials were administered in five blocks of 48trials. The breakdown of the types of trials within each block was basedon two visual fields (right, left), three cue types (valid, invalid, neutral)and two cue-target intervals (100 ms, 500 ms). For the visual field andcue-target interval variables, each of the associated levels occurred withequal frequency. For the cue type variable, valid cues occurred on twothirds of the trials, whereas invalid cues and neutral cues each occurredon one sixth of the trials. The main dependent measures for this taskwere the RTs associated with the various types of trials.

Turning Task. The Turning Task (Roberts & Aman, 1993) is thoughtto assess mental rotation, a process that has been shown to be sensitiveto the functioning of the right parietal lobe (e.g., Deutsch, Bourbon,Papanicolaou, & Eisenberg, 1988; Ratcliff, 1979). For this task, anisosceles triangle (1.2 cm wide at the base, 2.2 cm high) was displayedin the center of a video monitor. The triangle was displayed at one ofseven orientations, at 0° (facing straight up) and at rotated positions45°, 90°, and 135° to the left and right of 0°. A small dot was displayedapproximately 3.5 cm to the left or the right side of the base of thetriangle.

The children sat in front of a video monitor with their left and rightindex fingers resting, respectively, on left and right arrow keys. Thechildren were instructed to decide the shortest direction in which thetriangle should turn to point toward the dot and to press the correspond-ing arrow key as soon as they had made their decision. Prior to theexperimental trials, each child demonstrated his ability to reliably iden-tify the front, back, left, and right sides of the triangle at the varioustriangle orientations. The task consisted of 140 trials, with 10 left turnsand 10 right turns at each of the seven triangle orientations. For eachtrial, the computer recorded the child's RT and the accuracy of his turndecision, with the latter measure considered to be the primary dependentvariable.

Spatial Relations. Spatial Relations, a visual processing subtest fromthe Woodcock-Johnson Psycho-Educational Battery—Revised (Wood-cock & Johnson, 1989), was included as a second measure of mentalrotation. For each of the 33 test items, the child was presented with atarget figure and a series of potential component shapes, from which hehad to choose the two or three shapes that fit together to make the wholetarget shape. The dependent measure was the number of items solvedcorrectly.

Procedure

All children participated in a 1 -hr screening session during which theparents completed the research questionnaires and the children wereadministered the Edinburgh Handedness Inventory, the WISC-R Vocab-ulary and Block Design subtests, the WRAT-R Spelling subtest and theGray Oral Reading Test, and the SRED instrument. The ADHD boys

were tested off medication at this screening session because it wasthought that borderline reading-disabled boys would be more likely tofall below the cutoff when they were off medication, resulting in theirexclusion from the study. Also, it was thought that the IQ estimates forthe ADHD boys might be lower off medication and that matching thecontrol boys with the ADHD boys on their off medication IQ scoreswould result in a more conservative control group. For this session, themean number of hours since the ADHD children's last dose of medica-tion was 16.9 (SD = 20.62; Mdn = 9).

Children who met all of the criteria for the ADHD group or the controlgroup completed the series of experimental tasks twice, in 1- to 1.5-hrlong sessions approximately 1 week apart. The experimental tasks werepresented in the same fixed order for both of the experimental sessions(Antisaccade Task, Visual-Spatial Cuing Task, Spatial Relations, Stop-ping Task, Tower of Hanoi, and Turning Task). The ADHD boys wereon medication at the first experimental session and off medication at thesecond experimental session. For this study, a child was considered tobe ' 'off'' medication when he had not taken his medication for at least9 hr before the beginning of the testing session and "on" medicationwhen he had taken his medication approximately 1 hr before the begin-ning of the testing session. These time intervals reflect the findings thatthe sustained-release preparations of Ritalin or methylphenidate, Cylertor pemoline, and Dexedrine or dextroamphetamine have positive behav-ioral effects from approximately 1 to 9 hr after ingestion (Pelham et al.,1990), whereas the behavioral effects for the regular dose of Ritalin ormethylphenidate last from 1 to 4 hr after ingestion (Swanson, Kins-bourne, Roberts, & Zucker, 1978). The mean number of hours since theADHD children's last dose of medication was 1.27 hr (SD ~ .63; Mdn= 1) for Session 1 and 43.32 hr (SD = 35.20; Mdn = 24) for Session2. Six of the 22 ADHD boys did not take their medication for a fullweek prior to Session 2 because they were out of school for the summerand only took their medication regularly when school was in session.

Medication order was not counterbalanced across sessions because ofthe relatively small sample of children and the concern that counterbal-ancing the medication order might increase the variance of the resultsby introducing differences due to the drug order or its interactions. The"on" then "off " medication order was chosen because it was thoughtthat learning effects across the two sessions might minimize the expecteddecrease in performance off medication and thus reduce the chances ofdetecting differences between the medication conditions.

Results

Child and Family Characteristics

There were no significant differences between the ADHD andcontrol groups in terms of age, IQ estimate, Block Design orVocabulary standard scores, socioeconomic status (Hollings-head Index), mother's education, and father's education (seeTable 1).

Criterion and Related Measures

ADHD Measures

As expected, the ADHD group was rated higher than thecontrol group on the three ADHD criterion measures (see Table1).

Reading Measures

The two groups did not reliably differ in terms of RQ, GrayOral Reading age, or WRAT-R Spelling age; however, theywere reliably different on the number of problems indicated on


the CRQ, /(42) = 4.54, p < .001, with a greater mean numberof problems for the ADHD boys compared with the controls(see Table 1). It seems plausible that the higher score for theADHD group on the CRQ can be accounted for by the inclusionof ADHD-like symptoms involving inattention, distractibility,and fine motor control (e.g., problems remembering complexinstructions and keeping one's place on a page while reading,poor penmanship), rather than the presence of specific readingdifficulties, especially because the mean scores for both groupsare well below the clinical cutoff on the CRQ. Intercorrelationsbetween each of the ADHD criterion measures and CRQ scoressupport this hypothesis, all rs (42) > .50, ps < .001. In contrast,RQ, which is an objective measure of reading and spelling skills,did not correlate with any of the ADHD criterion measures.

