UNLV Theses/Dissertations/Professional Papers/Capstones

5-1-2014

WISC-IV Profiles in Children with Attention

Deficit Hyperactivity Disorder and Comorbid
Learning Disabilities
Elyse Parke
University of Nevada, Las Vegas, [email protected]

Follow this and additional works at: http://digitalscholarship.unlv.edu/thesesdissertations

Part of the Psychology Commons

Repository Citation
Parke, Elyse, "WISC-IV Profiles in Children with Attention Deficit Hyperactivity Disorder and Comorbid Learning Disabilities"
(2014). UNLV Theses/Dissertations/Professional Papers/Capstones. Paper 2130.

This Thesis is brought to you for free and open access by Digital Scholarship@UNLV. It has been accepted for inclusion in UNLV Theses/
Dissertations/Professional Papers/Capstones by an authorized administrator of Digital Scholarship@UNLV. For more information, please contact
[email protected].



WISC-IV PROFILES IN CHILDREN WITH ATTENTION DEFICIT

HYPERACTIVITY DISORDER AND COMORBID

LEARNING DISABILITIES

by

Elyse Martina Parke

Bachelor of Arts in Psychology

Westmont College

2009

A thesis submitted in partial fulfillment of

the requirements for the

Master of Arts - Psychology

Department of Psychology

College of Liberal Arts

The Graduate College

University of Nevada, Las Vegas

May 2014


 




THE GRADUATE COLLEGE

We recommend the thesis prepared under our supervision by

Elyse Parke

entitled

WISC-IV Profiles in Children with Attention Deficit Hyperactivity

Disorder and Comorbid Learning Disabilities

is approved in partial fulfillment of the requirements for the degree of

Master of Arts - Psychology

Department of Psychology

Daniel Allen, Ph.D., Committee Chair

Chris Heavey, Ph.D., Committee Member

Brad Donohue, Ph.D., Committee Member

Lewis Etcoff, Ph.D., Additional Committee Member

Tara Raines, Ph.D., Graduate College Representative

Kathryn Hausbeck Korgan, Ph.D., Interim Dean of the Graduate College

May 2014

ii






ABSTRACT

WISC-IV Profiles in Children with Attention Deficit Hyperactivity Disorder and

Comorbid Learning Disabilities
by

Elyse M. Parke, B.A.

Dr. Daniel N. Allen, Examination Committee Chair

Professor of Psychology
University of Nevada, Las Vegas

Attention Deficit Hyperactivity Disorder (ADHD) and learning disabilities (LD),

including Reading Disorder (RD), Disorder of Written Expression (DWE), and

Developmental Coordination Disorder (DCD) all co-occur at high rates. Previous

research indicates increased neurocognitive impairment in ADHD with the presence of

comorbid diagnoses. However, few direct comparisons between intellectual profiles of

children with one or multiple ADHD and LD diagnoses are available, specifically for the

Wechsler Intelligence Scale Fourth Edition (WISC-IV), despite its frequent and historical

use with this population. Profile analysis may contribute insights into spared and

impaired abilities. Therefore, the present study addressed these matters by comparing

WISC-IV profiles of children with ADHD and comorbid LD. Participants included 301

children with ADHD-Inattentive (n=101), ADHD-Combined (n=79), ADHD-DCD

(n=42), and ADHD-RD and/or Disorder of Written Expression (ADHD-RD-DWE)

(n=79). Children were 10.2 years old, 69% male, with a Full Scale IQ of 101.5.

Diagnoses of ADHD and learning disorders were established through comprehensive

evaluations including behavioral symptom ratings, interviews with parents, and

neuropsychological measures. Results indicated a significant group by Index score

interaction, which was primarily caused by the ADHD-RD-DWE group performing

iii






significantly worse (p<.05) on Verbal Comprehension Index (VCI) and Perceptual

Reasoning Index (PRI) than all other groups. This group demonstrated a relatively flat

profile, while the ADHD-DCD group demonstrated a sloping profile

(VCI>PRI>Working Memory>Processing Speed). Differences in ADHD presentations

were also found, with the ADHD-Inattentive group exhibiting slower processing speed

than the ADHD-Combined group. Findings indicate differences in intellectual profiles of

children with ADHD and LD as well as ADHD presentations. The combination of LD

and ADHD results in unique intellectual profiles, indicating clinical and theoretical utility

in distinguishing between these disorders. Further investigation is needed to determine

the extent to which these profiles are predictive of academic, social, and behavioral

outcomes.

iv






TABLE OF CONTENTS

ABSTRACT…………………………………………………………………….……......iii

TABLE OF CONTENTS………………………………………………………………...iv

LIST OF TABLES……………………………………………………………………….vii

LIST OF FIGURES……………………………………………………………………..viii

CHAPTER 1: INTRODUCTION………………………………………………………....1

CHAPTER 2: LITERATURE REVIEW………………………………………………….4

Attention Deficit Hyperactivity Disorder………………………………………………....4
Comorbid Learning Disorders…………………………………………………….………5
Etiological Theories……………………………………………………………….……....7
Wechsler Intelligence Scale for Children Profiles…………………………………….…..9
Attention Deficit Hyperactivity Disorder………………………..........................11
Reading Disorder...……………………………………………………………....14
Developmental Coordination Disorder………………………………………......16
Affected Brain Regions………………………………………………………………......18
Conclusion…………………………………………………………………….…………20
Research Aims and Study Hypotheses…………………………………………….….....21

CHAPTER 3: METHODOLOGY……………………………………………………....23
Participants………………………………………………………………………………23
Measures………………………………………………………………………………....23
Procedure.…………………………………………………………………….................26
Data Analysis …...………………………………………………………………………27
Data Entry and Screening……………………………………………………….27
Preliminary Analyses……………………………………………………………28
Primary Analyses……………………………………………………..................28

CHAPTER 4: RESULTS……………………………………………………………….29

CHAPTER 5: DISCUSSION……………………………………………………….…..38
ADHD Subtype Profile Differences…………………………………………………….39
Reading Disorder and Disorder of Written Expression Profile…………………….…...40
Developmental Coordination Disorder Profile………………………………………….43
Differences in Underlying Mechanisms for Working Memory and
Processing Speed Abilities among Groups……………………………………………...45
Clinical Applications…………………………………………………………………….47
Limitations and Future Directions……………………………………………………….48

