2 www-

w
a
w
u
n
m,
m,

~w-I. a

L".".P““""I" - - \>‘~-

 

'WEES

9099

lllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll

31293 0183419864“

LlBRARY

Michigan Stats
Universlty

;_._

 

This is to certify that the

thesis entitled

NEUROPSYCHOLOGICAL THEORIES OF ADHD:
EVIDENCE OF A LATERALIZED DEFICIT?

presented by

Lisa G. Blaskey

has been accepted towards fulfillment
of the requirements for

M . A. degree in Psychology

 

Majorp fe or

Date 7‘36 .. 7?

0-7539 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

PLACE IN REI'URN Box to remove this checkout from your record.
TO AVOID FINE return on or before date due.
MAY BE RECAILED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

DEW 5 290'?
MM: 3 9 arm
35$ ‘2 015ml]?7

 

 

 

 

H

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M c/CIRCIDIbDquS-pu

 

NEUROPSYCHOLOGICAL THEORIES OF ADHD: EVIDENCE FOR A
LATERALIZED DEFICIT?

By

Lisa G. Blaskey

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS
Department of Psychology

1 999

ABSTRACT

NEUROPSYCHOLOGICAL THEORIES OF ADHD: EVIDENCE OF A
LATERALIZED DEFICIT?

By

Lisa G. Blaskey

Attention deficit hyperactivity disorder (ADHD) is a behavioral syndrome
affecting some 3-5 percent of school-aged children and is a common reason for
referral to mental health services. The present study employed carefully
selected, brief, and validated clinical neuropsychological tests to evaluate
whether two competing neuropsychological theories of ADHD (Barkley’s Frontal
Inhibition theory and the Right Hemisphere theory) are applicable to
characterization of deficits in the combined type of the disorder. Findings
supported right hemisphere deficits in complex motor response, but also
supported Barkley’s suggestion of complex task difficulties in the visual motor
domain. Covariation of comorbid disorders and problems, especially reading
level, was necessary to reveal this pattern of findings. More detailed integration
of the Frontal Inhibition and Right Hemisphere theories with one another and with

theories of motor development is recommended.

ACKNOWLEDGEMENTS

First and foremost I would like to thank my advisor, Dr. Joel Nigg, for his
support and guidance throughout this process. Your knowledge, dedication, and
patience made this a tremendous learning experience for me. I truly appreciated
your commitment to this project and your ever-present willingness to “go the
extra mile" for your students.

I would also like to thank my committee members, Drs. Lauren Harris and
Norman Abeles, for their comments and suggestions. Your knowledge and
expertise were tremendous assets and definitely enhanced the quality of this
thesis.

To my family, friends, and labmates who supported me through this process
(especially Matt): your encouragement and support mean the world to me.
Thank-you for helping me meet my goals, for standing by me through trying

times, and for providing that “extra push” when necessary.

iii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................ v
LIST OF FIGURES .......................................................................... vi
INTRODUCTION ............................................................................ 1
HISTORY ....................................................................................... 3
DIAGNOSTIC ISSUES ..................................................................... 7
COMORBIDITY .............................................................................. 8
NEUROPSYCHOLOGICAL THEORIES OF ADHD ............................... 12
BARKLEY’S FRONTAL INHIBITION THEORY .................................... 13
THE RIGHT HEMISPHERE THEORY ................................................ 17
CURRENT STUDY ........................................................................ 30
METHODS
PARTICIPANTS ........................................................................... 31
PROCEDURE .............................................................................. 34
MEASURES ................................................................................. 34
SPECIFIC HYPOTHESIS PREDICTIONS .......................................... 46
PLAN OF ANALYSIS ..................................................................... 49
RESULTS
DATA REDUCTION AND CLEANING ............................................... 51
HYPOTHESIS 1 (FRONTAL INHIBITION THEORY) ............................ 52
HYPOTHESIS 2 (RIGHT HEMISPHERE THEORY) ............................. 59
DISCUSSION
HYPOTHESIS 1 (FRONTAL INHIBITION THEORY) ............................ 67
HYPOTHESIS 2 (RIGHT HEMISPHERE THEORY) ............................. 73
GENERAL CONCLUSIONS ............................................................ 78
LIMITATIONS .............................................................................. 81
FUTURE DIRECTIONS .................................................................. 82
APPENDIX A
THEORY OF ANOMALOUS DOMINANCE ......................................... 89
THE LEFT HEMISPHERE THEORY .................................................. 91
APPENDIX B ................................................................................ 93
APPENDIX C: DETAILED RELIABILITY DATA .................................... 97
REFERENCES ........................................................................... 101

iv

LIST OF TABLES

Table 1. Description of Groups ............................................................ 33
Table 2. Summary of Analyses and Data Analysis Plan ............................ 50

Table 3. Mean scores by group for all dependent variables (mean
raw:SD and mean age-corrected standardized residuals 1 SE) .................. 56

Table 4. Summary of Hypothesis 1 results, including changes in
effect size following covariance of estimated IQ, sex, and comorbid
LD/CD diagnosis .............................................................................. 57

Table 5. Summary of Hypothesis 2 results, including changes in
effect size following covariance of estimated IQ, sex, and comorbid
LD/CD diagnosis ............................................................................. 58

Table 6. Reliability and intra-class correlation estimates for the WRAVMA
Drawing and Rey-Osterrieth Complex Figure tasks ................................. 97

Table 7. Correlations among the live rater and five videotape coders on the
Denckla Time to Do Twenty tasks ....................................................... 98

Table 8. Alpha reliabilities for various combinations of raters on the Denckla
Time to Do Twenty Task ................................................................... 99

Table 9. Correlations among average times (across raters) for each of the 8
tasks on the Denckla Time to Do Twenty motor battery .......................... 100

LIST OF FIGURES

Figure 1. Nonsignificant group by task interaction for the WRAVMA

Drawing and Rey Complex Figure copy trial showing the signifcicant

group main effect and the larger deficit on the Rey than the

WRAVMA ........................................................................................... 64

Figure 2. Differences between ADHD and control subjects on the “ITO-20 left
foot tasks ........................................................................................... 65

Figure 3. Group by side interaction for grooved pegboard performance before
covariation of comorbid LD diagnosis and reading achievement .................... 66

vi

INTRODUCTION

The DSM-IV (American Psychiatric Association, 1994) describes Attention
Deficit Hyperactivity Disorder (ADHD) as arising in very early childhood and
occurring four to nine times more often in males, with excessive motor activity
often manifesting itself from toddlerhood. These children exhibit extreme
impulsivity, inattention, and hyperactivity, which often lead to impaired academic
achievement, peer rejection, and difficult family relationships. Symptoms of the
disorder usually remain stable and severe throughout childhood with many of its
sufferers experiencing a waning of symptoms in late adolescence and adulthood.
The syndrome, as currently defined, is diagnosed behaviorally in the presence of
six behavioral symptoms of either hyperactivity and impulsivity or inattention that
are extreme for age and onset before age seven; impairment from these
symptoms must be present in at least two settings for a period of at least six
months. Three sub-types are defined: Predominantly lnattentive Type,
Predominantly Hyperactive-Impulsive Type, and Combined Type. Though the
majority of children with ADHD exhibit the Combined Type of the disorder, this
classification also allows specifically for the differentiation of primarily inattentive
children who, arguably, have a distinct cognitive and behavioral profile and may
warrant a separate disorder classification (Barkley, 1997).

Accurately diagnosing ADHD can be a difficult task. Currently diagnosis is
based largely on parent-teacher behavioral ratings and reports that suffer from
halo and other confounds. Yet, developing objective and practical laboratory
methods of measuring ADHD symptoms has proven challenging. Due to the
frequent overlapping of symptoms of childhood psychological disorders,
differentiating ADHD from other common childhood disorders has further
complicated this process. Despite the behaviorally-based or diagnostic scheme

currently in place, throughout its history ADHD has been understood as a subtle

neuropsychological disorder (Barkley, 1990). As such, discovering
neuropsychological measures that reliably identify subtypes of ADHD could
prove extremely useful— both in evaluating theories and in clinical assessment.
Characterizing the specific neuropsychological functions impaired in ADHD
subgroups is one way to begin this process.

Currently, there are a number of neuropsychological explanations of the
functional impairment in ADHD. Each theory attributes the impairment to
different neurological substrates, and so, each theory, if taken to its logical
conclusion, would predict different ADHD deficits. Empirical findings in support
of these theories have produced conflicting evidence regarding the nature of
ADHD impairment. The symptom heterogeneity found in individuals with ADHD
diagnoses may partially explain this conflicting evidence. For instance, unique
substrates may underlie different constellations of ADHD symptoms and those of
its common comorbid disorders. Different neuropsychological theories of ADHD
may, therefore, pertain to different subtypes of the disorder. A priori hypothesis
tests of these neuropsychological theories remain rare, with much of the
empirical work atheoretical and much theorizing post hoc (Barkley, 1997).

Studying neuropsychological deficits in ADHD individuals while also
considering differences in symptom clusters may lead to better understanding of
the disorder’s causal or maintaining mechanisms. Neuropsychological theories
may also provide a means of reliably differentiating a "pure", or noncomorbid,
form of the syndrome from syndromes with similar symptoms but different
neuropsychological deficits. Identification of such characteristics could improve
reliability of ADHD diagnosis and help lead to discovery of the disorder's etiology.
The current study examined whether the different neuropsychological theories of
ADHD are applicable to characterization of deficits in the combined type of the

disorder. Conventional standardized clinical neuropsychological tests were

employed to evaluate two competing models of ADHD in children. Whether
these tests proved sensitive to subtle neuropsychological impairments in the
manner predicted evaluated the theories’ respective claims. After reviewing the
history of the ADHD syndrome, two neuropsychological theories will be
described and their key predictions summarized.

L'ILSM!

Because neuropsychological theories of ADHD have undergone
numerous reconceptualizations, it will be helpful to begin with a history of the
diagnosis. Descriptions of excessively hyperactive and impulsive children have
existed— at least in the medical literature— since the latter part of the nineteenth
century. Described by Clouston (1899) as a disorder of "simple
hyperexcitability," this cluster of symptoms was identified as lying on ”the
borderland of psychiatry,“ distinguishing it from other more severe mental
impairments such as mental retardation (Barkley, 1990). Earliest descriptions of
these ”hyperactive" children recounted marked restlessness and poor sustained
attention, which frequently led to school failure despite intact intellect (Still,
1902). Those suffering from these symptoms were more often males and were
recognized to exhibit poor long term outcomes despite attempted interventions.
Still (1902) spoke of the disorder‘s primary deficit as a "defect in moral control"
resulting in problems of "volitional inhibition” much like the problems of
behavioral inhibition and oppositional conduct noted in ADHD children today
(Barkley, 1996)

Reflecting the outlook of the times, psychological and social explanations
for hyperactivity were explicitly rejected during the beginning of the twentieth
century. Instead, the prevailing concept of social Danrvinism assumed heredity
and failure to acquire newly-evolved adaptive traits to result in deficiencies in

morality. Still (1902), for example, claimed that development of defects in moral

control bore little relationship to a child's home environment or exposure to
adequate discipline. He proposed that this condition resulted from a physical
abnormality, either inherited or resulting from peri or postnatal injury. Largely in
agreement with Still, Alfred Tredgold was perhaps the first to introduce the
concept of minimal brain damage in the syndrome. Tredgold (1908) asserted
that some forms of brain damage occurring at birth could result in behavioral
symptoms, despite absence of blatant neurological impairments. Also Clouston
(1899), in describing the disorder as a failure of higher centers of the brain to
inhibit activity, came extremely close to today's conceptualization of the
neurological deficits underlying ADHD (Barkley, 1990).

The focus on possible brain dysfunction associated with this cluster of
symptoms increased following the encephalitis epidemic in the United States
and Europe from 1917-1918. Children who survived encephalitis commonly
exhibited hyperactivity along with learning difficulties and marked changes in
personality. These children were also characterized by impaired attention,
impulse control, and regulation of activity (Ebaugh, 1923).

Hyperactivity has remained one of the primary symptoms of ADHD,
though its emphasis has varied throughout this century. The name hyperkinetic
disease , first used by Kramer and Polnow in the 1930's, reflected the centrality
of this symptom during the period (Sandberg & Barton, 1996). Also during this
time, theories as to the cause of extreme hyperactivity took further form, strictly
maintaining their focus on brain dysfunction. Kahn and Cohen (1934) linked
hyperactive behaviors to abnormal functioning of the activity modulating centers
of the brainstem~ the result of pre or perinatal brain injury. Perhaps most
notable were findings that the behavior of hyperactive children resembled that of
monkeys with frontal lobe ablations, thereby suggesting frontal lobe involvement

in the syndrome (Levin, 1938).

Findings of behavioral similarities in adults with acquired brain lesions and
brain-damaged children led, in the 1940's, to the idea that similar behavioral
patterns could emerge in children who did not exhibit signs of severe
neurological impairment (Strauss 8. Werner, 1943). In the 1950's and 60's, the
notion of a continuum of brain damage severity, postulated by Pasamanick and
colleagues, expanded upon the concept of minimal brain damage to allow for a
spectrum of behavioral outcomes. The more severe the damage during
development, the more severe the behavioral outcomes (Pasamanick, Rogers,
& Lilienfeld, 1956). The result was the replacement of the concept of minimal
brain "damage" with that of minimal brain "dysfunction". As such, it was
recognized that brain damage, in the usual sense, could not be inferred from
behavior alone. Rather, the idea emerged that hyperactive children could suffer
from minor neuropsychological deficits acquired during development (possibly
during fetal development), instead of brain damage, per se (Taylor, 1986).

The Brain Injured Child Syndrome and Minimal Brain Dysfunction
syndromes of the 1950's were the result of these biologically based models of
childhood dysfunction. However, these disorders encompassed symptoms
reflective of a large and diverse group of impaired children, including children
with learning disabilities, epilepsy, and conduct problems. Due to these
syndromes’ inability to accurately differentiate children with different kinds of
dysfunctions, as well as the realization that brain injury was absent in most
diagnosed children, attempts were made in the late 1950's and early 1960's to
classify the disorder more accurately (Chess, 1960).

The term hyperkinetic behavior syndrome was introduced by Laufer and
Denhoff (1957) to describe children with a cluster of symptoms having
hyperactivity as their hallmark. In addition, DSM-ll adopted the title Hyperkinetic

Reaction of Childhood (American Psychiatric Association, 1968). In contrast to

the biologically oriented models of the early part of the century, at the time of
DSM-ll the behavior was understood in relation to psychological conflicts.
Psychoanalytic and behaviorist schools of thought focused on parental causes of
childhood problem behavior (Bettelheim, 1973). Also becoming influential were *
theories of the environmental causes of hyperactive behavior. For example,
Feingold (1975) suggested that allergic reactions to substances in food,
especially food additives, could produce ADHD.

Interest in cognitive and neuropsychological correlates was revived in the
United States by a shifting focus during the 1970's towards sustained attention
deficits as the core symptoms of the disorder (Douglas, 1972). DSM-lll
(American Psychiatric Association, 1980) followed suit by adopting the title
Attention Deficit Disorder with or without Hyperactivity, indicating an attention
deficit-focused view of the disorder, with hyperactivity relegated, for the first time,
to a subsidiary role. Research in the late 1970's and early 1980's, however,
insufficiently supported the role of attentional dysfunctions as primary (Van der
Meere & Sergeant, 1987). DSM-llI-R (American Psychiatric Association, 1987)
subsequently renamed the disorder Attention Deficit Hyperactivity Disorder
(ADHD). Though symptoms of inattention were not excluded, hyperactivity was
seen as at least equally important in the diagnosis of the disorder. Notably, the
subtype of Attention Deficit Disorder without Hyperactivity was relegated to the
category of undifferentiated ADD, suggesting disenchantment with the role of
inattention.

The honing of diagnostic criteria for ADHD based on current empirical
findings continues, with the DSM-IV (American Psychiatric Association, 1994)
criteria for ADHD reflecting the attempt to differentiate more explicitly children
with different symptom patterns. New empirical data from field trials restored the

subtypes distinction absent from DSM-lll-R and created three subtypes based on

predominantly inattentive, predominantly hyperactive, and combined symptoms.
Over the last two decades a resurgence of the biological metaphor in the field of
psychopathology has returned neuropsychological theories of ADHD to
prominence. Enhancement of brain imaging and measurement techniques,
along with further experimental quantification of the disorder's cognitive deficits,
has led to increasing evidence of subtle neuropsychological dysfunction in ADHD
(Barkley, 1997; Pennington & Ozonoff, 1996; Casey et al., 1997).

Nevertheless, the symptom heterogeneity of diagnosed children continues
to create complications in identification of the disorder’s neurological correlates,
and diagnosis remains fully behaviorally-based due to lack of agreed-upon
neuropsychological measures. Neuropsychological deficits in ADHD have ample
historical and empirical support. Research now needs to focus on better
differentiating among proposed theories and on determining which theory most
accurately explains neuropsychological deficits. Furthermore, given the recency
of the new DSM-IV criteria, data regarding neuropsychological performance for
the DSM-IV is needed.

To set up the potential contribution of a neuropsychological study to better
ADHD nosology, a brief review of the merits and drawbacks of the current DSM-
lV classification system follows.

@nostic Issues

One strength of the DSM-IV diagnosis of ADHD is its empirically-based
descriptive classification, which provides a beginning means of organizing and
grouping symptom clusters to assist treatment strategy. This purely behaviorally
descriptive approach is also one of the DSM's main weaknesses. Holding
atheoretical classification as a central tenet, the DSM lacks etiological,
developmental, or other theory-based categorization (Follette & Houts, 1996).

The DSM's tacit adoption of a medical syndrome model fails to address possible

pathways to disorder—- including potentially numerous environmental and genetic
influences. Furthermore, as detailed momentarily, diagnosed children are a
heterogeneous group, often with different symptom clusters and different
comorbid diagnoses. These various groups might exhibit different functional
neuropsychological and cognitive deficits (Pennington, Groisser & Welsh, 1993).
Classification of the disorder according to identified neuropsychological
dysfunction could improve definition of the ADHD construct as well as its
reliability and meaningfulness. Such a method could differentiate ADHD into
subtypes with different neuropsychological correlates, thereby improving our
understanding and eventually our treatment of the disorder.

Cateqorv versus Dimension . Another issue in ADHD concerns possible
pathologizing of what may simply be an extreme of normal temperament
(Barkley, 1996). Due to the categorical classification system adopted by the
DSM, statistical rarity of symptom clusters as well as impairment determines
psychiatric diagnosis. Yet, as a result, no characteristics clearly distinguish
between those with and without the disorder. Identification of
neuropsychological mechanisms associated with the disorder may help to decide
whether ADHD represents a normal extreme of temperament or a distinct
disorder. Perhaps more importantly, however, identifying these mechanisms
may resolve another major weakness in ADHD diagnosis, that of comorbid
disorders.

Comorbiditv

Overview. One of the biggest problems in DSM-IV diagnosis of ADHD is
that the majority—68 percent according to Offord et al. (1987)-— receive one or
more additional diagnoses (Jensen, Martin, & Cantwell, 1997). Eighteen to
twenty-two percent of children experience a psychological disorder sometime

throughout their childhood (Nottleman & Jensen, 1995). For a large percentage

of afflicted children, these disorders do not generally represent fleeting
developmental phases, but instead show patterns of persistence that extend for
at least several years (Cohen, Cohen & Brook, 1993). In addition, psychological
disorders may impede the normal course of development, thereby creating
vulnerabilities to further disorder.

Comorbiditv in ADHD. ADHD appears to show particularly high rates of
comorbidity. It has been commonly found in conjunction with learning
disabilities, anxiety disorders, and other disruptive behavior disorders such as
Oppositional Defiant Disorder and Conduct Disorder (Biederman et al., 1992;
Faraone et al., 1993). ADHD most commonly occurs with Conduct and
Oppositional Defiant disorders, with studies finding average comorbidity rates of
between thirty and fifty percent and rates as high as ninety percent in some
samples (Jensen, Martin, & Cantwell, 1997). However, internalizing disorders
are also commonly diagnosed. Twenty to sixty percent of ADHD children
commonly exhibit anxiety disorders, while ten to thirty percent in some samples
have exhibited comorbid depression (Nottleman &Jensen, 1995).

In addition to high rates of comorbid psychiatric diagnoses, ADHD has
also commonly been associated with academic difficulties and learning
disabilities. According to a review by Hinshaw (1992), depending on the criteria
used to define learning disabilities, ADHD children have comorbidity rates of
between ten and thirty percent for formal learning disorders. School failure is
also very common in ADHD children, with one study finding over one half in need
of tutoring and one third placed in special classes or repeating a grade (Faraone
et al., 1993). Moreover, Faraone et al. found that comorbidity did not account for
the increased rate of academic difficulties, suggesting that mechanisms

producing ADHD symptomatology may also create learning difficulties.