Conduct Disorder Measures

Although the boys in the two groups were screened for overtbehavioral manifestations of delinquency (e.g., fire setting, re-current shoplifting) and previous conduct-disorder diagnoses,there were statistically reliable differences between the groupson two of the conduct-disorder measures (see Table 1). Specifi-cally, the ADHD boys received higher scores than the controlboys on the CBCL Aggression, /(42) = 7.40, p < .001, andDelinquency, f(42) = 6.70, p < .001, subscales. However, theADHD and control boys did not reliably differ on the RBPCSocialized Aggression scale, the SRED total score, or the SREDIllegal or Norm Violating scales. Furthermore, the mean scoresfor the two groups of boys in this study on these latter measuresare well below the mean scores for boys with known conductdisorder (RBPC: Quay & Peterson, 1987; SRED: Moffitt &Silva, 1988).

Experimental Tasks

For each of the experimental tasks, repeated measures multi-variate analyses of variance (MANOVAs) were conducted, withsession as a repeated measure (reflecting the ADHD boys' per-formances "on" vs. "off" medication) and group as a be-tween-subject variable. For analyses requiring the inclusion ofa second, multilevel repeated measure (e.g., the four stop-signalintervals for the Stopping Task) the Geisser-Greenhouse correc-tion was applied (Geisser & Greenhouse, 1958) to reduce theheightened Type I error rate.

Frontal Tasks

Stopping Task. The main dependent variables involve theprobability of inhibiting the button press to the presented letterat each of the stop-signal intervals and the slope of the inhibitionfunction. A MANOVA with session (two levels) and stop-signalinterval (four levels) as repeated measures and group as a be-tween-subject variable was conducted for the probability of inhi-bition given a stop signal. As shown in Figure 1, the probabilityof inhibiting increased with stop-signal interval for both of thegroups at both of the sessions, F(2, 90) = 291.42, p < .001.However, this effect was expressed differently for the ADHDand control groups and at the two sessions. These differenceswere shown by significant Group X Interval, F(2, 90) = 6.97,

p = .001, Session x Interval, F{3, 116) = 2.76, p = .05, andGroup x Session, F ( l , 42) = 7.11, p < .05, interactions. Thethree-way interaction involving group, session, and the linearcomponent of stop-signal interval was not statistically signifi-cant, F ( l , 116) - 3.70, ns. For the ADHD group, plannedunivariate comparisons revealed significantly worse ability toinhibit at Session 2 compared with Session 1 at the 250-ms,350-ms, and 500-ms intervals, all te(21) a 2.40, p < .05.Planned univariate comparisons for Session 2 showed that theADHD group performed significantly worse than the controlgroup at the 350-ms and 500-ms intervals, all /s(42) > 2.09,ps < .05; planned univariate comparisons for Session 1 yieldedno reliable differences between the two groups at any of thestop-signal intervals.

In general, the slope of the inhibition function across the fourstop-signal intervals for the unmedicated ADHD boys (i.e., atSession 2) was flatter relative to the slope for the control boysat both of the sessions and relative to their own slope when theywere medicated (at Session 1). This flatter inhibition functionis thought to reflect less efficient and slower inhibitory processesin the unmedicated ADHD boys (see Logan et al., 1984). Thegroup main effect was not significant, F ( l , 42) = 1.19, p —.28; however, the session main effect was significant, F ( l , 42)= 5.54, p < .05, indicating that performance was worse atSession 2 than at Session 1 because of the poorer performanceof the unmedicated ADHD boys.

The inhibition functions were plotted as a function of ZRFT(relative finishing times of the inhibitory process and the pri-mary task process converted to Z scores) to determine whetherthe flatter slope exhibited by the unmedicated ADHD group wasdue to an inefficient or a slow inhibitory mechanism rather thanto variability associated with primary task processes or withinternal reaction time. These functions, which reflect adjust-ments for differences in primary task variability and internalreaction time to the stop signal, are shown at the bottom ofFigure 1. As can be seen, the ZRFT correction does not com-pletely eliminate the differences between the inhibitory func-tions of the ADHD boys and the controls at Session 2. Thedifferences in the inhibitory functions remained because theADHD boys, while off medication, inhibited their responseswithin a smaller range of reaction times compared with thecontrols. A test of the linear component of the three-way interac-tion involving group, session, and stop-signal interval was sig-nificant, F ( l , 42) = 9.66, p < .01, indicating that the inhibitionfunctions of the ADHD boys were not as steep as those of thecontrol boys, but only when the ADHD boys were off medica-tion (Session 2) .

Antisaccade Task. For the Antisaccade Task, the variable ofparticular interest was the accuracy of the eye movements, orsaccades. Preliminary analyses revealed that the side on whichthe cue was presented was not associated with any significantmain effects or interactions, thus data were collapsed across thisvariable for the analyses reported below. The only repeatedmeasure included in the MANO\5\ for this task was session.

The saccade accuracy for the Antisaccade Task was quite lowfor both groups. The mean proportion of correct antisaccadesincreased for the control boys from .32 (SD = .22) to .43 {SD= .25) across the two sessions, whereas it decreased for theADHD boys from .38 (SD = .27) to .33 {SD = .20). The


Session 1 Session 2

1

0.9c 0.8oI 0.7I 0.6"5 0.5S 0.4| 0.3£ 0.2

0.10 100 250 350 500

Stop-Signal Interval (in ms)

1

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

-ADHD

-Control

100 250 350 500

Stop-Signal Interval (in ms)

1

0.9c 0.8I 0.7| 0.6"S 0.5{0.4| 0.3£ 0.2

0.1

0-0.5 0.5

ZRFT

1.5

1

0.90.80.70.60.50.40.30.20.1

0-0.5 0.5

ZRFT

1.5

Figure 1. (Top) Probability of inhibition on the Stopping Task as a function of session, group, and stop-signal interval. (Bottom) Probability of inhibition on the Stopping Task as a function of session, group,and Z score of relative finishing times (ZRFT) of the inhibitory and the primary task processes. ADHD =attention deficit hyperactivity disorder.