REFERENCES…………………………………………………………………………..51

v






Curriculum Vitae………………………………………………………………………...78

vi






LIST OF TABLES

Table 1 Demographic and Clinical Information of the Sample…..……………....29

Table 2 Descriptive Statistics for the WISC-IV Index Scores and Between
Subjects Effect…………………………………………………………...31
Table 3 Descriptive Statistics for the WISC-IV Index Scores and Within Subjects
Effects…………………………………………………..………………..34
Table 4 Descriptive Statistics for the WISC-IV Subtest Score…………………...36

vii






LIST OF FIGURES

Figure 1 WISC-IV Index Performance………………………..…………………..32

Figure 2 WISC-IV Subtest Performance Grouped by Index…..………………….37

viii






CHAPTER 1

INTRODUCTION

Children with Attention Deficit Hyperactivity Disorder (ADHD) exhibit a number

of cognitive and behavioral abnormalities that are the direct result of the disorder itself,

but that also result from coexisting learning disabilities (LD), which occur at a high

frequency. Research indicates that Reading Disorder (RD) and Developmental

Coordination Disorder (DCD) are among the most commonly co-occurring learning

disabilities with ADHD (Mayes & Calhoun, 2006). Children with these comorbid

conditions exhibit worse outcomes across a number of domains, including academic

success, social skills, and occupational outcome in adulthood (Glomb, et al., 2006; Eden

& Vaidya, 2008).

Intellectual assessment has been useful in characterizing the cognitive deficits

demonstrated by these children for both clinical and research purposes. From a clinical

perspective, results of intellectual assessments are commonly used to develop educational

assistance plans and intervention methods, as well as measure treatment outcomes. From

a research perspective, results of IQ batteries in combination with other

neuropsychological and achievement tests have provided a framework for LD taxonomies

as well as provided insight into brain regions that are differentially affected by the

disorders.

However, currently available information for the commonly used Wechsler

Intelligence Scale for Children (WISC) is limited in a number of respects, including

limited research examining the impact that co-existing learning disabilities have on index

and subtest score profiles. Previous research on the WISC-R and WISC-III indicates

1






children with comorbid ADHD and RD may demonstrate weaker performance on the

Freedom from Distractibility Index (FDI) than children with ADHD alone, but have

comparably low scores on the Processing Speed Index (PSI) (Rucklidge et al., 2002).

The extent to which verbal deficits independent of reading ability and working memory

abilities in children with ADHD and RD are impacted on the WISC-IV is currently

ambiguous. Furthermore, research is needed to examine perceptual reasoning and

working memory abilities in children with ADHD and DCD. For example, Loh, Piek,

and Barrett (2011) found that these children received lower scores on their WISC-IV

Perceptual Reasoning Index (PRI), compared to children with ADHD only. However,

this study used a small sample and a shortened form of the WISC-IV, not including the

Working Memory Index (WMI), which is an ability likely impacted by a diagnosis of

DCD (Alloway, 2011).

While there is abundant literature on the WISC-III, there is evidence that

revisions made in the development of the WISC-IV have altered some of the

characteristics of the test. For example, while the Perceptual Organization (POI) factor

has historically been sensitive to brain injury, recent studies of children with traumatic

brain injury (TBI) suggest that the WISC-IV Perceptual Reasoning Index (PRI) is no

more sensitive to brain injury than the Verbal Comprehension Index (VCI), which has

been typically viewed as being resilient in the presence of brain injury (Allen et al., 2010;

Donders & Janke, 2008).

Given the limitations apparent in the existing literature and the important role that

IQ testing has played in the identification, treatment and research of LD, the current study

examines WISC-IV performance in children and adolescents diagnosed with ADHD

2






only, compared to groups of children diagnosed with both ADHD and learning disorders.

In the following sections, information relevant to ADHD, learning disorders, and

intellectual assessment is discussed.

3






CHAPTER 2

LITERATURE REVIEW

Attention Deficit Hyperactivity Disorder

Attention Deficit Hyperactivity Disorder (ADHD) is a neurodevelopmental

disorder characterized by poor attention, excessive activity, and impulsivity (APA, 2000).

Developmentally inappropriate symptoms typically arise in childhood and may endure

throughout adolescence and adulthood (Biederman, 2005). Symptoms manifest across

more than one setting, such as home and school. This diagnosis is associated with poor

outcomes in academic performance and social functioning (Daley & Birchwood, 2010;

Barkley, 1990). ADHD is often the most commonly diagnosed psychological disorder in

childhood (Barkley, 1998), with prevalence rates ranging from 3% to 7% (APA, 2000).

Male to female ratios vary from 2:1 to 9:1, relative to differences in subtype and clinical

setting (APA, 2000; Cuffe et al., 2005).

ADHD is classified into the following subtypes: predominantly hyperactive

(ADHD-H), predominately inattentive (ADHD-I), and combined (ADHD-C). The

Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV; APA,

2000) defines ADHD-I as evidencing symptoms of inattention, such as poor sustained

attention, difficulty attending to detail, and distractibility. Additionally, symptoms of

hyperactivity (e.g., squirming) and impulsivity (e.g., difficulty awaiting turn and

tendency to interrupt others) must be present for a diagnosis of ADHD-C. The

predominantly hyperactive subtype is confined to symptoms of hyperactivity and

impulsivity. Research primarily examines ADHD-I and ADHD-C and suggests that

ADHD-C is often distinguishable from the other subtypes by increased externalizing

4






problems and diagnoses (Eiraldi et al., 1997; Morgan, Hynd, Riccio, & Hall, 1996).

These findings are not surprising given the increased symptoms required to meet criteria

for ADHD-C.

Comorbid Learning Disorders

Children with ADHD are a highly diverse group with various comorbid

conditions, including psychiatric diagnoses such as anxiety, mood, and substance use

disorders (Bierdman, Newcorn, & Sprich, 1991; Kessler et al., 2006). Learning disorders

co-exist at exceptionally high rates ranging from 25-40% (Barkley, 1998) within the

broad category of LD, from 25-40% within Reading Disorder (RD) (August & Garfinkel,

1990; Semrud-Clikeman et al., 1992), and as high as 50% for Developmental

Coordination (DCD) (Crawford & Dewey, 2008; Pitcher, Piek, & Hay, 2003). Reading

Disorder is defined as deficits in reading achievement in areas such as speed, accuracy, or

comprehension, compared to an individual’s age, grade level, and general intelligence

(APA, 2000). Disorder of Written Expression is characterized by deficits in writing, such

as spelling, organization, excessive grammatical errors (APA, 2000). Developmental

Coordination Disorder is characterized by deficits in the development of motor

coordination, which is unexplained by the child’s intelligence, neurological, or

psychiatric disorders (APA, 2000). Children with this disorder often show observable

behaviors such as poor posture, clumsiness, and difficulties holding a pencil.