With such high incidence of comorbidity, it is unclear whether ADHD is
really one unitary syndrome or, instead, a series of associated but independent
disorders lumped together under the same diagnostic heading (Jensen, Martin,
& Cantwell, 1997; Schachar & Tannock, 1995). For instance, diagnoses may be
made based on the behavioral manifestation of a disorder, but result from
different cognitive mechanisms than those of the ”pure" (noncomorbid) form of
the disorder (Pennington, Groisser, & Welsh, 1993; Schachar & Tannock, 1995).
Alternatively, multiple disorders may result from unique cognitive,
neuropsychological, or etiological features that are unrelated to those existing in
pure, noncomorbid forms of disorder. Thus, from a neuropsychological
perspective, DSM symptom categorization may unnecessarily differentiate
related dysfunctions into separate diagnoses. The importance of identifying
neuropsychological correlates associated with these disorders as a means to
differentiate and treat underlying impairments thus becomes particulariy salient.

Comorbid Ieaming disabilities in ADHD children exemplify some of these
comorbidity issues. Several different mechanisms for the increased rate of
comorbidity between ADHD and LD have been proposed. For instance, one
view is that symptoms of ADHD in those with LD result solely from LD-associated
behavioral problems, or phenocopy symptoms of ADHD (Cantwell & Baker,

1991 ). Noncomorbid ADHD and LD are thus believed to have very different
neuropsychological and cognitive causes (Pennington, Groisser, & Welsh, 1993).
A second view is that inattention inherent to disruptive behavior disorders
creates LD symptomatology and learning problems, thus positing a common
underlying cognitive or neuropsychological cause (DeLong, 1995). Lastly, other
findings support the coexistence of distinct neuropsychological deficits from both

pure forms of the disorder in its comorbid form, suggesting co-occurrence of

10

distinct neuropsychological or cognitive causes (Nigg, Hinshaw, Carte, &
Treuting, 1998).

The recent division of ADHD into three differential diagnoses
(predominantly Hyperactive/Impulsive, predominantly lnattentive, and Combined
type) also points to the utility of better characterizing the neuropsychological
mechanisms of the disorder as a way to resolve diagnostic ambiguities. For
example, the inattentive subtype may be a fundamentally distinct disorder in
cognitive terms (Cantwell & Baker, 1991). In order to study the relationship
between ADHD and its comorbid disorders it is crucial to consider possible
differences between ADD-Predominantly lnattentive Type (ADD/l) and ADD-
Predominantly Hyperactive/Impulsive (ADHD) or combined types.

For instance, some propose that the inattentive subtype is a later
developmental manifestation of the same disorder, while others propose that the
combined and inattentive types are distinct disorders with different underlying
causes (Hart, Lahey, Loeber, Applegate, & Frick, 1995; Nottleman & Jensen,
1995; Goodyear & Hynd, 1992; Biederman et al., 1992). Although not the focus
of this study, a brief discussion of the predominantly inattentive subtype of ADHD
further underscores the possible utility of better defining neuropsychological
mechanisms of ADHD as one way to resolve nosological problems.

Predominantly lnattentive ADHD. The distinction between the inattentive
and combined types of the disorder appears well-validated. Approximately sixty
percent of neuropsychological studies examining the differences between ADD/l
and ADHD have found clear differences (Goodyear & Hynd, 1992). Family
studies are also supportive— ADHD has been more commonly associated with
paternal ADD and hyperactivity and maternal substance abuse, while ADD/l is
more commonly associated with maternal anxiety disorders and Ieaming

disabilities in siblings (Biederman et al., 1992). Finally, Goodyear and Hynd

11

(1992) suggest different cognitive deficits in the two subtypes, with an anterior-
based right frontal inhibitory and attentional deficit in ADHD and a right posterior-
based attentional dysfunction in ADD/I. Given the evidence, then, it is surprising
that neuropsychological theories have not further elaborated or tested an
account of subtype. The application of current neuropsychological theories of
ADHD to an account of its subtypes may facilitate understanding of the
disorders underpinnings, as would the accurate differentiation of ADHD from its
comorbid diagnoses.

Summam . In summary, the DSM's failure to account for mechanisms of
disorder in ADHD has created within group symptom heterogeneity that could be
improved through use of a neuropsychological perspective. It is unclear
whether neuropsychological deficits in ADHD create vulnerability to further
impairment in the form of comorbid diagnoses, or whether distinct
neuropsychological impairment produces symptoms of ADHD leading to (a)
phenocopy and/or (b) dual diagnoses. Application of neuropsychological
theories to this comorbidity issue may prove useful in elucidating it. For
instance, different constellations of neuropsychological deficits may characterize
ADHD individuals with different comorbidities. The current study focused on the
combined rather than inattentive type of the disorder and controlled for comorbid
diagnosis as an initial step in addressing these issues. Discussion now moves to
a review of neuropsychological theories applicable to the combined type of
ADHD.

Neuropsychological Theories of ADHD
According to early neuropsychological theories, ADHD resulted from
over-arousal of the central nervous system. In the 1970's, however, many
studies of the psychophysiological characteristics of ADHD children found just

the reverse-- that hyperactive children exhibit slower responses to stimulation

12

(Sandberg & Barton, 1996). Cognitive theories concerning problems in rule-
governed behavior, self-regulation of arousal, and sensitivity to reinforcement
also arose in the 1970's and 1980's (Barkley, 1990). In the 1990's, theories have
focused on findings of executive dysfunction and frontal-striatal impairments in
the disorder.

Today, there are at least four major neuropsychological explanations of
the deficits in ADHD. As yet, however, little prospective, hypothesis-driven
research has explored their validity. These theories differ in terms of the
proposed mechanism underlying ADHD, so neuropsychological studies can help
to determine their relative merit. The theories can be generally organized
according to: (a) the area of the brain and (b) the primary deficit believed
involved in producing ADHD symptomatology. For heuristic purposes, I term
these the: Frontal Inhibition, Anomalous Dominance, Left Hemisphere, and Right
Hemisphere theories of ADHD. The current study focused on the Frontal
Inhibition and Right Hemisphere theories, which will now be reviewed. Review of
the Anomalous Dominance and Left Hemisphere theories can be found in
Appendix A.

Barklexs Frontallnhibition Theorv

Executive Functions. Barkley’s Frontal Inhibition theory was developed, in
part, to explain the multitude of findings suggesting executive function deficits in
ADHD. A brief discussion of executive function findings in ADHD is thus
discussed first. As seen from a neuropsychological perspective, many functional
processes fall under the rubric of executive functions, including set shifting,
planning, inhibition, and working memory. Primarily concerned with goal-directed
and problem solving behavior, executive functions are thought to play a role in a
wide range of adaptive and goal-directed behaviors (Pennington, 1997). In one

formulation, executive functions are believed responsible for context-specific

13

action selection, or the ability to select and implement appropriate action in the
face of numerous competing inappropriate actions (Pennington & Ozonoff,
1996). In the tradition of theories drawing links between ADHD's
symptomatology and similar effects of acquired brain damage, executive
functioning mechanisms are thought to be localized primarily to the prefrontal
cortex and/or associated thalamic and subcortical striatal areas (Pennington &
Ozonoff, 1996; Casey et al., 1997).

As Pennington and Ozonoff (1996) point out, however, the "frontal
metaphor" of executive functioning has proven conceptually problematic,
especially in childhood disorders, because it is "usually applied to patients whose
structural or functional neuropathology, if any, is unknown" (p. 51). As such,
executive functioning is subject to the criticism of serving as a post hoc
homunculus for an area of cognition still not well understood-- an area that
cognitive psychologists believe must occur between perception and action but
whose mechanisms remain insufficiently defined (Pennington & Ozonoff, 1996;
Pennington, 1997). Until these mechanisms are better identified, the executive
functioning explanation remains promising but limited.

Another problem with research in executive functioning in ADHD has been
that the performance of ADHD children on various executive function tests is not
uniform. For instance, studies with the Wisconsin Card Sort Test have yielded
inconsistent positive findings and small effects in the ADHD population, whereas
with other tests such as the Tower of London or Hanoi, the effects are larger and
more consistent (Pennington & Ozonoff, 1996). The inconsistency of these tests
has caused them to be of little clinical diagnostic value. Clearly, what is needed
is a theoretical perspective to drive and focus these studies in executive
functioning. Barkley's (1997) model is notable in its attempt to organize and

synthesize the multitude of executive function findings.

14

Front_a_l Inhibition Theorv. Building on the general theories of cortical brain
operation by Bronowski (1977) and Fuster (1989), Barkley (1997) proposed the
main deficit in ADHD to stem from failures in behavioral inhibition. Functionally
localized bilaterally to the prefrontal and motor cortex, behavioral inhibition allows
a delay in motor activity for further direction from the executive functions.
Barkley categorizes these executive functions into four main categories: working
memory, internalization of speech, self-regulation of affect-motivation-arousal,
and reconstitution. Failures in behavioral inhibition prevent the necessary delay
in action and lead to executive functioning deficits, illustrated by difficulties in
self-control and goal directed behavior. As a result, ADHD children exhibit
behavior influenced more heavily by immediate context than do others. For
example, Barkley predicts that ADHD children should show greater emotional
reactivity in response to contextual events, but no deficits in perception of
emotion (due to its nonexecutive nature).

In general, the Frontal Inhibition theory predicts that the more inhibitory
control required in the performance of a task, the greater the impairment
produced in ADHD individuals. As such, simpler tasks may not produce
evidence of the executive function deficits associated with more complex tasks.
For instance, tasks assessing simple motor speed, as in the grooved pegboard,
are suggested to be less impaired in ADHD than tasks requiring complex
sequences of movements. Barkley also predicts that, similar to deficits found
following prefrontal lobe lesions, ADHD individuals should have difficulty creating
numerous alternative response sequences in both verbal and nonverbal
domains. For example, deficits in tests of both verbal and figural fluency are
thus predicted (Barkley, 1997).

Inherent to the Frontal Inhibition theory is the suggestion that the

inattention found in the combined type of ADHD results from secondary

15

impairments stemming from poor behavioral inhibition and interference control
rather than from primary attentional deficits. Barkley draws upon sustained
attention and vigilance research (to be discussed in the review of the Right
Hemisphere theory) to postulate a developmental delay in the development of
self-regulatory mechanisms underlying sustained attention or persistence.
Barkley distinguishes this attentional deficit from that found in predominantly
inattentive children. The latter, he suggests, is a selective attention deficit
unrelated to inhibitory control. This claim further underscores Barkley’s belief
that combined type ADHD children exhibit a specifically frontal inhibitory deficit
and are differentiable from predominantly inattentive children.

The Frontal Inhibition theory is relatively new, so little empirical research
has directly put it to test. Mariani and Barkley (1997) assessed aspects of
neuropsychological functioning in ADHD and normal preschool children over a
number of domains and found that ADHD children were impaired on two of four
examined factors— in “motor control” and “working memory-persistence”, but not
in “verbal Ieaming-memory” or “picture identification-factual knowledge domains”.
Significant differences were found between ADHD and control children on the K-
ABC Hand Movements task and the Purdue Pegboard (with both hands only) in
the motor control domain and on the K-ABC spatial memory and number recall
tests in the working memory domain.

Barkley suggests that ADHD results from problems in neuropsychological
functioning particularly within the prefrontal cortex. Barkley does not specify the
anatomical pathways presumed to produce behavioral inhibition deficits.
However, Fuster's (1989) description of attentional difficulties and hyperactivity
following lesions to the dorsolateral prefrontal cortices may best apply to the
cognitive and behavioral symptoms of ADHD. Alternatively, Casey et al. (1997)

have obtained brain imaging data suggesting that orbito-prefrontal regions are

16

affected. The Right Hemisphere theory of ADHD also focuses on frontal lobe
functioning. Unlike the Frontal Inhibition theory, however, this theory examines
frontal functioning from the perspective of Iateralized deficits.

The “Right Hemisphere” Theorv

Lateralized deficits in ADHD may involve developmental alterations in
neural mechanisms that produce functional effects similar to those found
following unilateral brain damage. The Right Hemisphere theory examines this
possibility by proposing a predominantly right hemisphere dysfunction that is
localized primarily to the frontal lobes and its subcortical connections. Due to
this theory’s primary focus on the existence of a Iateralized impairment in ADHD,
a brief digression is made here to cover background on Iateralized functions
within the frontal lobes. For further details on brain Iateralization and its
development see Appendix B.

Lateral Sjeciflzation of the Fronta_l Lobes . According to Kolb and
Whishaw (1990), asymmetry in the frontal lobes is greatest in the prefrontal
cortices. As noted in the executive functioning literature, the prefrontal cortices
are considered to play a role in planning and problem solving (Pennington,
1997). Right hemisphere specialization for holistic processing and left
hemisphere specialization for analytic and sequential processing can be
theorized to be controlled by the prefrontal cortices of their respective
hemispheres. Furthermore, imaging studies in normal populations have
implicated right frontal regions in both sustained and selective attention as well
as episodic memory, while left frontal regions have been implicated in semantic
memory and language production (Cabeza & Nyberg, 1997).

Although the range of abilities affected by frontal lobe lesions is wide-
ranging, one consistent finding is that drastic changes in personality follow

lesions of the frontal lobes. Pennington (1997) reviews several of the social

17

deficit syndromes that follow frontal lobe damage. In particular, a
pseudodepressive syndrome marked by reduced awareness, lack of initiative,
unconcem, and blunting of emotional response has been associated with lesions
to the left medial prefrontal cortex or to the anterior convexity and frontal poles.
Lesions of the right orbital prefrontal cortex produce a pseudopsychopathic
syndrome marked by sporadic hypomania, childish humor, disinhibition of sexual
and eating behavior, lack of concern for others, and disregard for ethical
principles. Also associated with medial prefrontal lobe lesions (possibly in the
left hemisphere) is akinetic mutism, a disorder characterized by a serious deficit
in speech initiation and other spontaneous behaviors. Pennington suggests that
these syndromes may be linked to certain developmental disorders such as
autism and conduct disorder (Pennington, 1997; Pennington & Ozonoff, 1996).
Frontal lesions also produce impairment in the ability to maintain attention
to a stationary or fixed stimulus. Right frontally- mediated failures in focused
attention predict left-sided neglect as well as failure to maintain focused visual
attention in the left visual field (Colby, 1991). The neurological phenomenon of
neglect, for instance, results more often in individuals with right hemisphere
lesions (Hellman & Van Den Abell, 1980). As such, the role played by the right
frontal lobe in ADHD gains particular importance. lf ADHD is the result of
underactivation in right frontal attentional networks (as proposed by the Right
Hemisphere theory), then those with ADHD should exhibit failures in the ability to
maintain focused attention in general (due to right hemisphere dominance in
attention mechanisms) and, in particular, to the contralateral visual field. Failure
to maintain focused attention might also explain hyperresponsiveness to
environmental stimuli in that hypoarousal of frontal mechanisms could lead to
hyperarousal of posterior mechanisms (Tucker & Williamson, 1984). Anecdotal

reports of high rates of ADHD in those with nonverbal or right hemisphere

18

Ieaming disabilities further support the connection between right hemisphere
dysfunction and ADHD symptomatology. For instance, in a sample of twenty
children with nonverbal Ieaming disability, all also received a diagnosis of ADHD
(Gross-Tsur et al., 1995).

Due to connections between behaviors exhibited in ADHD and its related
disorders and those occurring following damage to the prefrontal cortices, ADHD
may thus result from problems in neuropsychological functioning particularly
within the right prefrontal cortex.

The Theory. Recently, some studies have examined the role of right
hemisphere functioning in ADHD. Voeller and Heilman (1988) observed that
ADD children with or without hyperactivity behave very similarly to adults with
right-hemisphere deficits, showing evidence of attentional disturbances through
the manifestation of left-sided neglect. This neglect was demonstrated through
the commission of more errors, particularly on the left side, in a letter
cancellation task. Though selected only according to their ADHD status, these
subjects also showed more left-sided neurological soft-signs. Heilman, Voeller,
and Nadeau (1991) suggest that their finding of neglect in ADHD children may
be explained through problems in right hemisphere mechanisms of attention,
arousal and motor activity. Findings of larger response inhibition deficits in
individuals with right hemisphere lesions were also drawn upon to support the
notion of a right hemisphere deficit in ADHD.

Voeller and Heilman's (1988) findings of neglect in ADHD children have
remained unreplicated. In fact, a later report by Voeller (1991) suggested that
this finding was attributable to a predominantly inattentive sample. A finding of
left-sided neglect was replicated by Malone, Coutis, Kershner, and Logan (1994)
in a letter-cancellation task, but was attributed to those ADHD children with

comorbid learning disabilities. As DeLong (1995) points out, however, symptoms

19

of inattention in ADHD may create LD symptomatology. Goodyear and Hynd
(1992) propose dysfunction in right posterior mechanisms to underlie the
predominantly inattentive type of ADHD and suggest that an increased incidence
of LD diagnosis exists in this group. Thus, a distinct group of children with right
posterior deficits may exhibit inattention symptoms which lead to behavioral
manifestations of learning disability but that do not result from the primarily left
posterior mechanisms underlying pure LD. These findings further underscore
the need to clearly distinguish between pure LD and predominantly inattentive
samples—especially when examining Iateralized deficits.

Motor Dysfunction . Some evidence suggests that ADHD children show
impairments in gross motor output (Denckla et al., 1985; Carte, Nigg, & Hinshaw,
1996). Heilman, Voeller, and Nadeau (1991) assert that right frontal deficits
produce failures in motor persistence in ADHD children compared to controls.
For example, the right frontal lobe and right striatum are hypothesized to be
important in the inhibition of unwanted responses to stimuli. Furthermore,
secondary motor areas controlling the nondominant side of the body are
proposed to modify movements in response to environmental or contextual
feedback. Thus, right hemisphere-mediated deficits in response inhibition and
motor control may lead to an inability to adequately plan motor responses or
sets. Heilman et al. concluded that ADHD results from a right hemisphere-
mediated, frontal-striatal deficit that prevents ADHD children from properly
responding to environmental stimuli and controlling their responses. These
inappropriate responses are manifested through their inattention, neglect, and
failure to inhibit behavior (Heilman et al, 1991).

Imaging Studies . Consistent with Heilman et al's theory (1991), imaging
studies in ADHD have largely supported the notion of a right frontostriatal

impairment in the disorder. Structural MRI studies have suggested a decrease in

20

size of the right prefrontal cortex in ADHD (Castellanos et al, 1996; Casey et al,
1997; Hynd et al., 1990), along with decreased arousal or activation (as
measured by EEG) in the right hemisphere (Crawford & Barabasz, 1996). In
support of the implications of right frontal-striatal Impairment, other studies have
found evidence of right striatal impairment in the disorder. Lou et al. (1989),
measuring cerebral blood flow, discovered bilateral hypoperfusion of the
striatum, which, for the right striatum, could not be reversed with
methylphenidate administration. Furthermore, a difference in anatomical size
has been found in the caudate nucleus of children with ADHD. ADHD children
exhibited a significant loss in the typical right-greater-than-left, MRI-measured,
asymmetry pattern in the caudate nucleus (Castellanos et al, 1996) as well as
failure to exhibit the normal size decrease with age of the right caudate nucleus
(Casey et al., 1997; Castellanos et al, 1996). Decreases in right globus pallidus
volume were also found (Casey et al., 1997; Castellanos et al, 1996). Hynd et
al. (1993) proposed an opposite pattern of caudate deficit in ADHD, claiming that
a left-less-than- right pattern of asymmetry was in contrast to its typical left-
greater-than-right activation. The reason for Hynd et al's findings of a different
asymmetry pattern in their control group is unclear. However, the use of females
in the study in addition to a much smaller number of subjects may explain these
contrasting results.

Findings by Casey et al. (1997) that abnormal brain regions in ADHD
(comprising the prefrontal cortices, globus pallidus, and caudate nucleus as

discussed above) were correlated with deficits on response inhibition tasks may

21

provide the first direct link between structural right hemisphere abnormalities and
functional deficits. More importantly, Casey et al. controlled for subtypes, adding
further support to this finding in the combined type of ADHD. Lending further
support to findings of right hemisphere anatomical differences in ADHD children
are a limited number of cognitive and behavioral studies that find specifically
right hemisphere-related impairments. These studies have focused primarily on
visual attention and motor functioning in ADHD.

_Deficits in Vigilance and Attention . In an extensive review by Weinberg
and Harper (1993), vigilance-- defined as "the state of being alert, awake, and
watchful (p. 59)"- was asserted to be a function of the right hemisphere,
probably the inferior parietal lobe or posterior parietal cortex. Deficits in vigilance
are characterized by symptoms such as daydreaming, loss of attention to detail
and failure to attend to all but novel stimuli. Those with deficits in vigilance often
exhibit extreme motor restlessness in their attempts to maintain alertness.
Deficits in vigilance are hypothesized to occur through dysfunctional activation of
the reticular activating system resulting from the right parietal lobe's release of
inhibition. When dysfunction in the right parietal lobe causes a release of this
activation, the reticular activating system puts the brain to "sleep.” Primary to the
activation of this pathway are the biogenic amines (norepinephrine, dopamine,
and acetylcholine), which play a role in wakefulness and arousal. Stimulant
medications, which increase biogenic amine activity, are often used in the
treatment of vigilance disorders. Their stimulation of the parietal lobe was

posited to re-establish inhibition of the reticular activating system.