Group X Session interaction was statistically reliable, F( 1, 40)= 11.86, p < .001. Planned univariate analyses for the Antisac-cade Task indicated that the control group's performance sig-nificantly improved from Session 1 to Session 2, t(l9) = 3.02,p < .01, whereas the ADHD group's performance stayed thesame, ((21) = 1.78, ns. The ADHD and the control groups didnot reliably differ from each other in terms of mean proportionof correct antisaccades at either of the sessions.

Other researchers who have studied the performance of chil-dren on the Antisaccade Task have found better overall abilityto suppress reflexive saccades than that obtained in this study(e.g., Rothlind, Posner, & Schaughency, 1991). One major dif-ference between our version of the task and those used in otherstudies may explain the discrepant findings. In the present study,the fixation point was extinguished before the cue was presentedrather than left on throughout the trial, as in other studies. Ourparticipants may have found it more difficult to refrain frommaking reflexive saccades in the direction of the cue becausethey did not have the fixation point on which to maintain theirgaze. In fact, Rothlind et al. found that the ADHD and thecontrol participants made at least 10% more reflexive saccades

when the fixation point was turned off than when it remainedon. Different versions of the Antisaccade Task consist of varia-tions in terms of scoring of saccades, including or excluding asecondary task (e.g., indicating the direction of the open sideof the target stimulus), and including or excluding "visuallytriggered'' saccades, all of which could contribute to dissimilarfindings regarding accuracy that have been observed acrossstudies.

Tower of Hanoi. The dependent measure for the Tbwer ofHanoi consisted of the total score (i.e., the sum of the scoresfor the individual three- and four-ring problems). The meantotal scores for the control boys were 19.6 (SD = 4.4) and 26.7(SD = 3.5) for Session 1 and Session 2, respectively; meanscores for the ADHD boys were 17.3 (SD = 4.5) and 20.9 (SD= 5.3) for Session 1 and Session 2, respectively. The controlboys exhibited better overall performance than the ADHD boys,F(1 , 42) = 12.10, p = .001, especially at Session 2, whenthe ADHD boys were off medication. Tn addition, both groupsdemonstrated improvement from Session 1 to Session 2, F ( l ,42) = 62.97, p < .001, but greater improvement was noted forthe controls across the two sessions, F( 1, 42) = 6.76, p < .05.

964 AMAN, ROBERTS, AND PENMNGTON

The primary type of error made by the ADHD boys on this taskinvolved impulsive responding or failure to think through theirplan before responding.

Parietal TasksVisual-Spatial Cuing Task. According to Posner (1988), pa-

rietal patients show a larger RT difference between validly andinvalidly cued trials presented to the ipsilesional visual fieldrelative to the contralesional visual field on the Visual-SpatialCuing Task. The primary dependent variable is the score re-flecting the difference in median response times on validly andinvalidly cued trials (i.e., the validity effect). Trials on whicha neutral cue was presented were, thus, not considered in thefollowing analyses.

MANOVAs conducted on the validity effect included delay(100 ms vs. 500 ms), target side (left vs. right) and session asrepeated measures and group as a between-subject variable. Theonly main effect that emerged involved session, F(l, 42) =4.46, /; < .05, and indicated that the validity effect was smallerat Session 2 than at Session 1. There were no other significantmain effects or interactions associated with this task. Of particu-lar note is that none of the interactions involving the groupvariable were significant, indicating that the groups did not reli-ably differ in terms of the validity effect on this task. There alsowere no interactions or main effects involving target side. Theseresults do not reflect the pattern of performance that Posner(1988) found for parietal patients; however, both groups ofparticipants did exhibit the "normal" pattern of results, withreaction times being longer for the invalidly cued trials com-pared with the validly cued trials.

Turning Task. Accuracy data for the Turning Task were col-lapsed across the side on which the cue occurred (i.e., thedirection in which the triangle should turn) as well as the presen-tation side of the triangle (i.e., rotations reflecting 45C, 90°,and 135° to the left or right of 0°), as these variables werecounterbalanced within participants and were not of specificinterest. The average number of correct responses for the homol-ogous left and right orientations were computed to equate thenumber of trials associated with each of the non-0° orientations(n = 40) with the number of trials associated with 0° (n - 20).

A repeated measures MANOVA, with orientation (four lev-els) and session as the repeated measures and group as a be-tween-subjects variable, was used to examine response accu-racy. Figure 2 shows that accuracy for both groups decreasedas orientation from 0° increased, F(2, 65) = 43.78, p < .001.In general, the control boys were more accurate than the ADHDboys, F( 1, 42) = 3.98, p — .05, but particularly at the orienta-tions further from 0°, as suggested by the significant Group xOrientation interaction, F{2, 65) = 3.45, p < .05. The GroupX Session x Orientation interaction was also significant, F(2,77) = 3.63, p < .05, indicating that the Group x Orientationinteraction differed at the two experimental sessions. This wasparticularly true for the control group, who demonstrated a moremarked drop-off in accuracy at the 135° orientation at Session1 than they did at Session 2. In summary, both groups performedat ceiling levels at 0° but not at orientations away from 0°, inwhich the groups were expected to perform dissimilarly. Also,both groups evinced improvement over the two sessions, butthis was more pronounced for the controls.

In terms of the response latencies on the Turning Task, maineffects for Orientation, F(1,61) = 65.17, p < .001, and Session,F(l, 42) = 45.98, p < .001, were statistically reliable, as wasthe Session X Orientation interaction, F(2, 87) = 9.57, p <.001. These results indicate that the response times for bothgroups increased as the triangle's orientation moved fartheraway from 0°, the response times were longer at the first sessionthat at the second session, and the increase in response timeswith increasing orientation was more dramatic at the first sessionthan at the second session. The ADHD and control groups didnot reliably differ in terms of response latencies, F ( l , 42) =0.52, and none of the interactions involving the group factorwere statistically reliable.