The presence of these comorbid conditions are associated with increased

cognitive, emotional, and academic impairments (Bonafina, et. al., 2000), as well as

overlapping symptoms among the disorders. For example, DCD has been linked to

similar psychosocial difficulties found in ADHD, such as emotional problems, behavioral

5






concerns, poor overall academic performance, and learning difficulties (Tseng, Howe,

Chuang, & Hsieh, 2007; Dewey et al., 2002). The combination of RD and ADHD has

been associated with poor motor coordination. For example, previous studies found

children diagnosed with ADHD and RD exhibited more difficulties with visual motor

coordination and planning compared to children only diagnosed with ADHD (August &

Garfinkel, 1990; Robins, 1992). Additionally, children with ADHD may show deficits in

motor performance and reading skills, yet not meet criteria for either LD diagnosis

(Mayes, Calhoun, & Crowell, 2000). Similarly, children with an LD often exhibit

symptoms of inattention without meeting criteria for ADHD. Because of the high

frequency of comorbidity of ADHD, DCD, RD, and the overlap of symptoms further

investigation of these disorders may provide insights into genetic susceptibility and

atypical cognitive development.

Although LD research has largely focused on RD, DCD, and math deficits, some

studies suggest that disorders of written expression exceed prevalence rates of

mathematics or reading disorder (Mayes & Calhoun, 2006). Disorders of Written

Expression also commonly co-occur with RD since both diagnoses result in deficiencies

in language skills and require similar cognitive abilities (Parodi, 2006; Lindstrom et al.,

2007). Writing is commonly understood as expressive language and reading understood

as a receptive language skill. However, these psycholinguistic processes are interrelated

in that the production of written language depends on comprehension capabilities

(Shanahan, 1997). Previous studies found positive correlations between reading and

writing performance (Stotsky, 1983; Shanahan & Lomax, 1986; Tierney & Shanahan,

1991), suggesting a common cognitive mechanism underlies both reading and writing

6






performance (Parodi, 2006; Sadoski & Paivio, 2001). Furthermore, it is relatively rare to

receive a diagnosis of Disorder of Writing Expression in the absence of a Reading

Disorder (APA, 2000). Research also indicates that this diagnosis may be a direct

consequence of language and word decoding deficits resulting in RD (Lindstrom, 2007).

Therefore, the present study has included Disorder of Written Expression and combined

cases with RD in subsequent theoretical understanding, hypotheses, and analyses.

Etiological Theories

The frequency of comorbidity of symptoms and diagnoses suggests implications

for etiology (Crawford & Dewey, 2008). Different theorists have made conjectures about

the established overlap in neurodevelopmental disorders. Gilger and Kaplan (2001)

proposed a generalized atypical brain development (ABD) framework that emphasizes

nonspecific developmental deficits rather than distinct diagnoses. They suggest ABD

should be used in conjunction with current diagnostic definitions as a guide to

understanding the proposed common etiology of these disorders. However, others have

suggested these disorders are independent in etiology and individual differences in

symptom profiles should not be overlooked (Cruddance & Riddell, 2006).

A substantial body of research has investigated the genetic relationship between

ADHD and RD. Willcutt and colleagues (2005) examined four common hypotheses

explaining this comorbidity. Founded on population base rates for the respective

disorders, the cross-assortment hypothesis suggests an individual with a diagnosis of

ADHD is more likely to choose a partner with RD and vice versa. Thus, genetic

susceptibility in both parents results in an additive combination of symptoms from each

disorder. Faraone and colleagues (1993) found this theory to best explain the

7






comorbidity rates of ADHD and RD when examining biological relatives of children

diagnosed with ADHD. However, following studies (Doyle et al., 2001; Friedman et al.,

2003) did not replicate these results indicating the cross-assortment hypothesis may not

provide a basis for most cases of comorbidity.

The phenocopy hypothesis proposes that a diagnosis of one disorder results in an

increased risk for manifesting symptoms of another disorder (Pennington, Groisser, &

Welsh, 1993). For example, a child with reading difficulties may exhibit symptoms of

inattention and hyperactivity due to frustration with reading rather than neurocognitive

difficulties related to RD. However, subsequent research in larger samples has not

supported the phenocopy hypothesis (Rucklidge & Tannock, 2002; Seidman, et al., 2001;

Willcutt, et al., 2001).

Another proposal is the cognitive subtype hypothesis suggesting children with

both ADHD and RD possess distinct etiological factors from children that develop

ADHD or RD alone. Thus, children with comorbid ADHD and RD would have different

symptomology and cognitive deficits than individuals with only one disorder. Research

regarding this hypothesis is inconclusive and requires further investigation. Rucklidge

and Tannock (2002) supported this hypothesis by finding their comorbid group

performed significantly worse on color naming measures than the groups with only

ADHD or RD. However, other studies found that groups with ADHD and RD exhibited

similar cognitive deficits suggesting an additive combination from each individual

disorder (Pisecco et al., 2001; Swanson, Mink, & Bocian, 1999; Willcutt, et al., 2001).

Finally, Willcutt and colleagues (2005) examined the common etiology influences

hypothesis describing a mutual genetic basis for the high co-occurrence of ADHD and

8






RD. This hypothesis assumes cognitive deficits in each disorder are partly shared;

meaning children with comorbid diagnoses will have similar weaknesses as children with

only one diagnosis (Willcutt et al., 2003). Twin studies have implicated that RD and

ADHD separately are highly heritable and polygenic (DeFries & Alarcón, 1996; Faraone

et al., 2001; Fisher & DeFries, 2002), as well as providing evidence suggesting common

genetic influences impacting comorbidity (Stevenson et al., 1993; Willcutt, et al., 2003).

Etiological research on DCD and ADHD has largely resembled the research on

co-occurring ADHD and RD. Evidence suggests DCD and ADHD have a shared genetic

heritability (Martin, Piek, & Hay, 2006). Others have suggested that there are differences

in cognitive symptomology providing evidence against the common etiology influences

hypothesis (Loh, Piek, & Barrett, 2011). These etiological theories provide theoretical

foundations for examining cognitive similarities and discrepancies in ADHD and LD.