22

According to Weinberg and Harper (1993), numerous disorders can be
linked to "secondary hypovigilance." Among these disorders are Ieaming
disabilities, particularly right hemisphere Ieaming disabilities, and narcolepsy,
which may result from an abnormality in the reticular activating system of the
brainstem. This abnormality leads to the production of rapid eye movement
(REM) activity that overrides the right brain's vigilance functions. Also implicated
in causing secondary hypovigilance are depression, mania, conduct disorder and
the neglect syndrome resulting from right hemisphere damage. Weinberg and
Harper suggest that the hypovigilance associated with these disorders also
creates the symptom complex of ADHD. Correct identification of hypovigilant
symptoms in disorders frequently co-occurring with ADHD will therefore produce
favorable outcomes in ADHD symptomatology. Due to the association between
right parietal functioning and vigilance disorders, if symptoms of ADHD are
produced by dysfunction in this vigilance system a global right hemisphere deficit
in the disorder could be suggested.

Visual Attention. Evidence for failures in vigilance and sustained attention
in ADHD have resulted from studies of visual-spatial attention. One study
concluded that ADHD children fail to show the typical right hemisphere
advantage in eye movements to left visual field stimuli under non-cued
conditions (Rothlind, Posner, &Schaughency, 1991). Instead, they showed no
Iateralized effect of orienting. The authors suggested that ADHD subjects
exhibited a right hemisphere- mediated failure to sustain attention.

Consequently, they responded similarly to targets in either visual field in fixation-

23

absent and present conditions. A similar result was found in an exogenously
cued covert orienting task. ADHD subjects exhibited a slowed movement of
attention to the left visual field when an unwamed target appeared, and was
attributed to possible hypoarousal of the right hemisphere (Nigg, Swanson, &
Hinshaw, 1996). In addition, Carter et al. (1995) posited a right hemisphere
based deficit in controlled (endogenous) visual-spatial attention processes,
based on the observation of reduced costs on orienting to invalidly cued left
visual field targets. They proposed that impairment in right hemisphere frontal-
striatal pathways produced this effect.

Evidence from visual attention studies thus provides some support for the
hypothesis that ADHD results from a right hemisphere dysfunction. Such a right
hemisphere deficit could occur somewhere in the course of development, linking
ADHD to atypical development of lateralization. Thus, further examination of
right hemisphere functioning in ADHD children may prove useful.

Comparison of the Theories:

Both the Frontal Inhibition and Right Hemisphere theories are similar in
their descriptions of functional impairments in ADHD. For instance, both theories
propose primary symptoms of ADHD to involve deficits in response preparation,
response inhibition, and sensitivity to errors. Such deficits are believed to make
ADHD children less able to sustain internally driven behaviors over time.

Where the two theories differ, however, is in the pathways proposed to
produce this functional deficit. The Frontal Inhibition theory proposes that a

primary deficit in behavioral inhibition (linked to orbital-frontal and dorsolateral

24

regions of the prefrontal cortex and their striatal connections) either directly
causes motor control/output deficits or causes executive functioning failures,
which then produce motor output/control deficits. Thus, in the Frontal Inhibition
theory, failures in behavioral inhibition are seen as the driving factor in the
production of other ADHD-associated deficits.

The Right Hemisphere theory, on the other hand, does not propose a
behavioral inhibition mechanism to account for ADHD symptoms. Instead, it
focuses on links between ADHD-related impairments and “right hemisphere
functions” (as documented in brain-injured populations and brain imaging
studies). Unlike the single mechanism proposed by the Frontal Inhibition theory,
the Right Hemisphere theory identifies three right hemisphere-mediated systems
that are disrupted in ADHD: sensory-attentional, motor-intentional, and arousal-
activation systems. Impairment in these systems is proposed to occur mainly
through disrupted connections between right prefrontal/medial frontal and other
distributed neural networks throughout the right hemisphere (e.g., striatal,
parietal, and limbic areas).

Perhaps most analogous to Barkley’s behavioral inhibition system is
Heilman and Voeller's “motor-intentional” system, which is proposed to play a
role in motivational and motor system functioning through a right dorsolateral
prefrontal cortex-striatal pathway that is mediated by the medial premotor cortex
(SMA). Voeller (1991) suggests that those with dysfunctions in this pathway may
represent an “anterior” group with motor impersistence, impulsivity, and deficits

in response inhibition. Despite the similarities between these two systems and

25

their presumed anatomical substrates, the two theories continue to differ in that
the Frontal Inhibition theory proposes behavioral inhibition to affect functioning
across a range of domains (e.g., Barkley's four executive function domains). In
contrast, the Right Hemisphere theory’s “motor-intentional system” focuses
solely on intentional behaviors and is only one of multiple systems that may be
Impaired in ADHD. Clearly, further testing of these two theories remains
necessary to clarify the nature of impairment in ADHD.

Traditional neuropsychological measures may provide one means by
which to evaluate these two theories. They are in wide clinical use, economical
to administer, and generally have validating literatures in brain injured
populations. They, therefore, can suggest underlying mechanisms. The
application of such measures to validating the construct of ADHD may help to
clarify the nature of neuropsychological and cognitive dysfunction in the disorder.
More importantly, it may help to differentiate core ADHD deficits in its subtypes
and distinguish these from similar symptoms associated primarily with comorbid
behavior and mood problems. The current study sought to take an initial step
towards these goals by examining neuropsychological deficits in the combined
type of the disorder when comorbid symptoms were controlled.

Several tasks were used to assess aspects of the two theories: fluency
tasks, simple and complex motor tasks, competing left-right output tasks, and
coverage of different output modalities (verbal, visuo-motor, and motor). Tests
designed to examine the Frontal Inhibition theory were selected according to

Barkley’s rationale that behavioral inhibition deficits should produce impairments

26

across a range of executive function and motor tasks. Content-matched control
tasks were paired with every executive function task (according to Denckla’s
[1996] suggestion). Denckla (1996) asserted that, “In each case the more EF
element demands more inhibition of prepotent automatic or overleamed
response(s) and more active, on-Iine, rule-govemed, future-oriented, goal-
oriented, time-limited response preparation than does the control task” (274).
Thus, tasks requiring greater executive functioning direction should show greater
deficits than otherwise similar but simpler tasks, presumably due to inhibitory
failures. Therefore, although inhibitory deficits were not studied directly,
comparison of EF tasks with content-matched simpler tasks was believed
capable of indirectly testing Barkley’s theory.

To examine EF impairment in the verbal domain, a verbal fluency and
vocabulary task were compared. A figural fluency task also examined fluency
deficits. According to Denckla (1996), fluency tasks are rule-govemed and
constraint-limited and, therefore, provide a good measure of executive function.
These tests are sensitive to working memory deficits and assess maintenance of
preparatory set, memory search efficiency, and rapid response output. Barkley
(1997) suggests that fluency tasks also require the creation of multiple novel
alternative response sequences and thus tap the EF domain of reconstitution.
The Frontal Inhibition theory predicts that inhibitory failures should prevent
operation of working memory and reconstitution operations necessary for optimal
performance on this task, and should thus result in fluency impairment,

manifested in fewer responses and increased repetition of non-unique responses

27

(i.e., perseveration). To differentiate EF from non-EF verbal impairments,
Denckla suggests assessing discrepancies between verbal fluency and
vocabulary performance.

A simple and complex drawing task were used to examine executive
function deficits in the visuo—motor domain. Barkley (1997) proposed working
memory to be involved in the development of organization and reproduction
strategies for complex visual-spatial material. Inhibitory deficits are proposed to
prevent temporal delays necessary for strategy formation and to impair ability to
mentally hold and manipulate aspects of the stimulus figure. Figure
reproduction, consequently, is likely to become more impaired with increasing
figure complexity. As impairment increases, drawings should demonstrate
poorer planning and organization, as well as overall poorer reproduction quality.
The Rey-Osterrieth Complex Figure was paired with a simpler drawing task to
examine the effect of increasing figure complexity on task performance.

The final comparison made for the Frontal Inhibition theory was in the
pairing of simple and complex motor tasks. Barkley (1997) asserts that inhibitory
and executive function deficits result in problems with motor presetting and
persistence, flexibility, failure to inhibit extraneous movements, and motor
overflow—all of which lead to motor slowing. These motor problems are
predicted to occur as a result of inhibitory and EF failures and are not believed to
result from specific impairments in premotor areas. As with other EF—dependent
tasks, Barkley predicts motor performance in ADHD children to become

increasingly impaired as motor demands increase. Thus, the more automatic

28

the behavior to be performed, the less likely that inhibitory failures will produce
deficits in task performance (Denckla, 1996). As motor behaviors require
increasing effort and persistence, susceptibility to interference created by
inhibitory failures also increases.

In order to examine the Right Hemisphere theory, content-similar tasks
with differential sensitivity to right versus left hemisphere functioning in brain
injured populations were used. Verbal and figural fluency tasks were used to
examine left versus right prefrontal functioning, respectively. Motor output tasks
(grooved pegboard and Denckla’s Time to Do 20 motor battery) were used to
examine Iateralized differences in motor ability. As is traditional in
neuropsychological assessment, slowed motor performance on one side of the
body was considered indicative of impairment in the contralateral hemisphere.

The Right Hemisphere theory proposes deficits in motor control to reflect
difficulty in gating sensory inputs Into motor output (involving deficits in right
frontal-striatal pathways), rather than in production of movement per se (left
hemisphere-produced apraxia). Voeller and Heilman (1988) assert that lesions
to the right hemisphere impair modification of movements according to
environmental cues. Therefore, decreased sensitivity to errors, motor
impersistence, and overflow movements all are proposed to slow motor output.
This motor slowing is considered indicative of dysfunction in the pathway

between the right dorsolateral prefrontal cortex and striatum.

29

Current Study

General Hypotheses: (specific predictions follow methods)

1. Frontal Inhibition Theory: ADHD involves a bilateral prefrontal deficit in
behavioral inhibition. ADHD children should thus exhibit deficits in tasks
assessing both right and left frontal functioning. Barkley predicts that there
should be deficits in both verbal and figural fluency due to difficulties in
producing alternate response sequences in verbal and nonverbal domains.
According to Barkley, the more complex a task, the more likely it is to be
impaired by inhibition deficits. For example, the Frontal Inhibition Theory would
predict deficits in: complex motor movements but not simple ones; verbal tasks
that require originality but not vocabulary tasks; and complex visual tasks
requiring organization but not simple visuo—spatial tasks. The Frontal Inhibition
theory applies only to children exhibiting both hyperactive and lnattentive
symptoms (ADHD). Consequently, hypothesis-testing was limited to those with

the combined type of the disorder.

2. Right Hemisphere Theory: Numerous anatomical and cognitive findings in the
ADHD literature suggest a right hemisphere deficit. The nature of this deficit
appears to involve primarily hypoarousal of right frontal-striatal networks. ADHD
children are thus predicted to show deficits on tasks relying upon right frontal
cortex, like figural fluency and left-sided motor programming. Voeller (1991) later
suggested that only ADD/l children exhibited right posterior deficits. As such,
hypothesis testing was limited to exploration of right frontal deficits in children

with the combined type of the disorder.

30

METHODS
Participants

Children were selected from volunteers recruited through the Lansing and
East Lansing public schools, a local pediatric clinic, and an ADHD parent support
group. Children with neurological impairments, other serious medical or
psychiatric conditions, Full Scale IQ below 70, or uncorrected visual or hearing
impairments, as well as those whose first language was not English, were
excluded from the study.

Qagnostic Criteria. For assignment to the ADHD group children meta
series of diagnostic criteria designed to avoid false positives. To be included in
the ADHD group children exceeded the empirically established diagnostic cut-
offs for: (1) the SNAP-IV by parent report (Pelham, Gnagy, Greenslade, & Milich,
1992), (2) the parent and teacher Child Behavior Checklist attention problems
subscale (Achenbach, 1991), or (3) the parent and teacher Conners Abbreviated
Symptom Questionnaire (Conners, 1997). After meeting these cut-offs,
diagnosis was formally made by the structured Diagnostic Interview Schedule of
Childhood for DSM-IV (DISC-IV; Shaffer et al., 1993), which specifically
assessed DSM-IV criteria for ADHD. ADHD diagnosis required that symptoms
had persisted for more than six months, been present in both home and school
situations, and appeared before seven years of age. The comparison group met
all inclusion criteria but was below cut-off on all ADHD measures. Comorbid
diagnoses were not excluded but were assessed to enable statistical control.

Among the comorbid disorders assessed were oppositional defiant, conduct

31

disorder, generalized anxiety disorder, obsessive-compulsive disorder, major
depression, and tics.

Description of Groups. The subjects were 55 children selected from a
larger sample of 73 children. Of these 73 children, 12 were excluded due to a
diagnosis of predominantly inattentive ADHD (ADD) and 6 due to factors such as
low IQ or presence of a confounding psychological disorder such as PTSD. The
full sample, however, was used to evaluate the reliability of the various measures
and for missing data mean replacement. Of the final sample (N=55), 78.2
percent were Caucasian (N=43), 10.9 percent African American (N=6), 7.3
percent Latino/Hispanic (N=4), and 3.6 percent Asian American (N=2).

Participants in the final sample were 29 ADHD (9 girls and 20 boys) and
26 non-ADHD control (12 girls, 14 boys) children with a mean age of 9.8 i 1.6
years. There were no significant age differences between the groups, with the
age of ADHD children ranging from 6.3 to 12.9 years (mean age: 9.7 1 1.9 years)
and that of non-ADHD controls from 6.9 to 13.3 years (mean age: 9.9 1- 1.2
years). The ADHD group did, however, have a lower mean IQ than the control
group [F(1,53)=5.54, p=.02] (Table 1). Three ADHD children met additional
criteria for a diagnosis of conduct disorder and four for a Ieaming disability. As
expected, groups differed on parent and teacher behavioral ratings of inattention,

hyperactivity, and oppositional/aggressive symptoms (See Table 1 below).

32

Table 1. Description of groups.

Comparison of mean age, estimated IQ, and ratings of hyperactivity, inattention,
ODD and CD symptoms for ADHD compared to control children by parent and
teacher report on the SNAP-IV. Mean SNAP ratings are out of a possible 3.0

 

 

Control ADHD Significance
(mean1SD) (mean1SD)
Age (months) 118.9 1 13.8 115.8 1 22.4 Nonsignificant
Estimated IQ 111.5 1 14.6 102.1 1 14.8 F(1,53)=5.54,
p=.02
SNAP Hyperactivity
Mother: .39 1 .46 1.9 1 .62 F(1,53)=92.70,
(n=25) (n=29) p5.001
Father: .32 1 .42 1.6 1 .77 F(1,53)=35.57,
(n=22) (n=19) p_<_.001
Teacher: .30 1 .57 1.18 1 .65 F(1,53)=18.78,
(n=21) (n=21) p_<_.001
SNAP lnattention
Mother: .59 1 .49 2.0 1 .69 F(1,53)=67.94,
(n=25) (n=29) p5.001
Father: .49 1 .42 1.69 1 .76 F(1,53)=34.24,
(n=22) (n= 19) p_<_.001
Teacher: .37 1 .41 1.94 1 .61 F(1,53)=63.01,
(n=21) (n=21) p5.001
SNAP ODD
Mother: 43 1 .46 1.571 .72 F(1,51)=44.81,
(n=24) (n=29) p5.001
Father: .39 1 .42 1.471 .82 F(1,39)=29.04,
(n=22) (n=19) p5.001
Teacher: .32 1 .66 .80 1 .70 F(1,40)= 5.12,
(n=21) (n=21) p=.03
SNAP CD
Mother: .07 1 .17 .68 1 .70 F(1,52)=18.22,
(n=25) (n=29) p5.001
Father: .05 1 .13 .54 1 .68 F(1,39)=10.90,
(n=22) (n=19) p=.002
Teacher: .05 1 .22 .33 1 .41 F(1,39)= 7.32,
(n=21) (n=20) p=.01

 

SNAP= Swanson, Nolan, and Pelham-IV checklist

CD=Conduct Disorder

ODD=Oppositional Defiant Disorder

33

Procedure

Children completed the research battery in Michigan State University
psychology department during visits to campus. They assented to the testing
procedure and were administered the tests in a standardized order; frequent
breaks were interspersed within the testing session. Those children taking short-
acting stimulant medications were tested, when possible, after a 24-hour
washout period. The assessments were administered by graduate students in
clinical psychology who were blind to formal diagnostic status in most cases.
Following the completion of the testing session children were permitted to
choose a prize and play computer games. Parents recruited from the community
were paid; those from a local ADHD clinic received clinical feedback regarding
their child’s test performance if they so desired.
Measures

Control Variables

(a)lntelligence: WISC III Vocabulary. and Block Design subtests.
Estimated lQ's were obtained from the vocabulary and block design subtests of
the Wechsler Intelligence Scale for Children (W lSC-lll) using procedures
documented by Sattler (1992). This short form of the WISC has been found to
have excellent validity (>90) in relation to the full battery (Sattler, 1992). For
purposes of clinical feedback, those children recruited from the pediatric clinic

often were given between four and six WISC subtests to estimate IO.

34

(b)Leaming Disabilities: Due to high rates of overlap between behaviors
shown by Ieaming disabled children and symptoms of ADHD, it is important to
control for reading disorders when examining an ADHD population.

The Woodcock Johnson Word Attack subtest (Woodcock & Johnson,
1989,1990) were used to measure phonological decoding ability, or the ability to
analyze unfamiliar nonsense words according to familiar letter-sound
correspondences. It consists of 32 nonwords that participants pronounce aloud.
Due to the difficulties in phonological decoding demonstrated by children with
learning disabilities, this task is believed to provide a good measure of reading
disorder. Internal consistency (split-half reliability) of this measure ranges from
.88 to .95 in children (N=908) between the ages of 6 and 18.

The Wechsler Individual Achievement Test (WIAT) Reading and Spelling
subtests (The Psychological Corporation, 1992) were also administered as part
of the assessment of reading and spelling abilities. These subtests require
participants to read and write words of increasing difficulty until a discontinue rule
is met. LD diagnosis was made in the presence of a significant discrepancy
between reading or spelling performance and estimated IO (15 or more points in
standard scores) combined with an absolute reading level less than 85. Average
split-half reliabilities for children between the ages of 5 and 17 (N=3899) on the
reading and writing subtests are .92 and .90 respectively.

(c)Handedness : Due to the incidence of reversed lateralization
(especially for language) in a subset of left handers, controlling for handedness

is important when assessing Iateralized processes. Eight questions about hand

35

use and two about foot use were taken from the Edinburgh Handedness
Inventory and utilized to obtain an index of hand/foot dominance for each child.
Children were asked to demonstrate each act and then for each act were asked
to rate whether they: always use their right/left hand, usually use it, or use both
hands to complete them. Footedness was determined by child demonstration.
In one study (N=735 adults), test-retest reliability on items included in this
questionnaire ranged from .42 to .80 (kappa coefficient) with 77 to 95 percent
agreement across tasks over the 18 month test interval (Ransil & Schachter,
1994)

Bishop, Ross, Daniels, and Bright (1996) have suggested that no
significant difference exists between right handers who use their right hand
exclusively (answering "always" on at least 80 percent of the items) and those
who are slightly more weakly right handed (answering "usually" on several
items). However, for purposes of comparison determination of strength of
handedness was made by averaging responses across the eight tasks
assessed. “Always” using the right hand/left hand was assigned a value of +2l-2
respectively; “usually” using the right/left hand was assigned a value of +1l-1;
and using “both” hands was assigned a value of 0. The average of these scores
served as an indication of magnitude of hand dominance and enabled
handedness to be covaried in secondary analyses.

Neuropsychological Measures
Selected tasks have received previous empirical support for their

measurement of primarily right or left hemisphere functioning. Though these

36

tests show sensitivity to Iateralized deficits in patients with acquired brain
damage, we still know very little about their sensitivity to normal variations in
lateralization or to deficits in developmental disorders. Nevertheless, these tasks
have been extensively validated in adults and are used with children in clinical
research and practice. Where possible, tasks in common clinical use were
selected to enhance the potential clinical utility of the findings. Although multiple
interacting neural proceSses contribute to performance on most of these tasks,
they provide a way to begin to evaluate current theories of ADHD. Several tasks
were needed to assess aspects of the various theories: fluency tasks (Frontal
Inhibition theory), simple and complex motor and drawing tasks (Frontal
Inhibition theory), and competing left-right output tasks (Right Hemisphere
theories). The following tasks were selected.