Spatial Relations. The dependent variable associated withSpatial Relations was the number of problems solved correctly.The mean number of correctly solved problems was 23.7 (SD= 3.1) and 24.0 (SD = 4.5) for the ADHD group at Sessions1 and 2, respectively, and 22.0 (SD = 4.1) and 25.0 (SD = 3.4)for the control group at Sessions 1 and 2, respectively. Therepeated measures MANOVA revealed a significant Group xSession interaction, F ( l , 42) = 5.82, p < .05, indicating thatthe control group evinced improvement on this measure at thesecond session compared to the first session, whereas the ADHDgroup did not. The session main effect was also significant,F ( l , 4 2 ) = 8.28, p < .01, with overall performance being betterat Session 2 than at Session 1.

Discussion

A neuropsychological approach was used in this study toexamine the frontal lobe and the right parietal lobe theories ofADHD. Several questions related to the involvement of thesetwo brain areas in ADHD were addressed.

The main question of interest was whether the frontal lobetheory or the right parietal lobe theory of ADHD would besupported. Generally, we found support for both theories, as theADHD boys demonstrated impairment when their off medica-tion performance was compared to their on medication perfor-mance and when their improvement across the two sessionswas compared to that of the control boys. Specifically, on thepurported tasks of frontal lobe functioning, the ADHD boysevinced less learning across the two sessions than the controlboys in terms of inhibiting reflexive saccades toward the cueon the Antisaccade Task and thinking through a plan prior toexecuting the plan on the Tower of Hanoi. The ADHD boys alsodemonstrated poorer performance off medication (Session 2)than they did on medication (Session 1) in terms of inhibitingbutton-press responses when a tone was presented on the Stop-ping Task.

On the purported tasks of right parietal lobe functioning,the ADHD boys showed less learning across the two sessionscompared with the controls in terms of mental rotation of objectsand shapes on the Turning Task and Spatial Relations. Therewere no apparent meaningful differences between the twogroups on the Visual-Spatial Cuing Task, and neither of thegroups produced patterns of performance resembling those ofparietal patients on this task. These latter findings are convergentwith those of Swanson et al. (1991), who also failed to find the"parietal" pattern of performance in ADHD children as well


2019181716151413121110

Session 1

fc=:::::t;—^x\

•

•

-

•

, , , ,

0 45 90 135

Orientation (in degreesi

20

19181716

151413

12Ui n

Session 2 - • - A D H D- o — Contrtf

*- - * ^ " ~ — ° ^ ^

0 45 90 135

Orientation (in degrees)

Figure 2. Number of correct turns on the Turning Task as a function of session, group, and orientation ofthe triangle. ADHD = attention deficit hyperactivity disorder.

as unequivocal differences between ADHD and control childrenon the Visual-Spatial Cuing Task.

There are several alternative interpretations or explanationsfor the findings across the two task domains included in thisstudy. One obvious interpretation is that the observed differencesbetween the ADHD and the control children on these taskssupport both the frontal and right parietal theories of ADHD.Assuming that these tasks do reflect the respective functioningof either the frontal or the parietal lobes, then deficient perfor-mance on them would imply dysfunction involving these brainregions. The failure to find differences between the groups onthe Visual-Spatial Cuing Task does not pose a problem for thishypothesis. It could simply mean that this task is only sensitiveto severe parietal lobe damage, such as would be associatedwith ablations or lesions, and that the right parietal dysfunctionin ADHD is much more subtle.

A second, less obvious, interpretation regarding the perfor-mance deficits and flatter learning effects demonstrated by theADHD children is that they are due to dysfunction of the frontallobes only rather than to dysfunction of both the frontal and theright parietal lobes. Support for this notion comes from researchindicating that frontal lobe patients often exhibit global cognitiveand behavioral deficits, including deficits that are typically as-cribed to right parietal lobe dysfunction or damage (e.g., Fuster,1985). Given the vast interconnections between areas of theprefrontal cortex and the posterior parietal cortex, it seems likelythat damage to the frontal lobes could affect the functioning ofthe parietal lobes. Along the same lines, the ADHD group'spoorer performance could be due to some other general or globalneurological deficit that affects both frontal and parietalfunctioning.

In addition, it may also be the case that the functioning ofother specific brain systems that were not examined in this studycould explain the deficient performance demonstrated by theADHD boys. For example, hippocampal cells in rats have beentied to behavioral inhibition as measured by go/no-go tasks(e.g., Sakurai, 1990), suggesting that the specificity of the Stop-ping Task as a measure of frontal lobe functioning is question-able. However, the Stopping Task does appear to be a fairlywell-validated measure of frontal lobe functioning, as other neu-

robiological evidence is lacking in terms of linking dysfunctionof the hippocampus to ADHD symptomatology.

Two assumptions made by the authors of the present studyare important to keep in mind when interpreting the results.First, the frontal versus parietal classification of the tasks wasassumed on the basis of evidence from previous research or onthe basis of the similarity of the tasks to other tasks that havebeen linked to either the functioning of the frontal lobes or theright parietal lobe. Second, it was assumed that each of theexperimental tasks is only sensitive to the functioning of one ofthese brain areas and not to both of them. Unfortunately, thespecificity of these tasks has not been explicitly tested in termsof frontal dysfunction versus parietal dysfunction. For example,Guitton et al. (1985) found that frontal patients demonstratedimpaired ability to inhibit reflexive saccades toward the cue sideon the Antisaccade Task but did not examine the performanceof parietal patients. The controls in their study consisted ofnonlesioned participants and patients who had sustained tempo-ral lobe lesions. Similarly, Drewe (1975) found impaired perfor-mance in frontal patients on go/no-go tasks, tasks on which theStopping Task is based, but compared the performances of fron-tal patients with those of nonfrontal lesioned patients, most ofwhom had temporal lobe rather than parietal lobe lesions. Inaddition, in Ditunno and Mann's study (1990), right parietalpatients were more severely deficient on mental rotation tasksthan were left parietal patients; however, they did not examinethe performance of frontal patients on the same mental rotationtasks. Given these two caveats regarding the assumptions madeby the authors of the present study, caution is advised wheninterpreting the results from the individual tasks included in thisstudy.