Wechsler Intelligence Scale for Children Profiles

Because ADHD, RD, DWE, and DCD are not only genetically, but also

cognitively related (Nigg, Hinshaw, Carte, & Treuting, 1998; Rucklidge & Tannock,

2002; Willcutt, et al., 2005), the existing literature has attempted to distinguish shared

and distinct cognitive deficits among these disorders. Research has used a variety of

cognitive measures including intelligence, academic achievement, and working memory

measures (Alloway, 2011). Particularly, the Wechsler Intelligence Scale for Children

(WISC) is widely used for this purpose because of its extensive history of use and

sensitivity to cognitive deficits within this population (Loh, Piek, & Barrett, 2011; Mayes

& Calhoun, 2006). One of the earliest examples is a study by Nelson and Warrington

(1974) that demonstrated children with LD commonly have lower WISC Verbal IQ,

9






relative to their Performance IQ. WISC profile analysis using composite scores as well

as individual subtest performance continues to be a widely used practice in determining

cognitive strengths and deficits within each disorder (Hale, Fiorello, Kavanagh,

Hoeppner, & Gaither, 2001; Kramer, 1993; Hjelmquist, & Gillberg, 2001).

Despite frequent use of the WISC in clinical populations, more research is needed

to examine the newest version of the WISC, the WISC-IV. The structure of the most

recent revision of the Wechsler Scales differs in notable aspects from its forerunners.

Earlier reorganization was a reaction to evidence supporting alternatives to the original

verbal-performance dichotomy of intelligence. Previous research supported three factor

latent structures comprised of Verbal Comprehension, Perceptual Organization, and

Freedom from Distractibility factors. Further research lead to a fourth processing speed

factor, which addressed evidence indicating the instability of the Freedom of

Distractibility factor (Chan, 1984; Stewart & Moely, 1983). Furthermore, efforts were

made to address theoretical considerations and align the WISC indexes with theories,

primarily the Cattel-Horn-Carroll (CHC) theory of intelligence (Keith et al., 2006). Thus,

the four revised index scores were derived from factor analytic work and theoretical

concerns, making the revised test an empirically salient assessment of actual underlying

cognitive functions.

Most recently, changes from the WISC-III to the WISC-IV include renaming the

FDI to the Working Memory Index (WMI) and replacing Arithmetic with Letter-Number

Sequencing within this index. This change is intended to better measure working

memory by decreasing the need for mathematical academic knowledge. Three of four

subtests on the WISC-III POI and VCI have been retained in the WISC-IV PRI and VCI.

10






Two timed visual-motor subtests (Picture Arrangement and Object Assembly) were

removed from the POI and replaced with two motor-free visual reasoning subtests

(Picture Concepts and Matrix Reasoning). Thus, there is less emphasis on motor

coordination and speediness on the PRI, which may be an advantage to groups with

limitations in these skills. On the VCI, the Information subtest was changed from a core

to supplemental subtest, increasing emphasis on verbal reasoning rather than verbal

academic abilities. Despite alterations to subtests and indexes the WISC-III and IV have

a high correlation (0.88) on Full Scale Intelligence Quotient (FSIQ) and Indexes ranging

from 0.72 to 0.88 (Wechsler, 2003). However, the extent to which alterations in the

WISC-IV impact scores in clinical populations is not yet fully understood.

Attention Deficit Hyperactivity Disorder

Extensive research has examined WISC-R and WISC-III profiles for clinical

cases. Various studies have demonstrated that children with ADHD have

characteristically lower FDI and PSI, relative to their scores on the POI and VRI

(Calhoun & Mayes, 2005; Naglieri, Goldstein, Iseman, & Schwebach, 2003; Wechsler,

1991). Additionally, children with ADHD demonstrate significantly worse performance

on the Coding subtest compared to Symbol Search subtest even thought these are the two

primary measures contributing to the PSI (Calhoun & Mayes, 2005; Mayes & Calhoun,

2003; Naglieri et al., 2003; Snow & Sapp, 2000; Wechsler, 1991). Extensive research

has shown that poor performance on the Coding subtest is strongly associated with

neurological dysfunction (Fiedorowicz et al., 2001; Hooper & Tramontana, 1997; Light,

Pennington, Gilger, & DeFries, 1995), which provides some support for neurobiological

theories of ADHD.

11






While considerable research has validated these WISC-III profiles, revisions to

the WISC-IV may affect their generalizability for children with ADHD and learning

disabilities. Currently, limited data are available regarding the effects of WISC-IV

revisions to profile analysis. The technical manual for the WISC-IV reported similar

patterns of performance as on the WISC-III for a small sample of children with ADHD.

This sample performed poorest on the WMI and PSI, indicating difficulties with short-

term storage and the manipulation of information (working memory), as well as quickly

performing simple clerical tasks (processing speed) (Wechsler, 2003). The highest scores

for this sample were for subtests comprising the VCI and PRI. The only study to date

comparing WISC-III and WISC-IV profiles in children with ADHD found that they

produced comparable results in that the lowest scores were obtained on subtests that

composed the FDI/WMI and PSI (Mayes & Calhoun, 2006). Furthermore, the WISC-IV

results demonstrated larger index discrepancies, indicating the new version might have

increased sensitivity to ADHD.

Furthermore, other measures confirm WISC findings of weaknesses in processing

speed and working in children with ADHD. For example, measures such as the Trail

Making tests (Shanahan et al., 2006), Stroop color word naming (Willcutt, 2010), and

reaction time on continuous performance tasks have also provided evidence for

processing speed deficits in ADHD (Wodka et al., 2007). Additionally, working memory

tasks such as the verbal working memory (VWM) test from the WRAML-2 (Sheslow &

Adams, 2003), spatial span (Alloway, 2011), and Corsi blocks have demonstrated

weaknesses in verbal and visuospatial working memory (Martinussen, Hayden, Hogg-

Johnson, & Tannock, 2005). These findings are relevant to understanding deficits in

12






children with ADHD, because processing speed and working memory abilities have

important implications for academic achievement. For example, decreased processing

speed can lead to poor reading fluency, which consequently impacts other cognitive

processes (Jacobson et al., 2011). Furthermore, working memory scores predict reading

achievement separate from phonological abilities in normal children (Swanson & Beebe-

Frankenberger, 2004). Difficulties with working memory have also been associated with

worse performance on mathematical computation and story problem solving (Bull &

Scerif, 2001; Geary, Hoard, & Hamson, 1999; Swanson & Sachse-Lee, 2001), and

reading (Alloway et al., 2009). Thus, working memory deficits present in children with

ADHD may also result in further academic difficulties, beyond symptoms directly related

to their diagnosis. These problems with working memory may stem from a lack of

behavioral inhibition within this population (Barkley, 1997; Pennington & Ozonoff,

1996).