(a)FIuency Tasks :

Controlled Oral Word Association Test (Benton & Hamsher, 1978). The
Controlled Oral Word Association Test (COWAT) was used to assess verbal
skills and left hemisphere frontal processes. Participants were required to say as
many words as they could think of that began with the letters C, F, and L,
respectively, on three different trials of sixty seconds duration. Impaired verbal
fluency has been associated particularly with damage to the left frontal lobe
immediately anterior to Broca's area. Though right hemisphere damage may
also produce depressions in verbal fluency scores, left hemisphere lesions
produce much greater impairments. One study found patients with left frontal

lesions to produce one third fewer words than did patients with right hemisphere

37

lesions (Benton, 1968, as cited in Lezak, 1995). Furthermore, numerous studies
using brain activation measurement techniques have found associations
between verbal fluency scores and activation in left frontal areas (Pujol, Vendrell,
Deus, & Kulisevsky, 1996; Elfgren, Ryding, Passant, 1996; Cantor—Graae,
Warkentin, Franzen, Risberg, 1993). Based on the evidence suggesting primary
involvement of left frontal regions in this task, any deficits on this task in an
ADHD sample will be assumed to result from left hemisphere dysfunction.
Deficits on this task will be considered to provide evidence in support of the left
hemisphere hypothesis. Norms for children, from grades K through 6, have been
reported by Schum, Sivan, and Benton (1989), thus enhancing the use of this
task for a childhood population. As no reliability data for use of this task in
children could be found, reliability data will be obtained as part of this study.

In the current study, reliability estimates were obtained from 71 children in

the full sample using individual trial scores (for each of three trials). Reliability

across the three trials was adequate (or=.80).1 The total score, comprising total
words generated across trials, was used in all analyses.

The Ruff Figural Fluency Test (RFFT; Ruff, Light, & Evans, 1987). This
test was developed as a counterpart to verbal fluency tests and requires subjects
to connect 5 dots or less to create a pattern within a box. Subjects produce as
many unique patterns as they can in ninety seconds. Some evidence suggests

that individuals with right hemisphere damage, especially those with right frontal

 

1 For a subset of children (N=18) seen for a follow-up visit approximately one
year after initial participation, test-retest correlations for the total score (sum of
scores across three trials) were weak (r = .25, a=.43).

38

lesions exhibit the lowest productivity in this task when compared with other brain
damaged individuals (Ruff, Allen, Farrow, Niemann, & Wylie, 1994). In addition,
individuals with right frontal and right central lesions produce the greatest
amount of perseveration, drawing the same pattern multiple times, despite
instructions to draw only unique patterns and not to repeat (Jones-Gotman, 1991
as cited in Lezak, 1995). In adults (N=95) the RFFT has been found to have a
test-retest reliability of r= .76. In a sample of intellectually bright children (N=62)
a test-retest reliability of r=.80 was obtained. Figural fluency has been found to
be uncorrelated to measures of both motor speed and verbal fluency. As such, it
appears to measure an independent domain which appears to be reliant on right
frontal processes.

(b). Visuo-Spjtiaj Tasks :

The Rey-Ostern’eth Complex Figure Test (RCF 7', Meyers &Meyers,
1995). Patients with both left and right hemisphere damage have been found to
exhibit deficits on this task. However, the nature of their exhibited deficits differ,
and those with right hemisphere damage show more severe deficits. Differences
between the two groups is most apparent during the recall trial of the test, with
left hemisphere damaged patients showing the ability to reproduce the figure as
a gestalt and right hemisphere damaged patients continuing to show pooriy
integrated figures that are even worse than those produced in the copy trial.
Other evidence of predominantly right-hemisphere involvement in this task
includes the finding of left visuo-spatial (left side of the figure) neglect in those

with right hemisphere lesions (Lezak, 1995). Deficits on this task have been

39

found in ADHD samples (Grodzinsky 8. Diamond, 1992), although conflicting
results have been reported (Carte et al., 1996; Nigg et al., 1998). Due to the
complexity of this task, its performance is expected to rely upon multiple neural
systems. However, different types of deficits may reflect impairment in different
neural regions. For example, the strategy used to reproduce the drawing may
require reliance upon frontal mechanisms.

The RCFT requires subjects to first copy a complex unfamiliar picture and
then to draw it again from memory three minutes later. Performance was
measured according to the Taylor scoring system (Spreen & Strauss, 1991). In
the normative sample for the RCFT, inter-rater reliabilities between .93 and .99
were established using a scoring system based on Rey and Taylor's (Spreen &
Strauss, 1991) scoring system as described Meyers and Meyers (1995). In this
adult sample, test-retest reliability correlations ranging from .76 to .89 were
obtained for the different trials of the test, with delayed recall producing the
greatest reliability. Norms for the RCFT have also been established for this task
in a representative sample of children (Meyers &Meyers, 1995).

In the present study, drawings were scored by three raters and scores
compared to establish adequate inter-rater reliability. Drawings were reordered
and given new identification numbers to assure rater blindness to subject
diagnosis and copy versus recall trial. Two raters coded all of the drawings
(3:117, including those from both the copy and recall trials), while the third rater
coded a subset (u=44) of drawings. Taylor system scoring criteria, as published

by Meyers and Meyers (1995), were used. Raters were trained and calibrated by

40

first having them independently rate ten drawings completed by siblings of the
research participants. Following the practice ratings, the raters established
consensus for their score on the 18 individual criteria comprising the total score
of each drawing. A similar calibration meeting took place between two raters
following the scoring of the first 20 drawings in the sample. To preserve
independence these latter drawings were not re-scored.

Reliability was excellent both for the raters scoring all of the drawings
(a=.98) and for all three raters (a=.99). lntra-class correlations (lCC’s) were
computed as a further check and were also excellent for the two raters scoring
all the drawings (1 = .96) and for all pairs of raters (5 =98, p.98, and [=.97;
Appendix C: Table 6). These reliabilities were similar to those reported by
Meyers and Meyers (1995). Analyses used the mean of all raters for the total
score on the c0py trial.

Wide Range Assessment of Visual Motor Abilities (WRA VMA): Drawing
Subtest (Adams & Shes/ow, 1995) . Designed specifically for use in children, the
WRAVMA Drawing test requires children to copy simple drawings (e.g., a circle,
square, or pyramid) without erasing any parts of their attempt. Drawings become
increasingly difficult throughout the test. According to its developers this task
provides a measure of integrated visual-motor ability. Test-retest and split-half
reliabilities in the normative sample (N=2282, aged 3-17) are .89 and .81,
respectively. Because the figures for this task are simpler than the Rey Complex

Figure (thereby requiring less behavioral inhibition for correct performance),

41

ADHD children are predicted to be relatively less impaired on this task compared
to controls according to the Frontal Inhibition theory.

In the current study, two independent raters coded the drawings using the
Adams and Sheslow (1995) scoring criteria. An acceptable alpha reliability
between the two raters was obtained (or = .84; intra-class correlation coefficient: _r
= .71 ). Inter-rater reliability as reported in the WRAVMA manual in a smaller
sample (N=39) was 5 = .96 and .97 between three examiners (Adams &
Sheslow, 1995). The average score (combining both raters) was used in all
analyses (Appendix C: Table 6).

(c) Motor Tests:

Lafayette Grooved Pegboard (Trites, 1977). Designed to measure fine
motor skills, this test requires subjects, with one hand, to insert grooved pegs
into holes with the grooved edge at varying orientations. Time to completion of
the second row and entire pegboard are measured for both dominant and
nondominant hands. In normal children, time to completion of the task has been
found to decrease with age indicating improvement with increasing development
of fine motor control. Improved performance also suggests increased
automaticity of the required skill with age, as well as improvement in other
developmental attributes such as visual attention, visually guided behavior, and
fine motor planning (Solan, 1987). In an adult population (N=110) test-retest
reliability correlations on the grooved pegboard have been found to be r= .72 and

.74 for the dominant and nondominant hands respectively (Ruff & Parker, 1993).

42

Some evidence suggests that ADHD children may not exhibit deficits in
fine motor skill (Denckla, Rudel, Chapman, & Krieger, 1985). Because the
grooved pegboard assesses the ability of each hand separately, it may provide
some evidence concerning Iateralized deficits in ADHD. If ADHD children have
deficits in either right or left hemisphere functioning, they may show impaired
performance In the contralateral hand. If they have a bilateral deficit, overall
performance should be impaired. If no differences are found, that will be a
further indication that fine motor skill in ADHD is unimpaired. Such a finding
would also provide evidence for lack of simple motor deficits in ADHD as
proposed by the Frontal Inhibition theory.

Correlations between times to completion of the second row and entire
pegboard (five rows) were calculated for both left and right hands. The
correlation between the 2nd row and end time for the right hand was high (5 =
.91 ), whereas that for the left hand was somewhat weaker (_r = .77). Correlations
between the left and right hands were also large, but stronger for total than 2"d
row times (2"d row: 5 = .62, a=.76; end: _r = .78, a=.87). Since the correlation
between 2"d row and total times for the left hand was weaker than that for the
right hand, analyses were checked using both 2“d row and total times.

Time to Do Twenty. This motor task is modeled on Denckla's (1974) Time
To Do 20 motor skills battery. Four motor tasks were selected, two involving
hand movement and two involving foot movement. The first task, hand patting,
required participants to pat their hand on their thigh close to their knee in quick

taps raised approximately two inches above the leg. The second task, hand

43

rotation, required similar patting of the thigh but with alternating pronation and
supinatlon of the hand. The two foot tasks were toe-tapping and heel-toe
rocking and closely modeled the movements required in the hand tasks. Each
task was performed first on the dominant and then the nondominant side.
Children were told to perform the movements as quickly as they could while
maintaining the integrity of the movement. Time, in seconds, to complete twenty
movements provided the score for each task. Inter-rater reliabilities were
obtained between the live rater present during task performance and five
reviewers who watched videotapes of the performance. Prior reports indicate
that reliability (alpha=.85) for these tasks is quite good (Nigg, Carte, Hinshaw &
Treuting, 1998).

This task has proven capable of differentiating ADHD children from
controls, suggesting delays in gross motor development as well as programmed
motor output in ADHD children (Denckla, Rudel, Chapman, & Krieger, 1985;
Carte, Nigg, & Hinshaw, 1996). Denckla et al. propose that ADHD children are
delayed in the normal cephalo-caudal development of motor abilities. Thus, in
the current study, differences in performance of hand and foot movements were
predicted to occur between ADHD children and same-aged controls, with foot
movements better differentiating the two groups. Performance of these
movements also involves attentional control and may be mediated by frontal
mechanisms (Denckla et al., 1985). Scores for right-sided and left-sided
movements, however, have typically been combined for each task, thereby

preventing analysis of Iateralized deficits or soft signs. Analysis of left and

44

right-sided movements independently was predicted to provide information about
the existence of right, left, or bilateral frontal deficits in ADHD children.

In the current study, two live raters and five videotape coders were
employed to obtain reliability estimates for this task. Inter-rater correlations and
reliabilities were calculated for a subset of 50 subjects in the full sample
(including those excluded from the final study group). Because the comparison
of interest was between recorded times of the live raters and those of the
videotape coders, the live raters were examined as one unit and compared to the
videotape coders. The number of subjects rated by each videotape coder varied
somewhat, with 4 coders rating at least 10 subjects, but with one coder (Rater 4)
rating only 4 subjects. Forty-two of the fifty subjects were coded by two
videotape coders as well as the live rater, with the remaining subjects coded by
one videotape coder and one live rater.

The average correlation between the live rater and the five videotape
coders was satisfactory (_r = .83) with correlations ranging from .76 (Rater 1) to
.88 (Rater 5). Correlations among videotape coders were also satisfactory,
ranging from .81 to .99 (Appendix C: Table 7). Alpha reliabilities were calculated
for all sets of overlapping raters and ranged from .86 to .95, with an average
reliability of .92 (Appendix C: Table 8). The average time (across raters) for
each of the 8 motor tasks was used in all analyses.

Correlations among the average times for the 8 motor tasks ranged from a
low of .07 (right hand tap versus left foot pronation-supinatlon) to a high of .77

(right versus left hand pronation-supination) with an average correlation among

45

the tasks of .40 (Appendix C: Table 9). Findings thus supported the differential
validity of the various motor tasks.

Specific Hypomjsis Predictions

Frontal Inhibition Theory:

1a. ADHD children will show greater deficits than controls, as measured
by age-adjusted standardized scores, on the Controlled Oral Word Association
Test than on the WISC Vocabulary test.

1b. ADHD children will show deficits compared to controls (as measured
by age-adjusted standardized scores) on both the Ruff Figural Fluency Test and
the Controlled Oral Word Association Test.

10. ADHD children will show greater differences than controls on the age-
adjusted standardized scores on the Rey-Osterrieth Complex Figure Test (using
the Taylor system copy trial score) than on the WRAVMA drawing test.

1d. ADHD subjects will show a greater deficit than controls on the
Denckla Time To Do 20 Battery (measured by mean time for left and right hand
and foot movements), than on the Grooved Pegboard (measured by time to
completion of second row and entire tasks for both left and right hand).
Differences on the Denckla battery are expected to appear on both left and right-
sided movements.
flight Hemisphere Theory:

2a. ADHD children will show greater deficits than controls on the Ruff

Figural Fluency Test than on the Controlled Oral Word Association test.

46

2b. ADHD subjects will show deficits on simple and complex motor tasks
measuring left-sided performance. Left-sided performance on the Time to Do
20 test and the Grooved Pegboard will be slower than controls to a greater
degree than right-sided performance.
Data Analysis

Between group differences were analyzed using Analysis of Covariance
covarying for intelligence, age, handedness, and comorbidity (both categorically
and dimensionally). Main effects of group and sex as well as the relevant
interactions were examined. Comorbid Ieaming disability classification (as
specified in methods) was covaried as well as comorbid ODD/CD diagnosis as
diagnosed on the DISC-IV. Covariance analyses for comorbid aggression and
oppositional/conduct disordered behaviors were also performed using composite
dimensional measures of these symptoms created from parent and teacher
ratings on the CBCL and SNAP. Effects of predominantly inattentive ADHD
were controlled by removing these subjects from the study sample.

With the sample size employed in this study, large effect sizes (I = .40)
were required for reliable detection of significant differences. For instance, a
value off = .40 (eta2 =.16) was required to achieve a power of .80 for a group of
25 subjects. Cohen (1992) reports that f-values of .10 (c_l=.25) are equivalent to
small effects, fs equal to .25 (g=.50) constitute medium effects, and fs of .40 or
larger (g=.75) are equivalent to large effects. Obtained effect sizes of most
significant results fell between f_= .25 and .32 (eta2 = .06 - .10), suggesting a

power of .76 for detecting their significance (given a two group repeated

47

measures design). Overall, however, effect sizes for deficits on the
neuropsychological measures used in this study were similar to those reported in
the literature for executive functioning tasks (typically d=.60 or eta2 = .08 or
larger) for ADHD versus normal group comparisons (Pennington & Ozonoff,
1996). Thus, the utility of these neuropsychological tests in identifying ADHD
children or clarifying ADHD deficits is equivalent to those of traditional executive
function measures.

Although the number of tasks used in this study somewhat increased the
risk of a Type-1 error, it is proposed that correcting for number of tasks by
adjusting the significance level (e.g., with a modified Bonferoni correction setting
alpha = .01) would not be helpful. Due to the study's small sample size,
adjusting the significance level would lower power and thus increase Type two
errors. Furthermore, specific predictions are defined a priori in accordance with
theoretical proposals, arguably reducing the need to control for chance findings.
Therefore, it was decided that the significance level not be altered from alpha =

.05.

48

Plan of Analysis (See Table 2 below for a summary)

1.

Verbal Fluency vs. Vocabulary: Task (2 level within) by Group (2 level
between: ADHD vs. control) mixed factorial ANCOVA. The task by group
interaction will test Hypothesis 1a, with a significant interaction supporting the
hypothesis.

Verbal Fluency vs. Figural Fluency: Task (2 level within) by Group (2 level
between: ADHD vs. control) mixed factorial ANCOVA. The Task by Group
interaction will test Hypotheses 1b and 2a (Hypothesis 1b predicts a main
effect of group but no interaction whereas Hypothesis 2a predicts a significant
interaction).

Rey-Osterrieth Complex Figure vs. WRAVMA Drawing: Task (2 level) by
Group (2 level between: ADHD vs. control) ANCOVA will test the interaction
effect on the Taylor score versus drawing in Hypothesis 1c.

Time to Do 20 Battery and Grooved pegboard: Task (2 level within subjects)
by Side (2 level within) by Group (2 level between: ADHD vs. control)
ANCOVA. The Task by Group interaction will test Hypothesis 1d.

Hypothesis 2c will be tested by the Side by Group interaction.

49

Table 2.

Summary of Hypotheses and Data Analysis Plan

 

 

 

 

 

Task Model 1 Model 2 Model 3 Model 4
Verbal Fluency Interaction: Main effect:
Hypothesis 1a Hypothesis 1b
Interaction:
Hypothesis 2a
WISC Interaction:
Vocabulary Hypothesis 1a
Figural Fluency Main effect:
Hypothesis 1b
Interaction:
Hypothesis 2a
Rey Complex Interaction:
Figure Hypothesis 1c
WRAVMA Interaction:
Drawing Hypothesis 1c
Time to Do Task
Twenty interaction
Hypothesis 1d
Side
interaction
Hypothesis 2b
Grooved Task
Pegboard interaction
Hypothesis 1d
m
interaction:
Hypothesis 2b
ANOVA MODELS

Model 1: Task (Verbal Fluency and Vocabulary) by Group

Model 2: Task (Verbal Fluency and Figural Fluency) by Group

Model 3: Task (Rey-Osterrieth Complex Figure and Drawing) by Group
Model 4: Task (TTD-20 and Grooved Pegboard) by Side (Left vs. Right) by

Group

50

Results
Data Reduction andfiData Cleaning

For some children data were obtained partially at initial testing and
partially at follow-up testing one year later. To remove the effects of age within
subjects, age-corrected standardized residual scores were created for each
variable by regressing them one at a time on age. To address missing data and
to allow for inclusion of all cases, means were substituted for missing values.
However, to compensate for the potential effects of mean replacement, a missing
data variable was created and covaried in all subsequent analyses as
recommended by Cohen and Cohen (1983), enabling the effects of substitution
to be both assessed and removed.

For several analyses, homogeneity of variance assumptions were
violated. In an attempt to rectify this problem when it occurred, several steps
were taken, in the following order: 1) Square-root transformations were made on
age-corrected standardized scores; 2) log transformations were made on age-
corrected standardized scores; and 3) data points lying more than 2 standard
deviations away from the group mean were replaced with the next closest score
on the distribution. Results were unchanged except where noted. Results based
on untransformed data are reported, with footnotes indicating notable changes
using transformed data.

Analyses
All analyses employed repeated measures analysis of covariance with
post-hoc multivariate analyses of variance as set forth in the Methods section.

51

Effect sizes in all analyses were represented by eta squared (interpreted like r-
squared). An alpha level of .05 was used to evaluate the significance of all
statistical tests. Descriptions of group means can be found in Table 3 and a
summary of all results can be found in Table 4 for Hypothesis 1 and Table 5 for
Hypothesis 2. Tables are placed between Hypothesis 1 and Hypothesis 2 results
on pages 56-58.

Hypothesis 1 (Front_al Inhibition Theory)

(1 a).WISQ Vocabuflary versus Verbal Fluency (1a). The interaction
between group and the verbal factor (comprising verbal fluency total score and
WISC vocabulary raw scores) was nonsignificant. However, ADHD children had
poorer performance overall, producing a significant main effect of diagnosis
[F(1,52)= 4.59, p=.04, etaz=.08]. Unsurprisingly, this effect was no longer
significant after covarying estimated IQ [F(1, 51 )=1 .25, p=.27, eta"’=.024].1 Post-
hoc examination of verbal fluency and WISC vocabulary performance was
conducted to locate the source of the main effect. ADHD children performed
significantly worse than controls on WISC vocabulary [F(1,52)= 6.37, p=.02,
eta2=.109] but not on total verbal fluency scores [F(1,52)=.96, p= .33, eta2=.018].

Conclusion. Findings of poorer performance by ADHD children on the
WISC vocabulary but not the verbal fluency test were in the opposite direction

than predicted. The prediction was, therefore, not supported.

 

1 Covarying child sex reduced the effect of group on the verbal factor so that it
was only marginally significant [F(1,51) = 4.48, p=.06, eta2 =.068]. The same
was the case after covarying missing data and comorbid LD and CD diagnoses
[F(1,50)=3.52, p=.07, eta2=.066], although the main effects of these comorbid
diagnoses themselves were not significant (LD: eta2=.029, CD: eta2=.006).