At least two explanations that are not specifically related tofrontal lobe or right parietal lobe dysfunction seem unable toaccount for the deficits exhibited by our ADHD group. Oneexplanation suggests that the ADHD children would performpoorly on all tasks, not just those that are linked to the function-ing of the frontal or the right parietal lobes. This explanationdoes not fit because the performance of the two groups did notreliably differ on all aspects of the tasks (e.g., there were noreliable differences between the groups at either Session 1 or


Session 2 on Spatial Relations). Furthermore, previous researchhas indicated that there are several cognitive tasks on whichADHD children do not demonstrate deficiencies, including thoseinvolving verbal memory and nonverbal memory (see Douglas,1988, for a review). There clearly are better ways to examinethe notion that the ADHD group would have performed poorlyon all tasks; unfortunately, such tests were not explicitly builtinto this study.

A second, related explanation that does not appear to be ableto account for our results involves a generalized motivationaldeficit in the ADHD group. Again, this explanation would besupported if the ADHD children demonstrated poor perfor-mance across all aspects of the experimental tasks. However,even though there were differences between the two groupson several of the experimental tasks, with the control groupperforming better than the ADHD group, the ADHD group wasoften very accurate. For example, the unmedicated ADHD chil-dren were very accurate, even at orientations up to 135° from0° on the Turning Task. This result would not be expected if theADHD children were unmotivated or were not trying to do wellon the tasks. Anecdotal evidence also suggests that the ADHDboys were trying to perform well, in that they frequently ex-pressed their disappointment after making an incorrect response(e.g., when failing to inhibit their button press response after atone was presented on the Stopping Task), and they often cor-rected themselves on Spatial Relations after providing an initialresponse and then changing their minds. Thus, it seems unlikelythat motivational deficits in the ADHD children can sufficientlyexplain the differences between the two groups on the experi-mental tasks.

In summary, the poorer performance of the ADHD group inthe present study can be interpreted as supporting both the fron-tal lobe and parietal lobe theories of ADHD, as supporting onlythe frontal lobe theory of ADHD, or as supporting a global braindamage theory of ADHD. It appears that the poorer performanceof the ADHD children cannot be explained in terms of eithergeneralized performance deficits or motivational deficits.

A second question addressed in this study involved the rela-tive severity of the deficits associated with the frontal and theparietal tasks. In general, the ADHD boys appeared to be moreseverely impaired on the frontal tasks than on the parietal tasks,although only moderately so. On the frontal tasks, the perfor-mance of the ADHD boys across the two sessions declined onthe Stopping Task and stayed the same on the Antisaccadc Task.Also, the ADHD boys demonstrated flatter learning effects thandid the controls on the Tower of Hanoi. On the parietal tasks,specifically the Turning Task and Spatial Relations, the ADHDboys tended to show nominal improvement across the sessions,reliably less than the controls.

Nevertheless, there are two reasons why these results shouldbe interpreted with caution. First, the ADHD group as well asthe control group performed very well on both the Turning Taskand Spatial Relations. It is unclear whether the relative severityof the frontal and the parietal deficits would be the same if moredifficult measures of parietal functioning had been included.Second, the two groups were matched on an estimate of 1Q,which was based on only two subtests from the WISC-R, Vo-cabulary and Block Design. Although the IQ estimate and Vo-cabulary standard scores were found to be unrelated to the

dependent measures from the experimental tasks, Block Designstandard scores were positively correlated with scores on SpatialRelations and the Turning Task. Even though the ADHD andcontrol groups did not reliably differ in terms of scores on BlockDesign, the ADHD group did exhibit poorer performance onthese two parietal tasks compared with the controls. Thus, it isunclear whether the ADHD children are truly more adverselyaffected on the frontal tasks compared with the parietal tasks orwhether the results obtained in this study regarding the relativeseverity of the deficits on the two types of tasks were an artifactof having matched the groups of boys on IQ, which was heavilyinfluenced by scores on Block Design.

Finally, the third research question addressed in this studywas whether stimulant medication would differentially affectperformance on the frontal and the parietal tasks. As notedpreviously, differences between the ADHD and control groupson both parietal and frontal tasks were more frequently foundat Session 2, when the ADHD children were off medication,compared to Session 1, when they were on medication. Thus,stimulant medication was associated with better performance ontasks from both the frontal and the right parietal domains.

In general, the medication findings for this study should beviewed with caution, however, as a fixed rather than randommedication order was used, confounding the medication condi-tion with the order of the experimental sessions. Consequently,it is important to consider alternative explanations that are inde-pendent of the medication condition for observed differenceswithin the ADHD group at the two sessions as well as differ-ences between the ADHD group and the control group that areevident at one of the sessions but not at the other. One alternativeexplanation for these differences, ignoring the medication condi-tion per se, involves the relative novelty or unfamiliarity of theexperimental tasks, as well as the testing environment in general,for the ADHD children at the first and second sessions. Supportfor this interpretation comes from research indicating thatADHD children tend to exhibit relatively mild behavioral symp-toms in novel or unfamiliar settings, or in situations involvingnovel tasks, that become increasingly more deviant as familiaritywith the setting increases (e.g., Zentall, 1985). Thus, the decre-ments in performance across the two sessions for the ADHDgroup or the smaller learning effects shown across the two ses-sions by the ADHD group compared to the control group canbe explained in terms of the experimental tasks being less noveland more familiar to die ADHD children at Session 2 than atSession 1. This explanation cannot be ruled out.

One interpretation or explanation that seems unlikely for themedication findings in this study involves the rebound effectsthat are associated with stimulant use. Rebound effects reflectthe deterioration in behavior, in excess of that expected frombaseline, that occurs as the beneficial effects of the stimulantmedication wear off. These effects are usually observed between5 and 10 hr following administration of the medication. If re-bound effects were a major factor in this study, the off medica-tion deficits observed in the ADHD children would be exagger-ated because they would reflect sub-baseline performance ratherthan baseline performance or ability. There are two reasons whythe rebound interpretation does not appear to be relevant in ourstudy. First, the median number of hours since the children hadlast taken their medication was 24, which is well beyond the time


during which rebound effects are typically observed. Second, ofthe two ADHD children for whom stimulant medication wasadministered within 10 hr of Session 2, 1 child did not performbelow average on any of the experimental tasks and the otherperformed no worse than the other ADHD children who hadnot taken their medication for more than 10 hr before performingthe tasks. Furthermore, recent findings by Johnston and her col-leagues (Johnston, Pelham, Hoza, & Sturges, 1988) suggestedthat rebound effects are not that widespread (occurring in onlyabout one third of ADHD children who take methylphenidate),and even when they do occur, their magnitude varies consider-ably from day to day.