Past research has also investigated differences between ADHD subtypes, but

found less consistent evidence for distinct neuropsychological profiles among these

groups. Numerous studies found no significant cognitive differences across ADHD

subtypes, which may call into question the validity of the subtypes (Chhabildas,

Pennington, & Willcutt, 2001; Hinshaw et al., 2002; Huang-Pollock, Nigg, & Carr,

2005). However, studies that found differences suggest children ADHD-I often exhibit a

sluggish cognitive tempo and slower processing speed than children diagnosed with

ADHD-C (Barkley, Grodzinsky, & DuPaul, 1992; Calhoun & Mayes, 2005; Hartman,

Willcutt, Rhee, & Pennington, 2004). Mayes and colleagues (2009) reported that the

WISC-III and WISC-IV captured this distinction by finding significantly lower PSI

13






scores in the ADHD-I group compared to the ADHD-C groups. However, other studies

do not find differences between ADHD subtypes on IQ and processing speed measures

(Chhabildas, Pennington, & Willcutt, 2001; Farone et al., 1998). Thus, further research is

required to determine whether profile differences do exist between the ADHD subtypes.

Reading Disorder

Because little research has investigated comorbid ADHD with RD and DWE,

research examining deficits in children with only RD may be useful in predicting

performance in comorbid groups. For instance, Kaufman (1981) suggests dyslexia, a

specific reading disability, is associated with poor performance on Arithmetic, Coding,

Information, and Digit Span subtests (ACID profile). Kaufman (1994) later altered the

profile with Symbol Search rather than Information appearing among the weakest subtest

in children with RD (SCAD profile). These profiles indicate that Verbal IQ (VIQ) will

be lower compared to Performance IQ (PIQ) in this population. However, this pattern of

performance is sometimes reversed, which may be due to the variability within children

with dyslexia (Thomson, 2003). Furthermore, studies on the WISC-R and WISC-III

indicate that children with RD perform significant worse on the FDI and PSI compared to

their VCI and POI (Thomson, 2003). Kaufman suggests that the Freedom from

Distractibility factor may “hold the key to competent LD assessment” (1981, p. 521).

Research investigating the WISC-IV demonstrated similar results for the WMI

and PSI. For example, De Clercq-Quegebeur and colleagues (2010) found children with

dyslexia performed significantly worse on the WMI with weaker performance on the PSI

compared to their VCI and PRI. Similarly, Wechsler (2003) found that children with RD

performed poorest on the WMI. These findings relate to the cognitive deficits of

14






phonological processing already associated with RD (Swanson, 1999). Specifically,

Digit Span and Letter Number Sequencing subtests require information to be processed in

the phonological loop of the working memory system (Baddeley, 2001).

In contrast to WMI and PSI findings, De Clerq-Quegebeur and colleagues (2010)

did not show significant impairment on the WISC-IV VCI and PRI measures. This result

also supports research showing advanced language abilities can develop in children with

RD (Vellutino et al., 2004). Despite these findings, Wechsler (2003) found that

Vocabulary was among the lowest scored subtests in children with RD. This population

has limited reading development, which may suppress their ability to gather and retain

verbal information (Stanovich, 1986).

Research on both ADHD and RD individually demonstrate deficits in working

memory and processing speed. While there is considerably less research investigating

WISC profiles in children with comorbid ADHD and RD, some studies have

demonstrated a distinct cognitive profile. For example, Rucklidge and colleagues (2002)

used a shortened version of the WISC-III (Vocabulary, Block Design, Arithmetic, Digit

Span, Symbol Search, and Coding subtests) to examine a group with ADHD, RD, and

comorbid ADHD and RD. Results indicated that children with ADHD showed

impairment on the PSI, while the children with only RD performed worse on FDI. The

group with comorbid conditions reflected the RD group and performed significantly

worse on FDI than the ADHD and control groups.

Further research found similar results using some subtests from the WISC-R and

WISC-III as part of a comprehensive battery examining processing speed and working

memory (Wilcutt et al., 2005). The results indicated that the comorbid groups performed

15






significantly lower on the working memory measures (Digit Span and Arithmetic), than

the ADHD only group. Although, each clinical group performed significantly lower on

PSI than the controls, there were not significant differences among the clinical groups.

Thus, the research suggests children with comorbid ADHD and RD exhibit worse

performance on the WISC-R and WISC-III FDI than children with ADHD (Rucklidge et

al., 2002; Wilcutt et al., 2005), yet have comparably low scores on the PSI (Shanahan et

al., 2006; Wilcutt et al., 2005). Furthermore, Wilcutt and colleagues (2010) found that

children with comorbid ADHD and RD performed worse on WISC-R subtests than

children with ADHD only, indicating verbal reasoning deficits were present and detected

by the previous WISC. However, it has yet to be determined if the WISC-IV is sensitive

to these verbal reasoning deficits in children with ADHD and RD.

Developmental Coordination Disorder

While DCD is primarily characterized by fine and gross motor deficits, some

research suggests children with DCD also exhibit deficits in visual perception

independent of motor functioning (Tsai, Wilson, & Wu, 2008; Piek and Pitcher 2004).

For instance, Tsai, Wilson, and Wu (2008) found this population performed worse than a

control group on a motor-free visual discrimination task as part of the Beery-Buktenica

Developmental Test of Visual–Motor Integration (VMI; Beery 1997) evaluating

perceptual abilities. This measure is of particular interest to the current study because of

its correlation (0.66) with the WISC (VMI Manual), indicating a similar cognitive

element in the VMI.