52

(1 b) Rey Copy versus WRAVMA Drawing. The interaction between
diagnosis and the two-level drawing factor (comprising WRAVMA drawing and
Rey copy age-corrected standardized scores) was nonsignificant [F(1,52)=1.86,
p=.18, eta2= .035]. However, a significant main effect of group was observed,
with ADHD children performing worse than controls [F(1,52)= 30.31, p=.000,
eta2=.368]. Post-hoc tests revealed significant effects of group on both
WRAVMA drawing [F(1,52)=11.51, p=.001, eta2=.181) and Rey-Osterrieth copy
[F(1,52)=30.86, p=.000, eta2=.372; Figure 1 at end of results section]
performance. The group main effect remained after independently covarying
estimated IQ, sex, and comorbid diagnosis.

lmportantly, covariance of endorsed ODD/CD symptoms on the SNAP-IV
questionnaire (by mother, father, and teacher) altered the group main effect.
After covariance of these symptoms (for dimensional symptom analysis), the
effect of diagnosis on WRAVMA drawing performance was no longer significant
(p=.14, eta2=.045) and its effect on Rey copy performance was reduced (p=.003,
eta2=.17).1 The same pattern held when CBCL aggression was covaried instead
of SNAP scores, with WRAVMA drawing performance again falling below
significance (p=.26, eta2 = .026) and Rey figure effects remaining significant

(p=.00, eta2 =.299).2 Homogeneity of variance assumptions were violated on the

 

1 Maternal ODD/CD symptom ratings contributed the largest effect (p=.01, eta2
=.12), with father and teacher ratings contributing modest nonsignificant effects
father: eta2 =.02, teacher: eta2 =.03).

A marginally significant effect of father aggression ratings (p=.07, eta2 =.064)
primarily produced this result with mother (eta2 =.035) and teacher (eta2 =.012)
ratings contributing non-significant effects.

53

Rey copy task and attempts to correct this problem were unsuccessful.
However, re-analysis after data manipulations did not significantly affect findings.

Conclusion. The lack of an interaction between WRAVMA Drawing and
Rey Copy tasks means the best test of the hypothesis failed. However, the Rey
effect was more robust to covariates than the Drawing effect and was about three
times as large, providing some support for the prediction.

(1c) Groofl Pegs End Row Time versus Denckla TTD-20 Tasks. A
non-significant interaction was found between group and the two-level repeated
measures “motor-pegs” factor (comprising the mean time for all tasks of the
Denckla Time to Do Twenty [mean of foot and hand tasks combined, a=.84] and
mean time for the Grooved Pegboard [mean end time for left and right combined,
or=.90], F[1,52]=.01, p=.91, eta2 =.00). However, there was a marginally
significant main effect of group on the motor-pegs factor [F(1,52)=2.76, p=.07,
eta2 = .061], with ADHD children performing worse overall than controls. 1 Post
hoc tests revealed that ADHD children performed nonsignificantly worse than
controls on TTD-20 performance (p=.11, eta2= .05) than on grooved pegboard
performance (p=.21, eta2= .03). However, neither of these differences reached
significance.

When covarying LD/CD diagnosis, however, a significant interaction was
found between the motor-pegs factor and LD diagnosis [F(1,50)=5.97, p=.03,
eta2 = .107] as well as a significant main effect of LD diagnosis [F(1,50)=3.67,

p=.03, eta2 =.088]. Post hoc tests revealed that comorbid LD diagnosis

accounted for more of the ADHD deficit on grooved pegboard performance than
on TTD-20 performance. For instance, there was a significant effect of comorbid
LD diagnosis on grooved pegboard performance [F(1,50)=8.67, p=.005, eta2 =
.148] but not on Denckla TTD 20 performance (eta2 =.001). Covariance of LD
diagnosis considerably weakened the effect of diagnosis on pegboard
performance (eta2 = .015) but not TTD-20 performance (eta2 = .044), although
effects still remained shy of significance. All results were essentially unchanged
when data were analyzed using the 2"d row score.

Conclusion. Although the difference between ADHD and controls was
greater on the TTD-20 tasks than on the grooved pegboard, the interaction was
nonsignificant and differences on the individual tasks were both nonsignificant.
The prediction is therefore not supported. Notably, however, covariance of
comorbid LD diagnosis significantly reduced the grooved pegboard but not the
TTD-20 effect size, suggesting that the TTD-20 effect was more robust to

covariates.

 

1 Independently covarying estimated IQ and sex each weakened the main effect
of diagnosis (eta2 =.040 and eta2 =.050 respectively). although the main effects of
these variables themselves were nonsignificant.

55

Table 3. Mean scores by group for all dependent variables (mean raw1SD and

mean age-corrected standardized residuals 1 SE).

 

 

Raw Scores Standasrdized

cores

COWAT Control 20.94 1 7.28 .13 1 .19
ADHD 18.71 1 7.34 -.14 1.18

WISC Vocabulary Control 8.88 1 11.69 .37 1 .18
ADHD 23.37 1 9.24 -.27 1 .17

Ruff Figural Fluency Control 13.50 1 5.93 .10 1 1.16
ADHD 12.39 1 5.19 -.06 1.91

Rey-Osterrieth Copy Control 29.93 1 4.50 .61 1 .52
ADHD 19.35 1 8.52 -.61 1 .99

Rey-Osterrieth Recall Control 16.99 1 7.85 .42 1 1.04
ADHD 11.25 1 6.75 -.32 1 .82

WRAVMA Drawing Control 15.28 1 2.32 .49 1 .92
ADHD 12.81 1 2.22 -.34 1 .90

TTD-20 Control 7.05 1 1.46 -.17 1 .17
(mean left 8 right: hand 8 foot) ADHD 8.06 i 2.09 .15 1 .79
TTD-20: Right Control 6.98 1 1.49 -.14 1 .61
(mean of all right-sided tasks) ADHD 7.79 i 2.10 .10 i .84
TTD-20: Left Control 7.13 1 1.48 -.19 1 .55
(mean of all left-sided tasks) ADHD 8.32 i 2.23 .17 i .76
TTD-20: Right hand Control 6.20 1 1.10 -.17 1 .61
(mean of right hand tasks) ADHD 6.98 i 2.16 .14 i 1.03
TTD-20: Left hand Control 6.53 1 1.23 -.16 1 .63
(mean of left hand tasks) ADHD 7.21 1 1.86 .13 i .95
TTD-20: Right Foot Control 7.74 1 2.34 -.12 1 .85
(mean of right foot tasks) ADHD 8.60 i 2.62 .06 j; .90
TTD-20: Left Foot Control 7.73 1 2.05 -.22 1 .66
(mean of left foot tasks) ADHD 9.43 1 3.11 .22 i .85
Grooved Peg: End Control 83.71 1 16.40 -.19 1 .48
(mean of left and right hand) ADHD 97.71 1 40.57 .15 1 1.26
Grooved Peg: 2"d row Control 32.54 1 8.37 -.17 1 .75
(mean of left and right hand) ADHD 37.24 1 11.35 .15 _'I_'_ .97
Grooved Peg: Rt. Hand Control 77.72 1 15.89 -.23 1 .49
(time to end) ADHD 97.97 1 43.25 .28 1 1.35
Grooved Peg: Lft. Hand Control 89.70 1 20.51 -.14 1 .59
(time to end) ADHD 98.07 1 39.02 .01 1 1.21

 

56

Table 4. Summary of hypothesis 1 results including changes in effect size
following covariance of estimated IQ, sex, and comorbid LD/CD diagnosis.

 

Effect size (etaT)

After covarying;

 

 

 

Effect
Hypothesis F (1 .52) Size
W l Sex Lp/cp
1a) WISC
Vocabulary vs. Interaction 1.47 .027 .000 .039 .044
COWAT .
Main effect 4.59 .081 .024 .068 .066
Post Hocs: .
Vocabulary 6.37 .109 .026
COWAT .96 .018 .010
1b) Rey Copy vs.
WRAVMA Interaction 1 .86 .035 .012 .024 .024
Drawing ..
Main Effect 30.31 .368 .309 .361 .369
Post Hocs:

Rey copy 3086” .372

"

Drawing 11.51 .181

1 c) Grooved

 

Pegboard vs. Interaction .01 .000 .001 .000 .002
Denckla TTD-20
Main Effect 2.76” .061 .040 .050 .043
Post Hocs:
Pegs 1.69 .03 .015
TTD-20 2.60 .05 .044
*p<.05
"pg .001
+ p<.10

57

Table 5. Summary of hypothesis 2 results, including changes in effect size
following covariance of estimated IQ, sex, and comorbid LD/CD diagnosis.

 

Effect size (eta’) after
covaging:

 

 

 

 

Effect
Hypothesis F (1, 52) Size
(e757) IQ Sex LDICD
2a) COWAT
vs. RFFT
Interaction 0.08 .002 .000 .001 .000
Main Effect 1.14 .022 .017 .016 .014
Post Hocs:
COWAT 0.96 .018
RFFT 0.34 .006
2b) TTD-20
vs. Grooved Group x ..
Pegboard task x side 6.96 .12 .104 .103 .105
interaction
group x side
interaction 1 .74 .03 .020 .027 .021
Main Effect 3.36+ .06 .040 .050 .043
Post Hocs:
Right 2.64. .05 .037
Left 3.98 .07 .061
TTD-20:
Right vs. Left Interaction 1.81 .03 .041 .028 .044
Main Effect 2.48 .05 .042 .041 .044
Post Hocs:
Right 1.43. .03
Left 3.96 .07 .066 .061 .068
Grooved
Pegs: Right Interaction 5.06' .09 .070 .076 .070
vs.Lefl
Main Effect 1.69 .03 .015 .024 .015
Post Hocs:
Right 3.42+ .062 .036 .05 .039
Left 0.34 .007 .001
*p 5 .05
**p5.01
+ p<.10

58

Hypo_thesis 2 (Right Hemisphere Theory)

(23) Verbal Fluency verStg. figural Fluency. The group by task interaction
(COWAT versus figural fluency) was nonsignificant [F(1,52)=.08, p=.78, eta2
=.002] as was the main effect of diagnosis [F(1,52)=1.14, p=.29, eta2 =.02].
Covariance for missing data, estimated IQ, and LD/CD diagnosis produced no
significant effects. There was, however, a marginally significant effect of missing
data on figural fluency performance [F(1, 52)=3.23, p=.08, eta2 =.057],
suggesting that those subjects missing larger amounts of data performed more
poorly on figural fluency. Post hoc analyses indicated a marginally significant
main effect of sex [F(1, 51 )=3.23, p=.078, eta2 = .06] on verbal fluency
performance, with boys performing worse than girls. No significant effect of sex
on figural fluency performance was obtained. lmportantly, covariance of parent
aggression scale ratings on the CBCL produced a significant effect of father
ratings on fluency factor performance [F(1, 32)=5.57, p=.02, eta2 =.148] that
erased the effect of group on fluency (eta2 =.000). Such a result was not found
for covariance of SNAP ODD/CD ratings.

Conclusion. The prediction was not supported.

(_ZQLGrooved Pegboarp versus Time to Do Twenty M The task
(grooved pegs versus TTD-20) by group by side (left versus right) interaction was
significant [F(1,52)= 6.96, p=.01, eta2 =.12]. In addition, a marginally significant

effect of group was found, with ADHD children performing worse than controls

59

[F(1,52)=3.36, p=.07, eta2 = .06].1 However, the predicted group by side
interaction was nonsignificant [F(1,52)=1.74, p=.19, eta2 =.03].

Post hoc analyses revealed that ADHD children performed significantly
worse than controls on left-sided tasks [F(1,52)=3.98,p=.05, eta2 =.07] but not on
right-sided tasks [F(1,52)= 2.64, p=.11, eta2 =.05]. lmportantly, a marginally
significant interaction between side and LD diagnosis was found [F(1,50)=3.85,
p=.06, eta2 =.071], such that there was a greater (though nonsignificant) effect of
comorbid LD on right-sided than on left-sided tasks (eta2 =.034 versus eta2
=.006).2 Since eariier analyses indicated differences between grooved pegboard
and TTD-20 performance based on LD effects, for purposes of comparison the
two tasks were next analyzed independently.

Right versus Left-Sided Performance on t_he TTD-29. The group by side
interaction was nonsignificant [F(1,52)=1.81, p=.18, eta2 =.03] as was the group
main effect [F(1,52)= 2.48, p=.11, eta2 = .05]. Covarying LD and CD diagnosis
produced no significant effects.

As a weaker test of the prediction, multivariate analyses of covariance
were used to independently examine the effects of group on right versus left
TTD-20 performance. ADHD children performed significantly worse than controls
on left-sided [F(1,52)=3.96, p=.05, eta2 :07], but not on right-sided TTD-20 tasks

[F(1,52)= 1.43, p=.24, eta2 = .03].3 There was no main effect of comorbid L0 or

 

1Covarying sex and estimated IQ along with missing data each independently
reduced the main effect (eta2 =.050 and eta2 =.040 respectively).

2 Covarying LD diagnosis reduced the main effect of group on right and left sided
task performance (eta2 =.037 and eta2 =.061 respectively).

3 Covariance of estimated IQ and sex each independently reduced the effect of
group on left-sided performance such that it was only marginally significant [le

6O

CD diagnosis and covarying for them only minimally affected the group main
effect on left-sided performance (eta2 =.068).

In light of Denckla et al.’s (1985) proposal that the motor deficit in ADHD
children Is related to cephalocaudal developmental delays, additional analyses
were performed to ascertain that Iateralized group differences were not obscured
by the combination of hand and foot tasks. Right-left differences between ADHD
children and controls on the hand tasks were nonsignificant [right F(1,52)=1.71,
p=.20, eta2 =.032; left: F(1,52)=1.67, p=.20, eta2 =.031]. A marginally significant
effect of comorbid LD diagnosis on right hand performance was found
[F(1,50)=3.08, p=.09, eta2 =.058] and further weakened the group effect after
covafiance.

The group by side interaction for foot tasks was only marginally significant
[F(1,52)=2.50, p=.12, eta2 = .046]. However, multivariate analyses of covariance
revealed that in contrast to hand task performance, ADHD children performed
more poorly than controls on left but not right foot tasks [Ieftz F(1,52)=4.44, p=.04,
eta2 =.079; right: F(1,52)= 0.57, p=.46, eta2 =.011]. Covarying comorbid LD
diagnosis led to a much larger effect of group on left foot performance
[F(1,50)=5.52, p=.02, eta2 =.099; Figure 2 at end of results section].

Conclusion: Findings of left-sided group differences, especially for left-foot
tasks, supported the prediction and were consistent with a delay in

cephalocaudal development.

 

p=.06, eta2 =.066; sex: p=.07, eta2 =.061]. lmportantly, no significant effect of
handedness on performance was found, although covarying for handedness also
reduced the left-sided group main effect (eta2 =.065).

61

Right versus Left-Sided Performance on t_he Groovep Pegboard (time to
my. On the grooved pegboard, the group by side interaction was significant
[F(1,52)= 5.06, p=.03, eta2 =.09], such that ADHD children performed significantly
slower with the right than the left hand in contrast to control children who
performed more slowly with their left than right hand (Figure 3 at end of results
section). No significant main effect of group was found [F(1,52)=1.69, p=.20,
eta2 =.03]. Covarying the significant main effect of comorbid LD diagnosis
[F(1,49)= 10.55, p=.002, eta2 =.18] along with Word Attack [F(1,49)=7.09, p=.01,
eta2 =.13] and WIAT reading scores [F(1,49)=4.76, p=.03, eta2 =.09] reduced the
size of the group by side interaction so that it was no longer significant
[F(1,49)=1.22, p=.28, eta2 = .o2;].

Post hoc tests were used to examine the effect of group on left versus
right hand grooved pegboard (time to end) performance. A marginally significant
effect of group was found for right-handed performance [F(1,52)=3.42, p=.07,
eta2 =.062], with ADHD children performing the task slower than control children.
1 No significant difference between ADHD and control children was found for left-
handed performance [F(1,52)=.34, p=.56, eta2 =.007]. A large and significant
main effect of comorbid LD diagnosis (with those with ADHD+LD significantly
slower) for both right and left hands [right F(1,50)=7.17, p=.01, eta2 =.125; left:
F(1,50)= 8.34, p<.01, eta2 =.143] reduced the main effect of group on pegboard

performance when it was covaried (right: eta2 =.039; left: eta2 = .001 ).

 

1 Covariance of estimated IQ and sex each independently reduced the group
main effect on right-handed performance (IQ: p=.17, eta =.036; sex: p=.11, eta2
=.05). A significant main effect of handedness on right-sided performance was

62

Covariance of comorbid LD diagnosis as well as Word Attack and WIAT reading
scores completely removed the right-handed group effect [F(1,49)=.003, p=.96,
eta2 = .00].

Conclusions: Findings of a significant left-footed effect on the TTD-20
tasks partially supports the hypothesis of right hemisphere deficit in ADHD.
Failure to find such a deficit on the hand task could be due to cephalocaudal
developmental pattern (Denckla et al., 1985). 1 Notably, despite a marginally
significant right-handed deficit on the grooved pegboard, covariance of comorbid
LD diagnosis as well as indices of reading achievement removed this effect.
Therefore, the right hand deficit appeared to be due to LD-related factors

whereas the left foot deficit seemed to result from ADHD-related factors.

 

also found [F(1,51)= 12.53, =.001, eta2 =.197] but only somewhat reduced the
right-sided group effect (eta =.060).

1 Homogeneity of variance assumptions were violated for both TTD-20 and
grooved pegboard tasks. For TTD-20 this violation was corrected through
square-root transformation of the dependent variable and the data re-analyzed
with no significant alteration of results. However, homogeneity of variance could
not be established for scores on the grooved pegboard. Scores for two ADHD
subjects were moved within two standard deviations of the group mean and data
re-analyzed. The pattern of results was unchanged.

63

9°
D

 

9°
E

 

9°
5"

 

 

N
9°

 

Rey
WRAVMA

 

N
Q’

     

Standardized scores (log transformed)
(it)
<:>

I9
4:.

 

 

 

   

 

control ADHD

Figure 1. Nonsignificant group by task interaction for the WRAVMA
Drawing and Rey Complex Figure copy trial showing the significant group
main effect and the larger deficit on the Rey than the WRAVMA

64

Mean Time (seconds)

10.0

 

9.5 u

9.0 -

8.5 .

8.0 .

 

7.5

.-_

control

 

ADHD

 

 

E Right
m Left

Figure 2. Differences between ADHD and control subjects on the
TTD-20 left foot tasks.

65

Ag e-corrected standardized means

 

 

 

 

.4
.3 i
,<>
/
/
.2 ' / /
/
/
/
.1 'I / /
/
/
/
-.0 I / / //<>
/ //,--//
-.1 l [ML/Q
_/
(VI/T / / Hand
/ / — —
-.2 I / o .
ht

<>’ _R’9
-.3 0’ Left
control ADI-D

Group

Figure 3. Group by side interaction for grooved pegboard performance before
covariation of comorbid LD diagnosis and reading achievement.

66

DISCUSSION

Results partially supported both the Frontal Inhibition and Right
Hemisphere theories. Predicted differences between ADHD and control subjects
on the WRAVMA Drawing and Rey Complex Figure provided the primary support
for the theory. In contrast, predicted deficits on left-sided TTD-20 motor
movements served as the main support for the Right Hemisphere theory. A
major finding was that controlling for comorbid diagnoses and symptoms altered
results for several key predictions. Implications of these findings are discussed
for each theory in turn, along with general limitations of the study and directions
for further research.

Mesis 1 (Frontal lnhibitjon Theory).

Barkley’s Frontal Inhibition theory (1997) predicted that deficits in
behavioral inhibition should lead to greater deficits on tasks requiring increasing
inhibitory control. Thus, complex tasks were predicted to exhibit greater deficits
than simple tasks in ADHD children. Barkley proposed that this deficit would
occur across a range of modalities, including visual-motor, fluency, and verbal
domains.

The Frontal Inhibition theory received mixed support in the current data.
Group differences for the WRAVMA drawing and Rey-Osterrieth copy tasks
provided the strongest support for the theory. ADHD children showed greater
impairment on the relatively more complex Rey copy than on the simpler

WRAVMA Drawing task, supporting Barkley’s prediction of a greater deficit on

67

the more complex visual-motor task.1 In addition, although nonsignificant,
comparisons of TTD-20 and grooved pegboard performance produced
differences in the predicted direction, with the simpler grooved pegboard test
revealing smaller effects than the more complex TTD-20 motor battery.

Controlling for comorbid diagnoses and dimensional symptom ratings
clarified the pattern of results that supported the Frontal Inhibition theory.
Covariance of ODD/CD symptoms on parent and teacher SNAP-IV ratings, as
well as scores on the CBCL aggression scale, reduced the WRAVMA drawing
effect so that it was no longer significant. In contrast, the Rey copy effect
remained robust and significant after covariance of these factors. On the motor
tasks, covariance of comorbid LD diagnosis also altered findings, producing a
greater effect on grooved pegboard than on TTD-20 performance and further
reducing group differences on the pegboard (although group differences on both
tasks remained nonsignificant).

Taken together these results suggest that comorbid symptoms account for
group differences on “simple” tasks that the Frontal Inhibition theory predicts will
be unaffected in ADH D, but do not account for effects hypothesized to be
impaired (i.e., “complex” tasks). One implication, then, is that group differences
on the simpler tasks may have resulted from deficits associated with comorbid

symptomatology rather than ADHD-specific deficits. These findings highlight the

 

1 lmportantly, findings of relative differences between Rey copy and WRAVMA
drawing performance were found only during post hoc analyses. The group by
task interaction was nonsignificant. Consequently, this finding provides weaker
support of the theory than predicted by the a priori hypothesis.

68

importance of accounting for these comorbid disorders when examining
neuropsychological deficits in ADHD.