Conclusion

The primary goal of this study was to examine the frontaland right parietal lobe theories of ADHD. Although our resultssupport both of these theories, the frontal lobe theory of ADHDwas more strongly supported than the right parietal lobe theoryof ADHD. Perhaps the best conclusion to draw from this studyis that it may be more important to determine why there aredecrements in performance in both task domains and how thesedecrements relate to brain functioning rather than to try to ex-plain the underlying deficit in ADHD in terms of either thefrontal lobe theory or the parietal lobe theory.

References

Achenbach, T. M., & Edelbrock, C. S. (1983). Manual for the ChildBehavior Checklist and Revised Behavior Profile. Burlington: Univer-sity of Vermont, Department of Psychiatry.

American Psychiatric Association. (1987). Diagnostic and statisticalmanual of mental disorders (3rd ed., rev.). Washington, DC: Author.

American Psychiatric Association. (1994). Diagnostic and statisticalmanual of mental disorders (4th ed.). Washington, DC: Author.

Andersen, R. A. (1988). The neurobiological basis of spatial cognition:Role of the parietal lobe. In J. Stiles-Davis, M. Kritchevsky, & U.Bellugi (Eds.), Spatial cognition: Brain bases and development (pp.57-80). Hillsdale, NJ: Erlbaum.

Barkley, R. A. (1990). Attention deficit hyperactivity disorder: A hand-book for diagnosis and treatment. New "tork: Guilford Press.

Barkley, R. A., & Edelbrock, C. S. (1987). Assessing situational varia-tion in children's behavior problems: The Home and School SituationsQuestionnaires. In R. Prinz (Ed.), Advances in behavioral assessmentof children and families (Vol. 3, pp. 157-176). Greenwich, CV. JAIPress.

Barkley, R. A., Fischer, M., Newby, R., & Breen, M. (1988). Develop-ment of a multimethod clinical protocol for assessing stimulant drugresponses in ADHD children. Journal of Clinical Child Psychology,17, 14-24.

Baynes, K., Holtzman, J. D., & Volpe, B. T. (1986). Components ofvisual attention: Alterations in response pattern to visual stimuli fol-lowing parietal lobe infarction. Brain, 109, 99-114.

Borys, S. V, Spitz, H. H., & Dorans, B. A. (1982). Tower of Hanoiperformance of retarded young adults and nonretarded children as afunction of solution length and goal state. Journal of ExperimentalChild Psychology, 33, 87-110.

Boucugnani, L. L., & Jones, R. W. (1989). Behaviors analogous tofrontal lobe dysfunction in children with attention deficit hyperactivitydisorder. Archives of Clinical Neuropsychology, 4, 161-173.

Breen, M. J. (1989). Cognitive and behavioral differences in ADHD

boys and girls. Journal of Child Psychology and Psychiatry, 30, 711 —716.

Chelune, G. J., Ferguson, W., Koon, R., & Dickey, T O. (1986). Frontallobe disinhibition in attention deficit disorder. Child Psychiatry andHuman Development, 16, 221—234.

DeFries, J. C , Olson, R. K., Pennington, B. F., & Smith, S. D. (1991).Colorado Reading Project: Past, present, and future. Learning Disabil-ities, 2, 37-46.

Deutsch, G., Bourbon, W. T, Papanicolaou, A. C , & Eisenberg, H. M.(1988). Visuospatial tasks compared via activation of regional cere-bral blood flow. Neuropsychologia, 26, 445-452.

Ditunno, P. L., & Mann, V. A. (1990). Right hemisphere specializationfor mental rotation in normals and brain damaged subjects. Cortex,26, 177-188.

Douglas, V. I. (1988). Cognitive deficits in children with attention deficitdisorder with hyperactivity. In L. M. Bloomingdale & J. Sergeant(Eds.), Attention deficit disorder: Criteria, cognition, and intervention(pp. 65-81). Oxford, England: Pergamon Press.

Drewe, E. A. (1975). Go-no-go learning after frontal lobe lesions inhumans. Cortex, 11, 8-16.

Edelbrock, C. S., & Costello, A. (1984). Structured psychiatric inter-views for children and adolescents. In G. Goldstein & M. Hersen(Eds.), Handbook of psychological assessment (pp. 276-290). New"fork: Pergamon Press.

Finucci, J. M., Isaacs, S. D., Whitehouse, C. C , & Childs, B. (1982).Empirical validation of reading and spelling quotients. DevelopmentalMedicine in Child Neurology, 24, 733-744.

Fuster, J. M. (1985). The prefrontal cortex, mediator of cross-temporalcontingencies. Human Neurobiology, 4, 169-179.

Geisser, S., & Greenhouse, S. W. (1958). An extension of Box's resultson the use of the F distribution in multivariate analysis. Annals ofMathematical Statistics, 29, 885-891.

Goldman-Rakic, P. S. (1987). Circuitry of primate prefrontal cortexand regulation of behavior by representational memory Handbook ofphysiology, Section 1: The nervous system: Vol. V. Higher functionsof the brain (pp. 373-417). Bethesda, MD: American PhysiologicalSociety.

Gorenstein, E. E., Mammato, C. A., & Sandy, J. M. (1989). Performanceof inattentive-overactive children on selected measures of prefrontal-type function. Journal of Clinical Psychology, 45, 619-632.

Gray, W. S. (1963). Gray Oral Reading Tests, Indianapolis, IN: Bobbs-Merrill.

Grodzinsky, G. M., & Diamond, R. (1992). Frontal lobe functioningin boys with attention-deficit hyperactivity disorder. DevelopmentalNeuropsychology, 8, 427-445.