Although evidence suggests deficits in visual perceptual reasoning in DCD, some

studies indicate this ability might be partially or fully preserved as well as independent of

16






visual motor difficulties in children with DCD (Schoemaker et al., 2001; Bonifacci, 2004;

Henderson et al., 1994). For example, children with visual motor integration difficulties

performed significantly worse on gross motor tasks, yet there were no significant

differences in perceptual abilities (Bonifacci, 2004). Furthermore, previous studies

indicate visual perceptual deficits in children with DCD may be accounted for by the

presence of comorbid conditions (Crawford & Dewey; 2008; Jongmans et al., 2003) and

the motor component frequently within perceptual measures (Schoemaker et al., 2001).

Studies using previous Wechsler scales found worse performance on the PIQ

(Coleman, Piek, & Livesey, 2001; Piek & Coleman-Carman, 1995; Henderson & Hall,

1982), and PSI (Smyth & Glencross, 1986; vanDellen & Geuze, 1988). However, these

findings may be due to the motor component in the subtests composing the PIQ

(Coleman et al., 2001). Studies examining co-occurring DCD and ADHD found

comparable results for visuoperceptual abilities, processing speed, and working memory.

For example, Loh, Piek, and Barrett (2011) found evidence that children with DCD and

comorbid ADHD demonstrated significantly worse perceptual reasoning abilities than a

sample of children with only ADHD on the WISC-IV. This result suggests visuospatial

difficulties present in DCD, which are not apparent in ADHD. Additionally, this study

found that the comorbid group performed significantly lower on PSI than the comparison

group, but not significantly different from the ADHD group. This finding resonates with

other studies indicating children with DCD generally work slower than typically

developing children (Piek & Skinner, 1999).

However, Loh and colleagues (2011) had small samples of children and used a

prorated version of the WISC, which did not include the WMI that may provide further

17






insights into working memory deficits detected by other measures (Alloway, Rajendran,

and Archibald, 2009; Piek, Dyck, Francis, & Conwell, 2007). For example, Alloway

(2011) found decreased performance on all working memory tasks in the Automated

Working Memory Assessment (AWMA; Alloway, 2007) in children with DCD.

Particularly, this group received lower scores on measures targeting visuospatial working

memory, while the group with ADHD performed poorly on both verbal and visuospatial

working memory measures. The author suggests that while ADHD and DCD share a

common deficit in working memory, the cognitive processes in each condition may be

different. Perceptual difficulties in children with DCD may result in worse working

memory performance by impacting the visuospatial sketchpad (visual storage system)

theorized to be a component of the working memory system (Piek, et al., 2007; Baddeley,

2003). Whereas, working memory deficits exhibited in ADHD may be a consequence of

the attention component required to perceive and manipulate information (Chhabildas

Pennington, Willcutt, 2001; Piek et al., 2007).

Affected Brain Regions

Some insight into cognitive profiles in these groups is provided by research

indicating that ADHD subtypes, RD/DWE, and DCD have distinct underlying biological

substrates and core behavioral symptoms. For example, neuroimaging and animal studies

indicate that the symptoms of ADHD are associated with the prefrontal cortex (PFC)

(Dalley et al., 2008; Seidmen et al., 2005; Dalley et al., 1999, & Whishaw et al., 1992),

caudate nucleus (Krain & Castellanos, 2006; Castellanos et al., 2002), and primary motor

cortex (Buchmann et al., 2006; Moll et al., 2000). The prefrontal lobes are implicated in

working memory abilities within this disorder (Klingberg, Forssberg, & Westerberg,

18






2002), while premotor circuits are more associated with slower processing speed in

ADHD (Jacobson et al., 2011). Furthermore, neuroimaging and neurophysiological data

provide evidence for neurobiological differences between ADHD subtypes (Mayes et al.,

2009; Barry et al., 2006; Clarke, Barry, McCarthy, & Selikowitz, 2001; Johnstone, Barry,

& Dimoska, 2003). Reading disabilities are characterized as a language-based disorder,

which results in pervasive symptoms of poor phonological processing, writing, working

memory, and processing speed (Bruck, 1992; Denckla, 1993; Maughan & Carroll, 2006).

Research indicates that cognitive symptoms in RD are a result of unique morphological

differences in the planum temprorale and perisylvian cortex (Galaburda, 1994) that are

not prevalent in children with only ADHD (Hynd et al., 1990). Furthermore, fMRI

studies indicate disruptions in specific left-parieto-temporal and occipito-temporal

circuits, which are thought to be unique to reading disorders (Shaywitz & Shaywitz,

2005). In contrast to ADHD, verbal working memory (Zayed et al., 2013), processing

auditory information (Desroches et al., 2013), and processing speed deficits (Breznitz,

2003) in RD are related to brain circuits involved in phonological processing. Finally,

writing requires coordination between multiple brain regions coordinating the

neuropsychological abilities of executive functioning, language development, and motor

output (Berninger, 1996). In DWE, the network functioning of these regions is disrupted,

meaning that multiple regions, including Broca and Wernicke’s areas for language

processing are affected (Bennett, McHale, & Soper, 2011). Furthermore, DWE may be a

direct consequence of language, phonological processing, and word decoding deficits

resulting in RD (Lindstrom, 2007; Hale & Fiorello, 2004). Thus, similar neurological

mechanisms involved in RD are likely disrupted in DWE. Finally, research indicates that

19






motor deficits in DCD are associated with cerbellar (Cantin et al., 2007; Zwicker,

Missiuna, & Boyd, 2009) and parietal lobe (Zwicker, Missiuna, & Boyd, 2009; Peters et

al., 2013) dysfunction, which may also account for poor perceptual abilities observed in

this population. Thus, there is both biological and clinical evidence supporting distinct

brain regions associated with ADHD and comorbid LD, which may result in distinct

cognitive profiles. Additionally, when a child has comorbid disorders, there may be

multiple dysfunctional circuits compounding deficits, resulting in lower scores on IQ

indexes.

Conclusion

Research indicates working memory and processing speed are both impacted by

separate diagnoses of ADHD, RD, and DCD. However, the current literature suggests

these deficits are due to distinct underlying causes of each disorder, which may result in

varying degrees of impairment. The degree to which these cognitive processes are

impacted by comorbid diagnoses is currently ambiguous in certain respects. For

example, consistent findings have shown children with ADHD perform worse on the

WMI and PSI compared to their VCI and PRI (Calhoun & Mayes, 2005). However,

distinct differences in ADHD-I and ADHD-C subtypes are currently unclear. Whereas

Mayes and colleagues (2006) found children with ADHD-I demonstrated a sluggish

cognitive tempo and consequently a slower WISC PSI, children with anxiety and

depression were included, which may impact performance. Furthermore, many studies

found no differences between subtypes on other cognitive assessments (Huang-Pollock,

Nigg, & Carr, 2005).