The finding of an ADHD deficit on the Rey Complex Figure was also
particularly noteworthy, since previous findings with the Rey have been mixed.
Grodzinsky and Diamond (1992) reported impaired ADHD performance on the
Rey, with greater deficits exhibited by older ADHD children. In contrast, a number
of studies have reported no deficits in ADHD children on Rey performance
(Carte, Nigg, and Hinshaw, 1996; Nigg, Hinshaw, Carte, & Treuting, 1998;
Reader et al., 1994). An important characteristic of the studies with
nonsignificant findings involved their use of the Weber and Holmes (1985)
scoring criteria for the Rey.

The Weber and Holmes (1985) criteria are said to emphasize
organizational qualities of the drawing and to provide a greater “executive
functioning” analysis of the figure. In contrast, the Taylor criteria used in the
current study emphasize accuracy and placement (Meyers & Meyers, 1995),
rather than organization. The Taylor scoring system was chosen for use in this
study, because it was believed to provide scoring criteria for the figure that were
similar to those used for the WRAVMA drawing task. Like the Taylor scoring
system, the WRAVMA drawing criteria evaluated accuracy and correct
placement of parts of the drawing in relation to each other. Because these tasks
were pitted against each other to evaluate differences in simple versus complex
task performance, the use of analogous scoring systems was considered crucial
in measurement of performance. Furthermore, it was believed that due to the
complexity of the Rey Figure, EF deficits would be manifested even through the

69

use of a simpler scoring system. Findings of greater deficits on the Rey Figure
than on the WRAVMA drawing task thus can be attributed to task demands
rather than differences in measurement methods.

What remains unclear from this study is the exact nature of the ADHD
deficit on the Rey Figure, and whether poor performance is simply an artifact of
organizational and inhibitory deficits, or the result of a distinct impairment in
visual motor perception. Since the Waber and Holmes criteria are purported to
measure largely executive functioning ability, large impairments on the Rey using
the Taylor criteria may suggest a subtle neuropsychological deficit in visual-motor
perception or integration. Obviously, further study is necessary to examine such
an assertion.

The strong current findings using the Taylor system indicate that the
accuracy and placement criteria may be more sensitive to detecting deficits in
ADHD. The Boston Qualitative Scoring System has also been reported to be
sensitive to deficits in ADHD, as well as to be able to discriminate ADHD from
control subjects with high sensitivity and specificity (Cahn, Marcotte, Stern,
Arruda, Akshoomoff, and Leshko, 1996). Like the Taylor system, this system
also places greater emphasis on attention to detail and accuracy rather than
organization. Examination of the consthcts measured by these three systems
may prove useful in further clarifying the type of ADHD deficit measured by Rey
copy performance.

In contrast to findings that support the Frontal Inhibition theory, several
hypotheses were not supported by the current data. Problematic for the Frontal
Inhibition theory were impairments in ADHD children on WISC vocabulary

70

performance paired with failure to find relative verbal fluency impairment on the
COWAT. In this case the “simpler" task revealed an ADHD deficit, whereas the
more “complex” one did not. In addition to a lack of significant differences in
verbal fluency, deficits in figural fluency also failed to receive support. These
findings contradict those predicted by Barkley.

Evaluation of the WISC vocabulary deficit in ADHD children is difficult
given the measure’s strong association with general intelligence. Findings of
lower IQ in ADHD children makes it impossible to determine whether this
vocabulary deficit is related to a specific verbal impairment or to decreased IQ
(especially given the number of children for whom lQ’s were estimated using only
the vocabulary and block design subtests of the WISC). Controlling for comorbid
LD diagnosis did not alter group differences, which suggests an ADHD-specific
impairment in either intelligence or vocabulary ability. Further clarification of the
nature of this verbal deficit in ADHD is necessary and should involve use of tasks
that can better discriminate verbal output and working memory deficits (i.e., on
complex verbal skills) from simpler verbal processing deficits.

The current study’s lack of support for verbal fluency impairments in
ADHD is consistent with prior findings and suggests that verbal fluency is either
unimpaired in ADHD or not complex enough to demonstrate an effect. Overall,
previous studies have produced mixed results regarding verbal fluency
impairments in ADHD children (Pennington & Ozonoff, 1996; Barkley, 1997).
Most studies have not found verbal fluency differences (Reader, Harris,
Schuerholz, & Denckla, 1994; Pennington & Ozonoff, 1996; Barkley, 1997).
However, Grodzinsky and Diamond (1992) described findings of verbal fluency

71

deficits in ADHD children on the COWAT (using the letters FAS) and noted that
group differences on the COWAT were smaller for older than younger ADHD
children compared to controls.

The current sample was closer in age to the “older ADHD” group in the
Grodzinsky and Diamond study. Consequently, failure in the current study to find
verbal fluency deficits, along with robust findings on the Rey, may be related to
the older age of our subjects. Denckla (1996) suggests that some tasks that are
sensitive to EF deficits during early stages of development will later lose their
novelty and become automatized. These tasks then move out of the EF domain
and become insensitive measures of EF deficits. Future studies may find it
useful to compare verbal fluency differences in younger versus older ADHD
children as a further means of determining the existence of verbal fluency, as
well as left frontal, deficits in ADHD.

Also important to consider is the possibility that Barkley’s review failed to
account for the confound of comorbid Ieaming problems in ADHD. If so, the
theory’s predictions of ADHD deficits on complex verbal tasks would be
unfounded. For example, although verbal fluency findings were nonsignificant,
controlling for comorbid LD diagnosis reduced group differences on this task so
that they were minimal (eta2 = .007). Furthermore, Pennington 8r Ozonoff (1996)
reported that deficits in ADHD children are not consistently found on verbal
measures, including verbal IQ, phoneme awareness, receptive language, and
story telling.

In addition to findings in the verbal domain, this study failed to support
ADHD deficiencies in figural fluency performance. Figural fluency differences in

72

ADHD have been largely unexplored prior to the current study. According to
Barkley (1997), only one study has reported deficits in ADHD children on a
nonverbal figural creativity task. Examination of performance on similar types of
tasks will be needed before anything definitive can be said about the existence of
these deficits in ADHD. Of greatest concern in the interpretation of these
findings is whether the task employed was sensitive to figural fluency or right
frontal deficits in children. The limitations associated with this task will be
discussed later.

In summary, current findings provide mixed support for Barkley’s theory.
Whereas there is some suggestion that ADHD children have more difficulty with
complex visuo—motor tasks, findings in the verbal domain were not supportive. In
the motor domain findings were weakly supportive when comorbid LD symptoms
were controlled (although the effect never reached significance). Taken together,
the current results suggest that the Frontal Inhibition theory alone cannot explain
all the neuropsychological deficits in ADHD. Instead, some combination of the
Frontal Inhibition and Right Hemisphere theories may be needed.

Whesis 2 (Right Hemisphere Theory):

Like the Frontal Inhibition theory, the Right Hemisphere theory received
only mixed support. Examination of right-left differences indicated deficits in
ADHD children on gross but not fine motor tasks. No deficit was seen on figural
fluency performance in ADHD children. Controlling for comorbid diagnoses and
symptoms again proved to significantly alter the pattern of results and highlighted

the importance of accounting for these differences in ADHD “subgroups”.

73

Grooved Pegboard findings. A three-way interaction between group, side,
and task (TTD-20 versus grooved pegboard) revealed a different pattern of
results on the grooved pegboard compared to the TTD-20 tasks. On the grooved
pegboard, ADHD children failed to show the typical right-hand faster than left-
hand control pattern. Instead, they performed the task at approximately the
same speed with both hands, with a significantly slower performance compared
to controls with their right but not left hand (inconsistent with a right hemisphere
deficit). ADHD children did not show an increased rate of left handedness
compared to controls, so results were not accounted for by subject handedness.

Controlling for comorbid diagnoses and dimensional symptom ratings
once again clarified the pattern of results. For instance, controlling for comorbid
LD diagnosis and reading achievement indices (on the WIAT reading and Word
Attack tests) removed the group difference on right hand speed on the grooved
pegboard. Such findings suggest that the anomalous pattern of performance in
ADHD children on the grooved pegboard resulted from left hemisphere
neuropsychological factors associated with Ieaming disabilities and reading
achievement, perhaps phonological processing skills in particular, rather than
from an ADHD-specific deficit.

Because no LD children were included in the control group, it could not be
determined whether this fine motor deficit was specific to Ieaming disabilities or
to the combination of LD and ADHD. Studies have indicated, however, that fine
motor development, as measured by the grooved pegboard, is correlated with
measures of reading readiness in kindergarten as well as with indices of reading
achievement in first and second graders (Solan & Mozin, 1986). Such findings

74

are consistent with current findings that controlling for indices of reading
achievement removed the ADHD effect on the grooved pegboard. Removal of
this LD and reading effect further underscores the importance of controlling for
comorbidity when examining neuropsychological deficits in ADHD. In this study,
controlling for comorbidity clarified that fine-motor speed impairments on the
right side were associated with comorbid LD and reading achievement apart from
ADHD.

TTD-20 Findings. In contrast to grooved pegboard performance, ADHD
children were slower on left than right foot TTD-20 tasks compared to controls
(thereby supporting the Right Hemisphere theory). The existence of global left-
sided gross motor output impairments was not supported, however, due to a
failure to find differences on the TTD-20 hand tasks. As with the grooved
pegboard, covariance of comorbid LD and reading achievement indices altered
the pattern of results. On the TTD-20 tasks, the size of the ADHD effect was
increased after covariance of these lndices.

Motor deficits in ADHD were hypothesized by the right hemisphere theory
to result from problems in the control of motor programming and persistence
rather than in execution of motor sequences (Voeller, 1991). Deficits on the
TTD-20 foot tasks support the appearance of impairment when the task becomes
more effortful and requires greater amounts of programming and persistence.
For instance, in addition to showing greater deficits on left foot tasks, a
marginally significant main effect indicated a tendency for ADHD children to be
slower overall on foot tasks (i.e., with left and right foot) compared to control
children. Controlled use of the hands, on the other hand, is more practiced and,

75

therefore, more automatic than is foot use. Consequently, the reduced
programming requirements for performance on the grooved pegboard and TTD-
20 hand tasks may account for failure to find group differences.

These findings may also be consistent with Denckla’s (1985) proposal of
cephalocaudal developmental delays in ADHD. Due to cephalocaudal
developmental patterns, performance of the foot tasks may require greater motor
programming (due to decreased automaticity) than does performance of the
hand tasks, and hence may be more sensitive to detecting Iateralized deficits.
Such a proposal is also consistent with Peters’ (1988) suggestion that foot
performance may be more sensitive to neurological impairments as a result of
better control of the upper extremities. Current findings of left-footed deficits in
ADHD children, therefore, may be strong indicators of right hemisphere
Impairment, despite the absence of hand deficits.

Notably, the failure to find deficits on TTD-20 hands tasks in the comorbid
ADHD+ LD subjects contrasted with findings of deficits on the grooved pegboard.
The apparent insensitivity of the TTD-20 hand tasks to left hemisphere
neuropsychological deficits associated with LD diagnosis or reading achievement
supports the differentiability of the skills required by the two tasks. One possible
distinction is between LD-related left hemisphere fine motor deficits (as
demonstrated on the grooved pegboard) and ADHD-related right hemisphere
motor programming deficits (as demonstrated on the TTD-20 tasks).

Prior studies have suggested that ADHD and LD- specific deficits exist in
ADHD children with comorbid LD (Pennington, Groisser, & Welsh, 1993). For
example, Denckla et al. (1985) found that a “dyslexia plus” group (described as

76

having discrepant reading abilities and attentional problems) had motor control
deficits on the TTD-20 battery that were not found in a pure dyslexia group.
Current findings provided evidence for motor deficits in a pure ADHD group that
are independent of LD symptomatology. Findings thus expand those of Denckla
et al. and provide evidence that some motor control deficits are specific to ADHD,
while other motor deficits are specific to L0, or at least to those with ADHD and
comorbid LD. Future studies may find it worthwhile to consider whether
ADHD+LD subjects can be differentiated according to primary “ADHD,” “LD,” or
“ADHD+LD” neuropsychological deficits.

In summary, an important finding in tests of the Right Hemisphere theory
was that, as with the Frontal Inhibition theory, factors associated with reading
achievement and comorbid LD significantly affected results, and would have
confounded interpretation of fine motor task results had they not been controlled.

Findings of slower left foot performance- paired with failure to find
differences on hand tasks-- did not support predictions of a global right
hemisphere motor impairment. However, consistent with predictions of the
Frontal Inhibition theory, results suggest that the more motor control and
programming required for the performance of a task, the greater the deficit. In
particular, left-sided tasks appear most affected, suggesting the possibility of a
right hemisphere substrate underlying motor programming and persistence (as
suggested by Heilman and Voeller).

These findings point to the possible usefulness of a synthesized Frontal
Inhibition-Right Hemisphere theory for a complete understanding of ADHD
deficits. Integration of the two theories may be best understood through the

77

incorporation of a “developmental delay" perspective on ADHD deficits. Such a
perspective would largely support a developmental right hemisphere model that
could be linked to the Frontal Inhibition theory in positing right hemisphere-
mediated delays in the development of behavioral inhibition.

General CODCIIfiIOflSZ

Both theories received only partial support. Findings suggest that a
combination of the two theories may provide the best explanation for current
results. As hypothesized by the right hemisphere theory, the deficit in ADHD
may most likely involve right frontal-striatal impairments in motor programming
and output. Impairment on these tasks may increase with task complexity,
causing greater deficits to appear when greater amounts of control are required.
For example, findings of deficits on left-footed motor tasks not only suggest a
right hemisphere deficit but also suggest that impairment increases as the need
for motor control and planning increases. Findings of deficits on complex
drawing tasks could also be consistent with a right hemisphere-mediated deficit
in behavioral inhibition.

With this said, a number of caveats should be kept in mind when
interpreting these theories. Both theoretical and empirical evidence suggests the
potential applicability of a neuropsychological laterality perspective to the study of
ADHD. However, any identification of a disorder as caused by a Iateralized
deficit should take into account the interaction between the hemispheres and the
effects that dysfunction in one hemisphere may produce in the other. For

instance, hypoactivation in one hemisphere could cause inhibitory failures that

78

produce hyperactivation of the other hemisphere-- thereby resulting in bilateral
functional deficits (Tucker 8 Williamson, 1984).

A Iateralized perspective also can be criticized for a focus on gross brain
regions rather than specific distributed neural systems. For example, the
anterior-posterior and lateral-ventral axes of the brain, as well as specific neural
networks for attention, vigilance, and working memory, also are clearly important
to understanding neural mechanisms in ADHD. The Frontal Inhibition theory
assumes a bilateral, anterior-based neurological substrate with a hypothesized
discrete neural system (e.g., behavioral inhibition). The theory thus can explain
deficits according to an impaired neuroanatomical region and, at the same time,
account for cognitive deficits associated with ADHD. However, the Frontal
Inhibition theory also draws criticism for its over-broadness of scope and lack of
specificity. Eventually, the most effective theory of ADHD will probably
incorporate hypotheses involving multiple axes of the brain and its neural
substrates. Current findings took one step towards this by suggesting the need
for integration of the Right Hemisphere and Frontal Inhibition theories.

In addition to combining behavioral inhibition and right hemisphere
perspectives, the results also suggest the need for incorporation of a
developmental component to both theories. Barkley (1997) suggests that ADHD
deficits result from delays in the development of behavioral inhibition
mechanisms. Children with ADHD are proposed to demonstrate behavioral
inhibition development with the same shape and trajectory as normal, younger
children’s development. Despite this assertion, little research has directly
compared cognitive or neuropsychological performance in older versus younger

79

ADHD children. Grodzinsky and Diamond’s (1992) findings of normal verbal
fluency and impaired Rey Figure performance in older children is one of the few
studies to compare older and younger children on traditional EF tasks. Whether
the suggestion of cephalocaudal developmental delays in ADHD stemming from
current findings (and suggested previously by Denckla et al.1985 and Nigg et al.,
1998) can be explained by behavioral inhibition deficits will be important to
explore.

There are two possible mechanisms that might relate behavioral inhibition
and right hemisphere development. The first possibility is that mechanisms of
behavioral inhibition may rely primarily on right hemisphere functions (as is the
case with attentional mechanisms), thereby producing greater deficits in right
than left hemisphere mediated processes. Right-hemisphere mediated vigilance
and arousal mechanisms could also moderate the effectiveness of inhibitory
function, which, In turn, might influence executive functioning (Van der Meere,
1996; Van der Meere, Vreellng, & Sergeant, 1992).

A second possibility is that development of behavioral inhibition
mechanisms occur in a left to right hemisphere direction, so that right
hemisphere functions remain impaired later into development, while left
hemisphere deficits abate. Best (1988) asserts that despite earlier morphological
development of right frontal-motor regions, a left-biased growth of tertiary regions
in the prefrontal cortex causes right prefrontal regions to develop later than left
prefrontal regions. Best further suggests that morphological development is
likely to be associated with functional maturation. Thus, earlier morphological
maturation of a region should be associated with earlier maturation of its

80

associated functions. If mechanisms of behavioral inhibition are localized to right
and left prefrontal regions, impairment should persist longer for the right
hemisphere. Obviously such interpretations are, at this point, speculative. For
clarification of this issue, further research should examine the existence of
Iateralized deficits in ADHD at different points in development.

lmportantly, despite the possible utility of a synthesized Frontal Inhibition-
Right Hemisphere theory for understanding ADHD deficits, much of the current
data remain unexplainable by either theory. Figural fluency findings provided no
support for either theory, and vocabulary and verbal fluency findings did not
support the Frontal Inhibition theory. Controlling comorbid LD diagnosis did not
alter the vocabulary effect in current findings. Thus, this deficit is indicative of an
ADHD-specific impairment in either IQ or vocabulary ability. If true vocabulary
deficits exist in ADHD, the right hemisphere theory is inadequate to explain them.
Limitations

Certain limitations in study design may have accounted for the study’s null
findings, so future studies will need to correct for these limitations to more
accurately assess impairments.

Validation of neuropsychological measures on populations with acquired
brain damage makes their application to the study of developmental disorders
somewhat problematic. Because brain injured populations are often described in
terms of broad regions of damage (e.g., right frontal or left posterior), there has
been little linking of impaired performance on these tasks to interrupted
neuroanatomical or cognitive pathways. Studies such as Casey et al.’s (1997)
neuroimaging study provide an ideal test of the relationship between

81

neuroanatomical and cognitive impairments. Future neuropsychological studies
must work to link neuropsychological test performance with specific
neuroanatomical substrates in children.

Another problem with the neuropsychological tests used in this study is
their limited normative information with children. This problem was perhaps most
evident on the figural fluency test, whose goal the children seemed to have
trouble understanding. Observations during the study suggested that children
often failed to understand that the task was to draw as many different simple
designs as possible and instead tried to make over1y intricate designs. As a
result, their performance suffered. This task, therefore, may have provided an
inappropriate test of right frontal lobe functioning in children.

Other factors that weakened current findings were failure to include a pure
LD group, as well as the small sample size and numerous post hoc statistical
tests employed. The small sample size created inadequate power for reliable
detection of small ADHD deficits. As a result, a number of group differences may
have been missed. At the same time, the large number of post hoc analyses
raises the likelihood that some significant findings reflect Type I errors. Future
studies can correct for these limitations by inclusion of a larger sample.

Future Directions

The results suggest many questions for further study. Of greatest value is
the need for the theoretical integration of the Right Hemisphere and Frontal
Inhibition theories to account for ADHD deficits. For instance, if substrates of
behavioral inhibition (or those underlying behavioral inhibition) can be localized
predominantly to the right hemisphere, then it may be possible to incorporate the

82

 

two theories into a new unified theory. Alternatively, if those functions
hypothesized to be impaired by the right hemisphere theory demonstrate a
simple versus complex impairment dichotomy then we would have further
evidence for the integration of the two theories. Some evidence for such a
dichotomy was found in the current study, with findings of impairment on foot but
not hand tasks.

The findings also highlighted the importance of further differentiating
neuropsychological deficits in LD+ADHD and pure ADHD. Results suggested
that LD and ADHD-specific impairments might be differentiable according to
demonstrated neuropsychological deficits in motor control. Similar differentiations
have been found in executive function and phonological processing (Nigg et al.,
1998; Pennington et al., 1994). However, whether comorbid ADHD and LD
involves deficits specific to both pure forms of the disorder or whether one
disorder produces "phenocopy" symptoms of the other remains to be proven.
The evidence to date supports both interpretations (Nigg et al., 1998; Pennington
et al., 1993; Denckla et al., 1985). In future studies it may be useful to examine
neuropsychological deficits in the comorbid as well as pure forms of the disorders
as a further means of clarifying this comorbidity issue.

A third direction suggested by the current findings is a need for better
neuropsychological tests of the hypothesized deficits. For instance, use of the
figural fluency and WISC vocabulary tests proved to be weak measures for
examining right frontal and simple vocabulary deficits in ADHD. Future studies
should use more accurate, sensitive, and developmentally appropriate measures.
For instance, tests of visual working memory may provide one means of tapping

83

right frontal impairments, while tests of vocabulary that are not highly correlated
with intelligence would provide a better means of assessing simple language
functioning.