Guitton, D., Buchtel, H.A., & Douglas, R.M. (1985). Frontal lobelesions in man cause difficulties in suppressing reflexive glances andin generating goal-directed saccades. Experimental Brain Research,58, 455-472.

Hallett, P. E., & Adams. B. D. (1980). The predictability of saccadiclatency in a novel voluntary oculomotor task. Vision Research, 20,329-339.

Harcherik, D. R, Cohen, D.J., Ort, S., Paul, R., Shaywitz, B. A.,Volkmar, F. R., Rothman, S. L. G., & Leckman, J. F. (1985). Com-puted tomographic brain scanning in four neuropsychiatric disordersof childhood. American Journal of Psychiatry, 142, 731-737.

Heilman, K. M., & Van Den Abell, T. (1980). Right hemisphere domi-nance for attention: The mechanism underlying hemispheric asymmet-ries of inattention (neglect). Neurology, 30, 327-330.

Herjanic, B. (1983). The Washington University Diagnostic Interviewfor Children and Adolescents. St. Louis, MO: Washington UniversityMedical Center.

Hollingshead, A. B. (1957). Two Factor Index of Social Position. NewHaven, CT: Yale University Press.


Hynd, G. W., Hern, K. L., Novey, E. S., Eliopulos, D., Marshall, R.,Gonzalez, J. J., & Voeller, K. K. (1993). Attention deficit hyperactiv-ity disorder and asymmetry of the caudate nucleus. Journal of ChildNeurology, 8, 339-347.

Hynd, G. W., Semrud-Clikeman, M., Lorys, A. R., Novey, E. S., & Elio-pulos, D. (1990). Brain morphology in developmental dyslexia andattention deficit disorder/hyperactivity. Archives of Neurology, 47,919-926.

Hynd, G. W., Semrud-Clikeman, M , Lorys, A. R., Novey, E. S., Elio-pulos, D., & Lyytinen, H. (1991). Corpus callosum morphology inattention deficit hyperactivity disorder: Morphometric analysis ofMRI. Journal of Learning Disabilities, 24, 141-146.

Jastak, J. E, & Wilkinson, G. S. (1984). Wide Range AchievementTest—Revised. Wilmington, DE: Jastak Associates.

Johnston, C , Pelham, W. E., Hoza, I , & Sturges, J. (1988). Psychostim-ulant rebound in attention deficit disordered boys. Journal of theAmerican Academy of Child and Adolescent Psychiatry, 27, 806-810.

Levin, H. S., Mendelsohn, D., Lilly, M. A., Fletcher, J. M., Culhane,K. A,, Chapman, S. B., Harward, H., Kusnerik, L., Bruce, D., &Eisenberg, H. M. (1994). Tower of London performance in relation tomagnetic resonance imaging following closed head injury in children.Neuropsychoiogy, 8, 171-179.

Levin, P.M. (1938). Restlessness in children. Archives of Neurologyand Psychiatry, 39, 764-770.

Logan, G. D., & Cowan, W. B. (1984). On the ability to inhibit thoughtand action: A theory of an act of control. Psychological Review, 91,295-327.

Logan, G. D., Cowan, W. B., & Davis, K. A. (1984). On the ability toinhibit simple and choice reaction time responses: A model and amethod. Journal of Experimental Psychology: Human Perception andPerformance, 10, 276-291.

Lou, H. C , Henriksen, L., & Bruhn, P. (1984). Focal cerebral hypoper-fusion in children with dysphasia and/or attention deficit disorder.Archives of Neurology, 41, 825-829.

Lou, H. C , Henriksen, L., Bruhn, P., Boraer, H,, & Nielsen, J. B. (1989).Striatal dysfunction in attention deficit and hyperkinetic disorder. Ar-chives of Neurology, 46, 48-52.

Mesulam, M. M. (1981). A cortical network for directed attention andunilateral neglect. Annals of Neurology, 10, 309-325.

Mishkin, M. (1964). Perseveration of central sets after frontal lesionsin monkeys. In J. M. Warren & K. Akert (Eds.), The frontal granularcortex and behavior. New York: McGraw-Hill.

Moffitt, T. E., & Silva. P. A. (1988). Self-reported delinquency: Resultsfrom an instrument for New Zealand. Journal of Criminology, 21,227-240.

Morrow, L. A., & Ratcliff, G. (1987). Attentional mechanisms in clini-cal neglect [Abstract]. Journal of Clinical and Experimental Neuro-psychoiogy, 9.

Oldfield, R. C. (1971). The assessment and analysis of handedness.Neuropsychologia, 9, 97-113.

Pelham, W. E., Greenslade, K. E., Vodde-Hamilton, M., Murphy, D. A.,Greenstein, J. J., Gnagy, E. M., Guthrie, K. J., Hoover, M. D., & Dahl,R. E. (1990). Relative efficacy of long-acting stimulants on childrenwith attention deficit-hyperactivity disorder: A comparison of standardmethylphenidate, sustained-release methylphenidate, sustained-releasedextroamphetamine, and pemoline. Pediatrics, 86, 226-237.

Pennington, B. E, Groisser, D. B., & Welsh, M. C. (1993). Contrastingcognitive deficits in attention deficit hyperactivity disorder versusreading disability. Developmental Psychology, 29, 511-523.

Piercy, M., Hecaen, H., & Ajuriaguerra, J. (1960). Constructionalapraxia associated with unilateral cerebral lesions—left and rightsided cases compared. Brain, 83, 225-242.

Posner, M. I. (1988). Structures and functions of selective attention. In

T Boll & B. Bryant (Eds.), Clinical neuropsychoiogy and brainfunction: Research, measurement, and practice (pp. 169-202).Washington, DC: American Psychological Association.

Posner, M. I., Tnhoff, A. W, Friedrich, F. J., & Cohen, A. (1987). Isolat-ing attentional systems: A cognitive-anatomical analysis. Psychobiol-ogy, 15, 107-121.

Quay,H. C.,& Peterson, D. R. (1987). Manual for tht Revised BehaviorProblem Checklist. Coral Gables, FL: Authors (University of Miami).