20






With regard to children with ADHD and RD, Rucklidge and colleagues (2002)

found that these children appear to perform poorly on the FDI, compared to children with

ADHD only. Thus, it is expected that this finding may translate into the WMI on the

updated WISC-IV. While it might be anticipated that children with RD would perform

worse on the VCI due to the reading capacities required for verbal development, some

research indicates otherwise (De Clerq-Quegebeur et al., 2010; Eckert et al., 2003).

However, few studies have investigated this phenomenon in children with both RD and

ADHD. Finally, Loh and colleagues (2011) found children with DCD demonstrated

lower PRI, compared to children with ADHD only. However, it is unclear the magnitude

to which perceptual reasoning is impacted in DCD, because other evidence suggests this

disorder is primarily confined to motor dysfunction rather than perceptual abilities

(Bonifacci, 2004).

Because of the limitations in the current research, it is imperative to examine

differences in IQ profiles across neurodevelopmental comorbid condition. This analysis

may provide a unique perspective into specific deficits related to each diagnosis, thus

offering evidence for distinct cognitive deficits as a result of each disorder.

Research Aims and Study Hypotheses

Given the limited research examining distinct comorbid learning disorders on

cognitive function in ADHD, research is needed to disambiguate cognitive profiles within

these populations. Thus, the aim of the current study is to examine the effect of comorbid

learning conditions on WISC-IV performance in children with ADHD. If distinct

learning disorders have a unique impact on WISC-IV profiles in ADHD, it is predicted

that:

21






Hypothesis 1

Children in the ADHD-Inattentive and ADHD-Combined subtypes will

demonstrate lower scores on the WISC-IV PSI and WMI compared to their VCI and PRI

scores, with the ADHD-I group exhibiting worse performance on the PSI, than the

ADHD-C group.

Hypothesis 2

The ADHD-RD group will exhibit worse performance on the WMI and VCI, with

comparable scores on the other Indexes, when compared to the ADHD only groups.

Hypothesis 3

The ADHD-DCD group will demonstrate worse performance on the PRI and

comparable scores on the other Indexes when compared to the ADHD only groups.

22






CHAPTER 3

METHODOLOGY

Participants

Children will be included in this study if 1) they had a diagnosis of ADHD-C,

ADHD-I, comorbid ADHD and DCD, or comorbid ADHD and Reading Disorder and/or

Written Expression Leaning Disorder; 2) they had no comorbid pervasive developmental

disorder, traumatic brain injury, or other neurological conditions; 3) they were

administered the WISC-IV as part of a clinical evaluation. Participants included 301

children with ADHD-Inattentive (n=101), ADHD-Combined (n=79), ADHD-DCD

(n=42), and ADHD, RD, and/or Writing Disorder (n=79). The sample was 69.1% male,

10.17 years of age on average, and 64.5% attended a private school. Information

regarding ethnicity was not available for this sample.

Measures

The WISC-IV is designed to assess cognitive abilities in children 6 to 16 years of

age, including verbal and perceptual reasoning abilities as well as working memory and

processing speed. These abilities are reflected in four index scores, including a Verbal

Comprehension Index (VCI), Perceptual Reasoning Index (PRI), Working Memory Index

(WMI), and Processing Speed Index (PSI). A Full Scale IQ (FSIQ) score may also be

calculated by combining these indexes, and serves as an estimate of general intellectual

ability. The Indexes and the Full Scale IQ are standard scores with means of 100 and a

standard deviation of 15. Index scores are measured using 15 subtests, 10 of which are

considered core subtests and 5 considered supplemental. These subtests have a mean of

23






10 and a standard deviation of 3. In the following sections a detailed description is

provided for each index score and its corresponding subtests.

WISC-IV Verbal Comprehension Index (VCI). This index involves the expression of

verbal concepts, application of previously acquired verbal knowledge, and academic

aptitude. These skills are greatly impacted by a child’s education and familiarity with

U.S. culture. The VCI is composed of the following subtests:

Vocabulary. This subtest requires a child to define words with increasingly

difficult vocabulary.

Similarities. This task assesses a child’s ability to recognize conceptual

similarities between words.

Comprehension. Children must answer questions related to their knowledge of

general information and social situations.

WISC-IV Perceptual Reasoning Index (PRI). The PRI assesses nonverbal reasoning

abilities requiring attention to visual elements, spatial skills, and forming abstract

concepts without words. This index is composed of the following subtests:

Block Design. This task involves arranging blocks to match a designated pattern

within a specified time limit.

Picture Concepts. The subtest asks children to choose pictures based upon their

abstract relationship to one another.

Matrix Reasoning. Participants choose pictures to complete a visual and

conceptual pattern.

24






WISC-IV Working Memory Index (WMI). The WISC-IV WMI assesses auditory short-

term memory, the ability to hold and manipulate information, and the effectiveness of

encoding strategies a participant may use. This index also requires good attention and

concentration. The WMI is composed of the following subtests:

Digit Span. Participants are asked to repeat an increasing series of numbers

forward and backwards.

Letter-Number Sequencing. Children mentally manipulate an auditory list of

letters and numbers and say them in ascending numerical and alphabetical order.

WISC-IV Processing Speed Index (PSI). This composite score measures the ability to

perform simple, clerical-type tasks quickly and efficiently.

Coding. Participants quickly copy geometric symbols or numbers that are paired

with numbers according to a key.

Symbol Search. Children identify the presence or absence of a target symbol in a

row of geometric symbols.

The WISC-IV was standardized on a nationally stratified sample of 2,200

children, who were selected based on the 2002 U.S. census data to provide a

representative sample of age, sex, race/ethnicity, geographic region, and socioeconomic

status (parental educational attainment) within the United States population. Samples

were obtained from states that represented each of the four major U.S. geographic regions

used in the 2000 U.S. census (Northeast, South, Midwest, and West). Categories for race

and ethnicity included White, African American, Hispanic, Asian, and Other.