Further analysis of verbal deficits in ADHD Is also desirable. For
instance, determining whether these deficits are due to working memory or motor
output impairments rather than phonological processing and language deficits
(as are present in Ieaming disabilities) would be particularly useful. Mariani and
Barkley (1997) reported finding deficient word reading/decoding ability in
preschool children with ADHD. The authors proposed that this impairment may
have resulted from impairments in working memory because the reading
measure strongly loaded on a working memory factor. They proposed that early
word recognition might depend on working memory rather than on phonological
processing factors. Further examination of these findings would prove useful.
Also important to consider is whether Barkley’s prediction of verbal deficits in
ADHD is an artifact of the high degree of comorbidity between ADHD and LD
rather than an ADHD-specific cognitive impairment (Nigg et al., 1998).
Pennington and OzonofI's (1996) assertion that verbal deficits are not
consistently found in ADHD supports this possibility.

Further examination of deficits on Rey Complex Figure performance
would also prove useful. Analysis of the approach used to draw the figure (for
instance, holistic versus sequential) may provide another means of examining
whether Impairments on the task result from Iateralized deficits. Such studies
could provide a clearer picture of the deficit on this task as well as examine
whether current findings of impairment on the two drawing tasks resulted from

84

specifically right hemisphere impairments in addition to increased behavioral
inhibition requirements.

Lastly, the current study attempted to explore primarily right frontal
performance in ADHD by using neuropsychological tasks asserted to tap right
frontal mechanisms. However, possibilities for right posterior deficits exist as
well (Aman, Roberts, & Pennington, 1998). Voeller and Heilman’s theory of right
hemisphere dysfunction is not incompatible with the prediction of global right
hemisphere impairments. Voeller (1991) suggested that posterior impairments
might better describe predominantly inattentive children, and others have argued
that combined and predominantly inattentive type children may be differentiable
according to the anterior-posterior locus of their neuropsychological impairment
(Goodyear & Hynd, 1992). Aman, Roberts, and Pennington (1998) suggest that
posterior deficits in ADHD could be artifacts created by predominantly frontal
Impairments. However, the possibility for independent posterior impairment
cannot be ruled out. Attempting to differentiate ADHD children according to their
neuropsychological test performance could prove interesting, particularly in
differentiating predominantly inattentive from combined-type children.

Along with differentiation of subtypes according to comorbidity and
symptom clusters, gender differences will be another important area to consider
in future studies. Some studies have suggested that different patterns of deficits
may exist in girls with the disorder (Ernst, Liebenauer, King, Fitzgerald, Cohen, 8
Zametkin, 1994). Consequently, an account of gender when considering ADHD

subtypes will be important. Any comprehensive neuropsychological theory will

85

need to explain these different patterns of deficits, as well as the higher
occurrence of the disorder in boys.

The current study limited its scope to consideration of deficits in only the
combined type of the disorder. Even an integrated Right Hemisphere-Frontal
Inhibition model may prove incapable of explaining all ADHD deficits, particularly
symptom clusters in its subtypes or comorbid forms. Consequently, it will be
important for future studies to consider the possible contributions of the other two
other major neuropsychological theories of ADHD, the Anomalous Dominance
and Left Hemisphere theories (see Appendix A).

The Left Hemisphere Theory of ADHD may be most applicable to current
findings that controlling for comorbid LD and reading ability often altered the
pattern of results. Although recent studies have given greater consideration to
comorbid disorders, comorbid LD has not often been controlled. Current findings
suggested that left hemisphere-related impairments in ADHD could be accounted
for by controlling for comorbid LD diagnosis and reading ability. Consequently,
the Left Hemisphere theory may be an artifact of these uncontrolled studies or
may be applicable to describing a unique group with coexisting ADHD and LD.
Future studies will need to examine whether those with comorbid LD and ADHD
represent a distinct group with left hemisphere impairment or whether behavioral
symptoms of ADHD are phenocopy symptoms related to primary LD cognitive
impairments.

Similar considerations are relevant to studies of the Anomalous
Dominance theory, whose primary application, to this time, has been towards the
explanation of Ieaming disabilities. In the current study, failure to find left

86

hemisphere impairment in a noncomorbid ADHD group does not support
predictions of the Anomalous Dominance theory. Given the Anomalous
Dominance theory’s lack of support in prior studies of ADHD, this theory may
have limited applicability to describing deficits in the disorder. The theory’s main
contribution to studies of ADHD thus may be its emphasis on the role of
developmental factors in the creation of childhood disorders. Current findings
also highlighted the importance of incorporating developmental considerations
into neuropsychological theories.

Summagy.

The current study’s use of widely employed clinical neuropsychological
tests suggests that such tests could prove sensitive to deficits in ADHD, and also
help to evaluate complex theories. Findings also suggest that some traditional
neuropsychological tests are better at evaluating deficits than others, with some
tests exhibiting problems in their applicability to a child population. The
development of age-appropriate neuropsychological tests eventually could lead
to the development of more sensitive measures for the evaluation of ADHD in
clinical practice and in research.

Examining neuropsychological functioning in ADHD provided valuable
information towards the understanding of neuropsychological mechanisms of the
disorder. It also proved useful in differentiating those ADHD children with
different patterns of neuropsychological deficits. Further work in this area would
prove useful. For instance, as was found in the current study, children exhibiting
different comorbid diagnoses or symptoms may exhibit different patterns of
Iateralized dysfunction, thus allowing for differentiation according to

87

neuropsychological profile. Incorporation of the anterior-posterior cerebral axis
as well as identified cognitive substrates, as in the Frontal Inhibition theory, may
provide further differential power, particularly with regard to differentiating
predominantly inattentive from combined type children. Such findings not only
can enhance diagnostic specificity but also the ability to identify other psychiatric
disorders representing phenocopies of ADHD (thereby excluding false positive
diagnoses). In the current study, both the Frontal Inhibition and Right
Hemisphere theories received some support, with an integrated approach
appearing to provide the most useful explanation of ADHD deficits. Further
studies will need to evaluate the ability of such a synthesized theory to explain

patterns of neuropsychological impairment in ADHD.

88

 

APPENDIX A

Out of the literature on development of lateralization has come theories
concerning the role played by lateralization in developmental disorders. Although
difficult to verify empirically, such theories provide potentially useful heuristics by
which to conceptualize the development of possible Iateralized deficits in ADHD.
The theory of anomalous dominance arises most directly from this literature and
provides the third of the four neuropsychological theories of ADHD.

Theory of fliomalous Dominance

Two versions of anomalous dominance warrant mention-— Geschwind's
(1985) Anomalous Dominance theory and Annett’s (1984) Right Shift theory.

Geschwind . Geschwind and Galaburda (1985) proposed a theory of
abnormal development of lateralization which appears consistent with recovery of
function findings in acquired brain damage. Although they did not set out to
explain ADHD, their thinking is readily applied to ADHD. Geschwind and
Galaburda hypothesized that fetal exposure to prenatal testosterone modulates
the development of lateralization primarily by slowing the development of the left
hemisphere, creating vulnerabilities towards left hemisphere deficits. As in
acquired hemispheric lesions, the left hemisphere incurs an insult for which the
right hemisphere is believed to attempt to compensate. The result is what
Geschwind and Galaburda term "anomalous dominance." Anomalous
dominance comprises six factors: reversed handedness, reduced degree of
handedness, reversed language dominance, reduced degree of language
dominance, reversed dominance in favor of right hemisphere functions, and
reduced dominance in favor of right hemisphere functions. In other words,
language, handedness, and visuo-spatial functions are theorized to have
reduced or opposite dominance to those found in most individuals.

In addition, Geschwind and Galaburda theorize that the right hemisphere
undergoes compensatory development due to left hemisphere growth

89

retardation, resulting in "giftedness" for those skills in which the right hemisphere
is involved. lmportantly for the discussion of ADHD, however, testosterone is
also thought to produce delays in anterior right hemisphere development or right
frontal delay (McManus & Bryden, 1991). Extrapolating from this theory then,
ADHD children might exhibit deficits in right frontal functioning along with
enhanced right posterior functioning. However, right posterior compensation
could also lead to overactivation of posterior attentional processes leading to
enhanced responsivity to environmental stimulation in ADHD.

Though its mechanisms do not appear clearly elucidated in Geschwind's
theory, these effects are proposed to occur at a testosterone sensitive period in
fetal development. Differences in fetal testosterone are theorized to result
through both genetic and environmental mechanisms and, thus, can account for
heritable or environmental risks for disorders presumed to result in Iateralized
deficits. Learning disabilities and other language based disorders, for example,
are hypothesized to be affected by this particular growth pattern. Though this
theory was developed specifically to explain Ieaming disabilities, individuals with
ADHD may exhibit similar patterns of lateralization. The much greater ratio of
males with the disorder suggests that testosterone may play a role in producing
ADHD deficits. Furthermore, the delays in anterior right hemisphere
development proposed by Geschwind and Galaburda are consistent with brain
imaging findings of right frontal lobe abnormalities in ADHD(CastelIanos et al.,
1996; Casey et al., 1997), and may provide some rationale for a specifically right
hemisphere frontal deficit as reviewed below.

m . Annett's Right Shift Theory (1984) suggests genetic mechanisms
linked to the development of lateralization and may be relevant to the study of
Iateralized deficits in ADHD. Annett and Kilshaw (1984) posit a genetic effect for
the determination of right handedness and magnitude of left hemisphere
Iateralized speech and language. For example, the average individual is
believed to receive a dominant gene (RS+) for left hemisphere language
specialization and "right handedness" from one parent, but no such gene from
the other parent. Some individuals, however, may either inherit the dominant

90

 

"right" gene from both parents or from neither parent. Individuals with both
"right" genes (RS+RS+) are hypothesized to be highly Iateralized towards the left
hemisphere and may actually exhibit some impalrrnents in right hemisphere
functioning. Individuals without any "right” genes (RS-RS-), however, would form
hemispheric specialization according to environmental influences. In the
absence of strong environmental influences about half of these individuals are
expected to develop left hemisphere dominance, and half to develop right
hemisphere dominance (Annett & Kilshaw, 1984). It may be interesting to
consider, however, the effect of prenatal testosterone on the RS-RS- individual
and whether Geschwind's theory could explain the development of lateralization
in these individuals. Such a theory is relevant to ADHD in that certain genes or
gene combinations may produce alterations in neural mechanisms underlying
right hemisphere Iateralized processes.

These two theories together provide a means to conceptualize possible
pathways to atypical lateralization in ADHD. Annett’s and Geschwind's theories
are not well-validated empirically and are based primarily on post hoc use of data
on developmental disabilities. Nevertheless, existence of Iateralized functional
differences during development suggests that some factors do exist that could
influence the development of lateralization and, consequently, produce
development of atypical hemispheric functioning. Such factors may produce
lateralization deficits in ADHD. These theories underscore the theoretical
plausibility that deficits in lateralization could occur somewhere in the course of
ontogeny, linking ADHD to atypical development of lateralization.

The "lfift Hemisphere" Theory

Malone, Kershner, and Swanson (1994) proposed that ADHD involves a
deficit in the left frontal-striatal dopamine pathway producing relative over—arousal
of the right posterior-noradrenergic pathway. Such overarousal is thought to
result from the left hemisphere's failure to inhibit the right's activation. These
authors drew upon Tucker and Williamson's (1984) revision of Pribram and
McGuinness' (1975) attentional model. In Tucker and Williamson's formulation,
internally focused, tonic, routinized behavior is attributed to preferential left

91

hemisphere mediation and externally focused, phasic, novelty-oriented behavior
to primary right hemisphere control. According to the logic that ADHD involves
left hemisphere underarousal, these children should exhibit deficits in sustained
attention, along with a bias towards change and novelty produced by the relative
dominance of the right hemisphere arousal system. This theory thus accounts
for failures in focused attention as well hyper-responsivity to external stimuli in
ADHD children.

Studies of left hemisphere mechanisms in ADHD have consisted primarily
of those on visual attention. A finding of right visual field/left hemisphere deficits
in visual orienting under effortful processing conditions supports the hypothesis
of left hemisphere dysfunction in ADHD (Swanson et al., 1991). ADHD subjects
were found to orient more quickly to targets in the left visual field following invalid
cues to the right visual field/left hemisphere, suggesting a failure to maintain
focused attention- consistent with problems in the functioning of left frontal
dopamine pathways (Tucker & Williamson, 1984). Notably, however, this finding
has not replicated in other studies of visual attention.

It Is interesting to consider whether the high degree of comorbid Ieaming
disabilities in Swanson et al's ADHD sample may have accounted for this finding
of left hemisphere dysfunction. If so, findings of a left hemisphere deficit in
ADHD may be the result of group heterogeneity, especially of inadequate
differentiation between LD and ADHD groups . While this comorbid Ieaming
disability group accounted for only about fifty percent of the sample (8 of 17), the
majority of the remaining subjects had been given a co-diagnosis of oppositional
defiant disorder (7 of 17). Consequently, comorbid disabilities rather than ADHD
symptoms themselves may primarily have accounted for the study's results.

92

APPENDIX B

Background on Lateralization
Functional Effects of UnilateraLBrain Damage. Most functions attributed

to Iateralized processes have been defined according to deficits following
unilateral brain lesions. As reviewed by Springer and Deutsch (1989) for
example, certain left hemisphere lesions produce deficits in language functions
such as speech, writing, and reading that are not found after comparable right
hemisphere damage. On the other hand, many abilities thought to be mediated
primarily through the right hemisphere, such as spatial tasks, singing, musical
ability, and emotion perception are impaired following right (but not left)
hemisphere lesions. According to Springer and Deutsch (1989), this right-left
functional dichotomy has also been conceptualized according to processing
strategies utilized by each hemisphere. The left hemisphere has been proposed
to dominate during tasks requiring analytic and sequential processing while the
right hemisphere utilizes gross, holistic processing mechanisms. Normal
individuals have been found to demonstrate such advantages on dichotic
listening and tachistoscopic tasks designed to tap Iateralized processes (Springer
& Deutsch, 1989). Further evidence for functional hemispheric specialization has
been obtained through study of the nature of deficits displayed in split brain
patients who possess no interhemispheric communication (Kolb 8r Whishaw,
1990) and from functional brain imaging data in normal individuals (Cabeza &
Nyberg, 1997).

Development of Lateraligm . Also important in the consideration of a
Iateralized deficit and to theories in ADHD is the role played by development in
the formation of Iateralized processes. Genetic influences or environmental
insults to the brain during development may create impairments similar to those
following acquired brain damage. Alternatively they may create subtle changes
in brain mechanisms or chemistry such that more subtle, but similar, deficits may
occur. Because ADHD is presumably a developmental disorder, understanding

93

 

the mechanisms by which development of a Iateralized deficit could produce the
symptoms of ADHD is essential.

Lateralization is currently thought to be present in the infant from before
birth. Evidence suggests that genetic determinants of maternal cytoplasmic
constituents act to repress or activate DNA segments coding for levo or dextro
proteins in the developing embryo. Levy (1977) asserts that the biological
activity of a protein is a function of its sthcture and, therefore, that asymmetric
proteins could easily affect the resulting asymmetrical morphology of the
individual. Several different forms of behavioral evidence have been drawn upon
to support this assertion. First, infants exhibit Iateralized functioning from a very
early age. For instance, infants as young as three months old have been shown
to exhibit right ear biases for verbal sounds in dichotic listening tasks; while
infants as young as one week old have exhibited greater evoked potential
amplitudes in one or the other hemispheres depending upon the nature of the
stimuli. Furthermore, infants just a few days old often exhibit right sided head
turning preferences (Springer 8 Deutsch, 1989).

Other examinations of lateralization in infancy have found right
hemisphere advantages in four to ten month old infants' recognition of their
mother's faces and in identification of spatial location. The left hemisphere, on
the other hand, was found to specialize in componentlal discrimination, or
discriminating parts of objects. Thus, these infants were unable to process
components of faces when presented to the right hemisphere despite showing an
enhanced ability to recognize familiar faces and the spatial relation of facial
components to each other (de Schonen 8 Mathivet, 1990; Dereuelle 8 de
Schonen, 1995). In the domain of emotional functioning, infants have been
found to exhibit lateral asymmetries in the frontal lobes associated with response
to maternal separation. Infants with relative increased right frontal activation
cried during maternal separation, while those who did not cry exhibited greater
left frontal activation (Davidson 8 Fox, 1989). Whether these Iateralized findings
increase in magnitude or remain the same throughout development is open to
some debate with some studies reporting increases in the magnitude of the

94

 

Iateralized response with development and other studies failing to find such
differences (Springer 8 Deutsch, 1989)

Additional evidence for the existence of lateralization early in life derives
from studies on hemidecortication in infants. Along with reports on the effects of
early unilateral brain lesions in infants, the apparent absence of long-term effects
of hemidecortication (the removal of one hemisphere) has often been used to
argue for an absence of lateralization in the young child's brain. Using these
findings, arguments for social Ieaming or environmentally reinforced
determinants of lateralization have been made. Nevertheless, such arguments
appear unfounded (Springer 8 Deutsch, 1989). Infants or young children who
have unilateral brain damage experience largely the same effects as do adults
with similar damage-— despite having lost the hemisphere at birth. For example,
findings indicate that children do exhibit Iateralized deficits, such as aphasia,
immediately following unilateral damage. Long-term outcomes in children,
however, often indicate a recovery of functions by the alternate hemisphere such
that no lasting deficits exist in adulthood (reviewed in Springer 8 Deutsch, 1989).
Springer and Deutsch (1989) suggest that such recovery of function is more
consistent with effects of brain plasticity in early childhood than with a lack of
lateralization. The recovery of functions, therefore, apparently supports the
assertion that one hemisphere has taken over the damaged hemisphere's
functions, not that these functions existed in both hemispheres at this stage of
development or that no hemispheric specialization existed.

Further, some studies suggest limitations in the extent of recovery of
function following hemidecortication. For example, Dennis and Whitaker (1976)
found that although no differences in verbal and performance IQ existed in right
and left hemidecorticates, left hemidecorticates performed worse on a test of
complex syntax. This finding suggests that when faced with more complex tasks,
the right hemisphere is unable to perform processes specialized for left
hemisphere mechanisms. Further related to these findings are studies in
children with closed head injuries which suggest that cognitive impairments are
dependent upon the age at which injury occurs (Scheibel 8 Levin, 1997).

95

Abilities which normally develop after the age at which the head injury is received
may be impaired while earlier processes remain intact. As such, impairment in
children with closed head injuries often may not become apparent until later in
development (Scheibel 8 Levin, 1997). Findings such as these point to the
persistence of Iateralized deficits following brain injury even in children.

 

96

APPENDIX C: DETAILED RELIABILITY DATA

Table 6. Reliability and intra-class correlation estimates for the WRAVMA
Drawing and Rey-Osterrieth Complex Figure tasks.

 

 

WRAVMA Drawing

Rey Complex
Figure:

2 raters (N=117)
Rey Complex
Figure:

3 raters (N=44)

 

Alpha ICC
.84 .71
.98 .96
.99 .98, .98, and .97

(comparing each rater to
one another)

 

97

Table 7. Correlations among the live rater and five videotape coders on the
Denckla Time to Do Twenty tasks.

 

 

Lowest Highest Average N
correlation correlation correlation
(task) (task)

Live Rater vs. .60 .96 .76 11
Coder 1 (LFPS) (RHPS)
Live Rater vs. .77 .93 .84 34
Coder 2 (RHTAP) (RHPS) I
Live Rater vs. .44 .99 .87 13 I
Coder 3 (RFPS) (RFTAP)
Live Rater vs. .66 .97 .84 5
Coder 4 (LHTAP) (RFPS)
Live Rater vs. .61 .98 .88 13
Coder 5 (RFPS) (LHTAP)
Coder 1 vs. .98 1.0 .99 11
Coder 2 (LF PS) (RHPS)
Coder 2 vs. .59 .99 .90 7
Coder 3 (RHPS) (RFTAP)
Coder 2 vs. .49 .98 .81 15
Coder 5 (RHPS) (LFPS)
Coder 3 vs. .75 .99 .90 5
Coder 4 (LHPS) (RHTAP)
Coder 3 vs. .84 1.0 .90 3
Coder 5 (RHPS) (RHTAP)

 

98

Table 8. Alpha reliabilities for various combinations of raters on the Denckla
Time to Do Twenty Task.

 

 

Reliability N

(alpha)
Live Rater with Rater 1 and Rater 2 .93 11
Live Rater with Rater 3 and Rater 4 .95 4
Live Rater with Rater 3 .86 10
Live Rater with Rater 2 and Rater 5 .94 12
Live Rater with Rater 5 .91 13

 

99

 

Table 9. Correlations among average times (across raters) for each of the 8
tasks on the Denckla Time to Do Twenty motor battery.