Ratcliff, G. (1979). Spatial thought, mental rotation and the right hemi-sphere. Neuropsychologia, 17, 49-54.

Roberts, R. J., Jr., & Aman, C. J. (1993). Developmental differences ingiving directions: Spatial frames of reference and mental rotation.Child Development, 64, 1258-1270.

Roberts, R. J., Jr., Brown, D., Wiebke, S., & Haith, M. M. (1991).A computer-automated laboratory for studying complex perception-action skills. Behavior Research Methods, Instruments & Computers,23, 493-504.

Roberts, R. J., Jr., Hager, L. D., & Heron, C. (1994). Prefrontal cognitiveprocesses: Working memory and inhibition in the antisaccade task.Journal of Experimental Psychology: General, 123, 374-393.

Rothlind, J. C , Posner, M. I., & Schaughency, E. A. (1991). Lateralizedcontrol of eye movements in attention deficit hyperactivity disorder.Journal of Cognitive Neuroscience, 3, 377-381.

Sakurai, Y. (1990). Hippocampal cells have behavioral correlates duringthe performance of an auditory working memory task in the rat.Behavioral Neuroscience, 104, 253-263.

Schachar, R., & Logan, G. D. (1990). Impulsivity and inhibitory controlin normal development and childhood psychopathology. Develop-mental Psychology, 26, 710-720.

Schneider, W. (1988). Micro experimental laboratory: An integratedsystem for IBM PC compatibles. Behavior Research Methods, Instru-ments & Computers, 20, 206-217.

ShalUce, T. (1982). Specific impairments of planning. PhilosophicalTransactions of the Royal Society of London Bulletin, 298, 199-209.

Shaywitz, B. A., Shaywitz, S. E., Byrne, T, Cohen, D. J., & Rothman,S. (1983). Attention deficit disorder: Quantitative analysis of CT.Neurology, 33, 1500-1503.

Shue, K. L., & Douglas, V.I. (1992). Attention deficit hyperactivitydisorder and the frontal lobe syndrome. Brain and Cognition, 20,104-124.

Simon, H. A. (1975). The functional equivalence of problem solvingskills. Cognitive Psychology, 7, 268-288.

Snow, J. H. (1990). Investigation of a mental rotation task with school-age children. Journal of Psychoeducational Assessment, 8, 538-549.

Stuss, D. T, & Benson, F. (1984). Neuropsychological studies of thefrontal lobes. Psychological Bulletin, 95, 3-28.

Swanson, J., Kinsboume, M., Roberts, W., & Zucker, K. (1978). Time-response analysis of the effect of stimulant medication on the learningability of children referred for hyperactivity. Pediatrics, 61, 21-29.

Swanson. J., Posner, M., Potkin, S., Bonforte, S., Youpa, D., Fiore, C ,Cantwell, D., & Crinella, F. (1991). Activating tasks for the studyof visual-spatial attention in ADHD children: A cognitive anatomicapproach. Journal of Child Neurology, 6 (Suppl.), S119-S127.

Voeller, K. K. S. (1986). Right-hemisphere deficit syndrome in children.American Journal of Psychiatry, 143, 1004-1009.

Voeller, K. K. S. (1991). What can neurological models of attention,intention, and arousal tell us about attention-deficit hyperactivity dis-order? Journal of Neuropsychiatry, 3, 209-216.

Voeller, K. K. S., & Heilman, K.. M. (1988). Attention deficit disorderin children: A neglect syndrome? Neurology, 38, 806-808.

Wechsler, D. (1974). Manual for the Wechsler Intelligence Scale forChildren-Revised, New York: Psychological Corporation.

Woodcock, R. W, & Johnson, M. B. (1989). Woodcock-Johnson Psy-


cho Educational Battery—Revised. Allen, TX: DLM Tfeaching Bialer (Eds.), Advances in learning and behavioral disabilities (Vol.Resources. 4, pp. 273-343). Greenwich, CT: JAI Press.

Zametkin, A. J., Nordahl, T. E., Gross, ML, King, A. C , Semple, W. E.,Rumsey, J., Hamburger, S., & Cohen, R. M. (1990). Cerebral glucosemetabolism in adults with hyperactivity of childhood onset. New En- Received August 4, 1994gland Journal of Medicine, 323. 1361-1366. Revision received October 10, 1997

Zentall, S. S. (1985). A context for hyperactivity. In K. D. Gadow & I. Accepted October 10, 1997 •

AMERICAN PSYCHOLOGICAL ASSOCIATION

SUBSCRIPTION C L A I M S INFORMATION Today's Date:

We provide this form to assist members, institutions, and nonmember individuals with any subscription problems. With theappropriate information we can begin a resolution. If you use the services of an agent, please do NOT duplicate claims throughthem and directly to us. PLEASE PRINT CLEARLY AND IN INK IF POSSIBLE.

PRINT FULL NAME OR KEY NAME OF IKSTTTUTION MEMBERORCUSTOMERNUMBER(MAYBEFOUNDONANYPASTISSUELABEL)

ADDRESS DATE YOUR ORDER WAS MAILED (OR PHONED)

PREPAID CHECK CHARGECHECK/CARD CLEARED DATE:_

STATE/COUNTRY ZIP(If possible, send a copy, front and back, of your cancelled check to help us in CUT researchof your claim.)

YOUttNAMEANDPHONENUMBEK ISSUES: MISSING DAMAGED

TITLE VOLUME OR YEAR NUMBER OR MONTH

Thank you. Once a claim in received and resolved, delivery of replacement issues routinely lakes 4-6 weeks.

^ — ^ — (TO BE FILLED OUT BY APA STAFF) "

DATE RECEIVED: DATE OF ACTION: _ACTION TAKEN: INV. NO. & DATE:STAFF NAME: LABEL NO. & DATE:

Send this form to APA Subscription Claims, 7S0 First Street, NE, Washington, DC 20002-4242

PLEASE DO NOT REMOVE. A PHOTOCOPY MAY BE USED.

A Neuropsychological Examination of the Underlying … are thought to play a major role in...

Documents

Transcript of A Neuropsychological Examination of the Underlying … are thought to play a major role in...