25






The sample was divided into 11 age groups, which allows for calculation of age

corrected standard for subtest, Index and IQ scores. Across all age groups, the reliability

(internal consistency) of the Composite Indexes ranged from .88 to .97, and the reliability

for the subtests ranged from .79 to .89 (The Psychological Corporation, 2003, Table 4.1,

p. 34). Test-retest reliability ranged from .86 to .93 across indexes establishing stability

of the measure (The Psychological Corporation. 2003, Table 4.4, p. 40). Evidence for the

strong validity of the test scores was based on test content, response processes, internal

structure, and intercorrelation studies. Additionally, there is strong evidence indicating

the WISC-IV is a reliable and valid measure for children with ADHD and LD (The

Psychological Corporation. 2003, Table 5.29 & 5.30, p. 87-88).

Procedure

Participants in this study were selected from a consecutive series of 619 cases that

were referred for neuropsychological evaluation over a period of 11 years. Children were

primarily referred to the neuropsychologist primarily because they were experiencing

academic problems, but presented with multiple complaints, such as learning difficulties,

attentional deficits, mood and anxiety symptoms, and behavior disturbances in the home

and at school, among others. Children were assessed by a neuropsychologist in private

practice, in which only cash reimbursement was accepted. The neuropsychologist

diagnosed children with ADHD and learning disorders according to DSM-IV diagnostic

criteria based on parent and child interviews, neuropsychological testing, behavioral

assessment, and other relevant information from medical and educational records. The

WISC-IV was administered as part of larger battery, which included the Behavior

Assessment System for Children (BASC), DSM-IV ADHD Symptoms Rating Scale,

26






Woodcock Johnson Achievement-Third Edition (WJ-ACH-III), and Wide Range

Assessment of Memory and Learning (WRAML).

All assessments were administered according to standardized procedures, as

indicated in testing manuals, by a neuropsychologist or clinical psychology doctoral

candidates under supervision of the neuropsychologist. All measures were scored

according to standardized procedures by a pediatric neuropsychologist or clinical

psychology doctoral candidates under supervision of the neuropsychologist. Children

were individually assessed in a quiet room in a private practice setting. Assessments took

place within one day with total assessment times ranging from 3-6 hours. Children were

given short breaks throughout the assessment to maintain their effort and attention toward

testing materials.

Data Analyses

Data Entry and Screening

Data were double entered into a database and analyzed by SPSS version 19.0.

During the preliminary data screening process, frequency distributions for all variables

were inspected for out of range variables, which would indicate the presence of a data

entry error. Also, all variables were evaluated as potential outliers to guarantee proper

scoring and entry into the database. Outliers were defined as having a score ± 3.0

standard deviations above or below the mean. There were no outliers identified in the

initial screening of the data. Skewness and kurtosis were also evaluated to ensure normal

distribution of the data and were within accepted limits (skewness < +/1, kutosis < +/-

1.5). Given the absence of outliers and the normal distribution of the data, parametric

analyses were used to test the main study hypotheses.

27






Preliminary analyses

Prior to analyses on the primary hypotheses, descriptive statistics were calculated

for each group on demographic variables, including age, gender, type of school (public or

private), and current full-scale intelligence quotient (FSIQ). ANOVA and chi-square

analyses were used to determine whether the four groups significantly differ on these

variables.

Primary Analyses

The general approach to the analysis was to evaluate the study hypotheses using a

mixed model ANOVA that contains group as a between subjects factor, and index score

as a within subjects factor (repeated measure). Given that WISC-IV profiles differences

are hypothesized for each of the groups, it was anticipated that the results of this analysis

will produce a significant main effect for group, a significant main effect for WISC-IV

Index, as well as a significant Group X WISC-IV Index interaction effect. Following a

significant result for these initial analyses, ANOVAs and other appropriate procedures

were used to test the specific predictions made in each hypothesis, by comparing index

score differences within and between groups as appropriate.

28



 


CHAPTER 4

RESULTS

Descriptive statistics for the ADHD and LD groups are presented in Table 1.

Table 1

Demographic and Clinical Information of the Sample

Variable Mean SD Range

Age (years) 10.2 2.8 6.0-16.4
Grade 4.5 2.7 K-10
FSIQ 101.5 11.9 61-142
VCI 104.8 13.7 67-148
PRI 103.4 12.1 71-137
W
WMI 97.6 12.4 62-144
PSI 95.9 12.6 70-136
Frequency
Diagnosis (%)
ADHD-I 33.6%
ADHD-C 26.2%
ADHD-RD/DWE 26.2%
ADHD-DCD 14.0%
Gender (%)
Male 69.1%
Female 30.9%
School (%)
Public 64.5%
Private 35.5%
Comorbid Diagnoses
ODD ( n = 16) 5.3%
Anxiety Disorder (n=21) 7.0%
Mood Disorder (n = 11) 3.7%
Adjustment Disorder (n = 70) 24.3%

Note. FSIQ = Wechsler Intelligence Scale for Children Fourth Edition (WISC-IV) Full scale IQ.
VCI = Verbal Comprehension Index; PRI = Perceptual Reasoning Index; WMI = Working
Memory Index; PSI = Processing Speed Index ADHD-I = Attention Deficit Hyperactivity
Disorder Predominantly Inattentive Type. ADHD-C =Attention Deficit Hyperactivity Disorder
Combined Type. ADHD-DCD = Attention Deficit Hyperactivity Disorder and Developmental
Coordination Disorder; ADHD-RD/DWE = Attention Deficit Hyperactivity Disorder and
Reading Disorder and/or Disorder of Written Expression; ODD = Oppositional Defiant Disorder.

29






Chi-square analyses indicated significant differences among the ADHD-I, ADHD-C,

ADHD-RD/DWE, and ADHD-DCD groups with regard to gender, χ2 (3) = 14.33, p <

.01. Post hoc analyses indicated that the ADHD-DCD group had significantly fewer

females than the other groups. There were no significant differences among groups with

regard to type of school (public/private), χ2(3) = 2.18, p = .54. One-way ANOVAs

indicated that the groups differed significantly with regard to age, F(3,297) = 10.66, p <

.01. Post hoc comparisons indicated that the ADHD-C group was significantly younger

than the ADHD-I and ADHD-RD/DWE groups. Because the scores used in the main

analyses were corrected for age and gender differences are not present on the WISC-IV

scores, these differences were not considered in the main analyses.

Descriptive statistics for the ADHD and LD groups on the WISC-IV Index scores

are presented in Table 2, as are the results for the mixed-model ANOVA examining

potential differences between the groups on the WISC-IV Index scores.

30