 

LFPS LFTAP LHPS LHTAP RFPS RFTAP RHPS RHTAP
LFPS 1.0
LFTAP .25* 1.0
LHPS .74“ .21 1.0
LHTA .08 .19 .25* 1.0
P
RFPS .76" .43** .67“ .27" 1.0
RFTA .37“ .55“ .35“ .24* .45** 1.0
P
RHPS .65" .18 .77“ .53" .56" .29" 1.0
RHTA .08 .14 .22 .87“ .28" .23 .41** 1.0
P
*p<.05; **p<.01
LFPS: Left foot pronation supination RFPS: Right foot pronation supination

LFTAP: Left foot tap RFTAP: Right foot tap
LHPS: Left hand pronation supination RHPS: Right hand pronation supination
LHTAP: Left hand tap RHTAP: Right hand tap

100

REFERENCES

Achenbach, TM. (1991). Manual for t_he Child Behavior Checklist /4-18
and 1991 Profile. Burlington, VT: University of Vermont Department of
Psychiatry.

Adams, W. 8 Sheslow, D. (1995). WRAVMA: Wide Range Assessment of
Visual Motor Abilities (manual). Wilmington, Delaware: Wide Range, Inc.

Aman, C.J., Roberts, R.J., Pennington, B.F. (1998). A neuropsychological
examination of the underlying deficit in ADHD: The frontal lobe vs. right parietal
lobe theories. Developmental Psvcholgqv. 34, 956-969.

American Psychiatric Association. (1968). _Di_agnostic and statistical
manual of mental disorders (2nd ed.). Washington, DC: Author.

American Psychiatric Association. (1980). D_iagnost_ic and statistical
mantmf mental disorders (3rd ed.). Washington, DC: Author.

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

American Psychiatric Association. (1994). _Di_agnostic and statistical
manual of mental disordgs (4th ed.). Washington, DC: Author.

Annett, M. 8 Kilshaw, D. (1984). Lateral preference and skill in
dyslexics: Implications of the right shift theory. flimal of Child Psychology and
Psychiatry. 25(3), 357-377.

Barkley, RA. (1997). Behavioral inhibition, sustained attention, and
executive functions: Constructing a unifying theory of ADHD. Psychological
Bulletin 121(1), 65-94.

Barkley, RA. (1996). Attention-Deficit/Hyperactivity Disorder. In: E.J.
Mash and RA. Barkley (Eds.). Child Psychopathology. New York: Guilford
Press.

Barkley, RA. (1990). Attention Deficit Hyperactivity Disorder: A Hand_book
for Diagnosis and Treatment. New York: Guilford.

Benton, AL. 8 Hamsher, K. deS. (1978). mltilingual aphasia
examination. Iowa City: AHA Associates.

Best, CT. (1988). The emergence of cerebral asymmetries in early
human development: A literature review and a neuroembryological model. In
D.L. Molfese 8 SJ. Segalowitz (Eds.). _Brain Lateralization in Children:
Developmental lmplicatiLns (pp. 5-34). New York: Guilford.

Bettelheim, B. (1973). Bringing up children. La_dies Home Journal. 90,

 

 

28.

Biederman, J., Faraone, S.V., Keenan, K., Benjamin, J., Krifcher, B.,
Moore, C. et al. (1992). Further evidence for family-genetic risk factors in
attention deficit hyperactivity disorder: Patterns of comorbidity in probands and
relatives in psychiatrically and pediatrically referred samples. Archives of
general Psychiatrv,49, 728-738.

101

Bishop, D.V.M., Ross, V.A., Daniels, M.S., 8 Bright, P. (1996). The
measurement of hand preference: A validation study comparing three groups of
right-handers. British Journal of Psychology 87, 269-285.

Cabeza, R. 8 Nyberg, L. (1997). Imaging cognition: An empirical review
of PET studies with normal subjects. Journal of Cognitive Neuyoscience.9(1),1-
26.

Cahn, D.A., Marcotte, A.C., Stern, R.A., Arruda, J.E., Akshoomoff, NA,
and Leshko, LC. (1996). The boston qualitative scoring system for the rey-
osterrieth complex figure: A study of children with attention deficit hyperactivity
disorder. The Clinical Ngropsychologist, 10(4), 397-406.

Cantor-Graae, E., Warkentin, S., Franzen, G., 8 Risberg, J. (1993).
Frontal lobe challenge: A comparison of activation procedures during rCBF
measurements in normal subjects. Neuropsychiatm, Neuropsychology. 8
Behavioral Neurm. 6(2), 83-92.

Cantwell, DP. 8 Baker, L. (1991). Association between attention deficit-
hyperactivity disorder and Ieaming disorders. upumal of Learninggisabilitjfi,
23(2), 88-95.

Carte, E.T., Nigg, J.T., 8 Hinshaw, SP. (1996). Neuropsychological
functioning, motor speed, and language processing in boys with and without
ADHD. J_0I_Imal of A_bnonna_l Child Psychology, 24(4), 481-498.

Carter, C.S., Krener, P., Chaderjian, M., Northcutt, C., 8 Wolfe, V. (1995).
Asymmetrical visual-spatial attentional performance in ADH D: Evidence for a
right hemisphere deficit. Biological Psychiatry. 37, 789-797.

Casey, B.J., Castellanos, F.X., Giedd, J.N. et al. (1997). Implication of
right frontostriatal circuitry in response inhibition and attention-deficit/hyperactivity
disorder. Journal (flhe American Academy of Child and Adolescent Psychiatfl,
36, 374-383.

Castellanos, F.X., Giedd, J.N., Marsh, W.L., et al. (1996). Quantitative
brain magnetic resonance imaging in attention-deficit hyperactivity disorder.
Archives of General Psychiatm, 53, 607-616.

Chess, C. (1960). Diagnosis and treatment of the hyperactive child. _Ne_w
York State Journal of Medfine, 60, 2379-2385.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155-
159.

Cohen, R, Cohen, J., 8 Brock, J. (1993). An epidemiological study of
disorders in late childhood and adolescence- ll. persistence of disorders.
Journal of Child Psychology and Psychiatry. 34(6), 869-877.

Cohen, J. 8 Cohen, P. (1983). Applied multiple regression/correlating
analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum.

Colby, CL. (1991). The neuroanatomy and neurophysiology of attention.
Journal of Child Neurology, 6(Suppl), 890-8116.

Conners, CK. (1997). Conners Rating Scales-Revised. Toronto: Muli-
Health Systems, Inc.

Crawford, H.J. 8 Barabasz, M. (1996). Quantitative EEG magnitudes in
children with and without attention deficit disorder during neurological screening
and cognitive tasks. Child Study Journal, 26(1), 71-85.

102

 

Davidson, R.J. 8 Fox, NA. (1989). Frontal brain asymmetry predicts
infants' response to maternal separation. qumal of A_bnorma| Psycholo 98,
127-131.

DeLong, R. (1995). Medical and pharrnacologic treatment of Ieaming
disabilities. J_ournal of Child Neurology,10(suppl. 1), $92-$95.

Denckla, MB. (1996). A theory and model of executive function: A
neuropsychological perspective. In: G.R. Lyon and NA. Krasnegor (Eds.).
Attent_ign. Memory, and Executive Function. Baltimore: Paul H. Brooks.

Denckla, MB. (1974). Development of motor coordination in normal
children. Developmental Medicine and Child Nflrolpqy. 16, 729-741.

Denckla, M.B., Rudel, R.G., Chapman, 0., 8 Krieger, J. (1985). Motor
proficiency in dyslexic children with and without attentional disorders. Archives of
Neurology, 42, 228-231.

Dennis, M. 8 Whitaker, H. (1976). Language acquisition following
hemidecortication: Linguistic superiority of the left over the right hemisphere.
Brain and Language, 3, 404-433.

Deruelle, C. 8 de Schonen, S. (1995). Pattern processing in infancy:
Hemispheric differences in the processing of shape and location of visual
components. lnfant§ehavior and Developmentfi_8_, 123-132.

de Schonen, S. 8 Mathivet, E. (1990). Hemispheric asymmetry in a face
discrimination task in infants. Chil; Development. 61, 1192-1205.

Douglas, V.l. (1972). Stop, look and listen: The problem of sustained
attention and impulse control in hyperactive and normal children. Canadian
Journal of Behavioural Science. 4(4), 259-282.

Ebaugh, F.G. (1923). Neuropsychiatric sequelae of acute epidemic
encephalitis in children. American Journal of Diseases of Children. 2_5, 89-97.

Elfgren, C.l., Ryding, E., 8 Passant, U. (1996). Performance on
neuropsychological tests related to single photon emission computerised
tomography findings in frontotemporal dementia. British Journal of Psychiatg,
£914), 416-422.

Ernst, M.E., Liebenauer, L.L., King, A.C., Fitzgerald, G.A., Cohen, RM, 8
Zametkin, A.J. (1994). Reduced brain metabolism in hyperactive girls. Journal
of the American Academy of Child and Adolescent Psychiatry. 33(6), 858-868.

Faraone, S.V., Biederman, J., Lehman, B.K., Spencer, T., Norman, D., 8
Seidman, L.J. et al. (1993). Intellectual performance and school failure in
children with attention deficit hyperactivity disorder and in their siblings. Journal
ofipnonnal Psychology. 102(4), 616-623.

Feingold, B. (1975). Why Your Child Is Hyperactive. New York: Random
House.

Follette, WC. 8 Houts, AC. (1996). Models of scientific progress and the
role of theory in taxonomy development: A case study of the DSM. Journal of
Consulting and Clinical Psychology. 64(6), 1120-1132.

Geschwind, N. 8 Galaburda, AM. (1985). Cerebral lateralization:
Biological mechanisms, associations, and pathology: H": A hypothesis and a
program for research. _A_rchive43 of Neurology,42, 428-459, 521-552, 634-654.

103

 

Goodyear, P. 8 Hynd, G.W. (1992). Attention-deficit disorder with
(ADD/H) and without (ADDNVO) hyperactivity: Behavioral and
neuropsychological differentiation. Jourpaj of Child Psychology, 21(3), 273-305.

Grodzinsky, GM. 8 Diamond, R. (1992). Frontal lobe functioning in boys
with attention-deficit hyperactivity disorder. Developmental Neuropsychology,
8(4), 427-445.

Gross-Tsur, V., Shalev, R.S., Manor, 0., 8 Amir, N. (1995).
Developmental right-hemisphere syndrome: Clinical spectrum of the nonverbal
Ieaming disability. qumal ofueaming Disabilities, 28(2), 80-86.

Hart, E.L., Lahey, B.B., Loeber, R., Applegate, B., 8 Frick, P.J. (1995).
Developmental changes in attention-deficit hyperactivity disorder in boys: A four-
year longitudinal study. dpurnal of A_bnormal Child Psychology, 23, 729-750.

Heilman, K.M., Voeller, K.K.S., 8 Nadeau, SE. (1991). A possible
pathophysiologic substrate of attention deficit hyperactivity disorder. Journal of
Dhild Neurology,§(Suppl), $74-$79.

Heilman, K.M. 8 Van Den Abell, T. (1980). Right hemisphere dominance
for attention: The mechanisms underlying hemispheric asymmetries of inattention
(neglect). Neurology, 30, 327-330.

Hinshaw, SP. (1992). Academic underachievement, attention deficits,
and aggression: Comorbidity and implications for intervention. Journal of
Consulting and Clinical Psychology. 60, 893-903.

Hynd, G.W., Semrud-Clikeman, M., Lorys, A.R., Novey, ES, 8 Eliopulos,
D. (1990). Brain morphology in developmental dyslexia and attention deficit
disorder/hyperactivity. Archives of Neurology. 47, 919-926.

Hynd, G.W., Hem, K.L., Novey, E.S., Elioipulos, D., Marshall, R.,
Gonzalez, J.J., 8 Voeller, K.K. (1993). Attention deficit-hyperactivity disorder
and asymmetry of the caudate nucleus. Journal of Child Neurology, 8, 339-347.

Jensen, P.S., Martin, D., 8 Cantwell, D. (1997). Comorbidity in ADHD:
Implications for research, practice, and DSM-IV. J_oumal of Lhe A_merican
Academy (Lthld and Adolescent Psychiatry. 36(8), 1065-1079.

Kahn, E. 8 Cohen, L.H. (1934). Organic drivenness. A brain stem
syndrome and an experience with case reports. New England Journal of
Medicine 210, 748-756.

Kolb, B. 8 Whishaw, l.Q. (1990). Efldamentals of Human
Neuropsychology, 3rd edition. New York: W.H. Freeman and Company.

Laufer, M 8 Denhoff, E. (1957). Hyperkinetic behavior syndrome in
children. Journal of Pediatrics 50, 463-474.

Levin, PM. (1938). Restlessness in children. Archives of Neurology and
Psychiatgy, 39, 764-770.

Levy, J. (1977). The origins of lateral asymmetry. Hemispheric
Asymmetry: What's Right and What's Left. In: J.B. Hellige and SM. Kosslyn
(eds.). Cambridge: Harvard University Press.

Lezak, MD. (1995). Neuropsychological Assessment (3rd ed.). New
York: Oxford University Press.

Lou, H.C., Henriksen, L., Bruhn, P., Bemer, H, 8 Nielsen, J.B. (1989).
Striatal dysfunction in attention deficit and hyperkinetic disorder. Archives of

Neurology, 46, 48-52.

 

 

104

 

Malone, M.A., Couitis, J., Kershner, JR, 8 Logan, W.J. (1994). Right
hemisphere dysfunction with attention-deficit/hyperactivity disorder. Journal of
Child and Adolescent Psychophannacolggy. 4(4), 245-243.

Malone, M.A., Kershner, JR, 8 Swanson, J.M. (1994). Hemispheric
processing and methylphenidate effects in attention-deficit hyperactivity disorder.
Journal of ChiltLNeuroIogy. 9, 181-189.

Mariani, MA. and Barkley, RA. (1997). Neuropsychological and
academic functioning in preschool boys with attention deficit hyperactivity
disorder. Developmental Neuropsycholgqy. 13(1), 1 11-129.

Mattson, A.J., Sheer, DE, 8 Fletcher, J.M. (1992). Electrophysiological
evidence of Iateralized disturbances in children with Ieaming disabilities. Journal
of Clinical and Experimental Neuropsychology, 14(5), 707-716.

McManus, LC. 8 Bryden, MP. (1991). Geschwind's theory of cerebral
lateralization: Developing a formal, causal model. Psychological Bulletin,
_1__1_0_(2), 237-253.

Meyers, J.E. 8 Meyers, KR. (1995). Rey Complex Figure Test all
Recognition TrialzProfessional Manual. Odessa, Florida: Psychological
Assessment Resources, Inc.

Nigg, J.T., Hinshaw, S.P., Carte, E.T., 8 Treuting, J.J. (1998).
Neuropsychological correlates of childhood attention-deficit/hyperactivity
disorder: Explainable by comorbid disruptive behavior or reading problems?
dpumal of Abnormal Psychology. 107, 468-480.

Nigg, J.T., Swanson, J., 8 Hinshaw, S.P. (January, 1996). Covert visual
attention in boys with attention deficit hyperactivity disorder: Lateral effects,
methylphenidate response, and results for parents. Neuropsychologia, 35, 165-
176.

Nottleman, ED. 8 Jensen, PS. (1995). Comorbidity of disorders in
children and adolescents: Developmental perspectives. Advances in Clinical
gild Psychology. 17. 109-155.

Obrzut, J.E., Conrad, P.F., Bryden, MP, 8 Boliek, CA. (1988). Cued
dichotic listening with right-handed, left-handed, bilingual and Ieaming-disabled
children. Neuropsychologia, 26(1), 1 19-131.

Oldfield, RC (1971 ). The assessment and analysis of handedness: The
Edinburgh Handedness Inventory. Niiropsychologia. 9, 97-113.

Offord, D.R., Boyle, M.H., Szatmari, P., Rae-Grant, N.|., Links, P.S.,
Cadman, D.T. et al. (1987). Ontario Child Health Study: II. Six-month
prevalence of disorder and rates of service utilization. _A_rchives pf General
Psychiatry. 44. 832-836.

Pasamanick, B., Rogers, ME, 8 Lilienfeld, AM. (1956). Pregnancy
experience and the development of behavior disorder in children. American
Journal of Psychiatry. 11_2_, 613-618.

Pelham, W.E., Gnagy, E.M., Greenslade, L., 8 Milich, R. (1992). Teacher
ratings of the DSM-III-R symptoms for the disruptive behavior disorders. Journal
of_the American Academy of 9M and Adolescent Psychiatry. 31. 210-218.

Pennington, B.F. (1997). Dimensions of executive functions in normal
and abnormal development. In: Development of t_he PrefrontaJ Cortex: Evolution,

105

 

Neurobiology, and Behavior, Eds: N. Krasnegor, R. Lyon, and P. Goldman-
Rakic. Baltimore: Brookes Publishing Company.

Pennington, BF. 8 Ozonoff, S. (1996). Executive functions and
developmental psychopathology. Journal of Child Psychology and Psychiatm,
31(1), 51-87.

Pennington, B.F., Groisser, D., 8 Welsh, MC. (1993). Contrasting
cognitive deficits in attention deficit hyperactivity disorder versus reading
disability. Devel0pmental Psychology, 29(3), 51 1-523.

Peters, M. (1988). Footedness: Asymmetries in foot preference and skill
and neuropsychological assessment of foot movement. Psychological Bulletin,
fie), 179-192.

The Psychological Corporation (1992). Wechsler Individual Achievement
Test Screener. San Antonio: Harcourt Brace.

Pujol, J., Vendrell, P., Deus, J., 8 Kulisevsky, J. (1996). Frontal lobe
activation during word generation studied by functional MRI. Acta Neurologica
Scandinavica. 93(6), 403-410.

Ransil, B.J. 8 Schachter, SC. (1994). Test-retest reliability of the
Edinburgh Handedness Inventory and Global Handedness Preference
measurements, and their correlation. Percegttgl arui Motor Skills. 79. 1355-
1372.

Reader, M.J., Harris, E.L., Schuerholz, L.J., Denckla, MB. (1994).
Attention deficit hyperactivity disorder and executive dysfunction. Developmental
Neuropsychology, _1__Q(4), 493-512.

Rothlind, J.C., Posner, MI. 8 Schaughency, EA. (1991). Lateralized
control of eye movements in attention deficit hyperactivity disorder. Journal of
Cognitive Neuroscience&(4), 377-381.

Ruff, R.M., Allen, C.C., Farrow, C.E., Niemann, H., 8 Wylie, T. (1994).
Figural fluency: Differential impairment in patients with left versus right frontal
lobe lesions. Archives of Clinical Neurppsychology. 9, 41-55.

Ruff, R.M, Light, R., 8 Evans, R. (1988). The Ruff Figural Fluency Test:
A normative study with adults. Developmental Neuropsychology. 3. 37-51.

Ruff, RM. 8 Parker, 83. (1993). Gender- and age-specific changes in
motor speed and eye-hand coordination in adults: Normative values for the finger
tapping and grooved pegboard tests. Perceptual and Motor Skills 76, 1219-
1230.

Sandberg, S. 8 Barton (1996). Historical development. In: S. Sandberg
(Ed). Hyperactivig Disorders of Childhood. Cambridge University Press.

Sattler, J.M. (1992). Assessment of Childfil‘l (3rd Ed.). San Diego:
Author.

Schachar, R. 8 Tannock, R. (1995). Test of four hypotheses for the
comorbidity of attention-deficit hyperactivity disorder and conduct disorder.
Journal of t_he Amenfln Aca_demy of Child and Adolescent Psychiatm, 34(5),
639-648.

Scheibel, RS. 8 Levin, HS. (1997). Frontal lobe dysfunction following
closed head injury in children: Findings from neuropsychology and brain imaging.
In: N.A. Krasnegor, G.R. Lyon, and PS. Goldman-Rakic. Development of th_e

Prefrontal Cortex: Evolution, Neurobiology, and Behavior. Baltimore: Brookes.
1 06

 

 

Schum, R.L, Sivan, AB, 8 Benton, A. (1989). Multilingual Aphasia
Examination: Norms for children. Clinical Netlropsychologist. 3(4), 375-383.

Shaffer, D., Schwab-Stone, M., Fisher, P. et al. (1993). The Diagnostic
Interview Schedule for Children— Revised Version (DISC-R): I. Preparation, field
testing, interrater reliability, and acceptability. Journal offithe A_merican Acadfemy
of Chich and Adplescent Psychiatry. 32, 643-650.

Solan, HA. (1987). Perceptual norms in grades 4 and 5: A preliminary
report. Journal of the American Optometric Association. 58(12), 979-982.

Solan, HA 8 Mozlin, R. (1986). The correlations of perceptual-motor
maturation to readiness and reading in kindergarten and the primary grades.
J_Qumal of the A_merican Optometric Association. 57(1), 28-35.

Springer, SP. 8 Deutsch, G (1989). Left Brain, Right Brain. Eds.: R.C.
Atkinson, G. Lindzey, and RF. Thompson. New York: W.H. Freeman and
Company. Chs. 8 and 10.

Springer, SP. 8 Eisenson, J. (1977). Hemispheric specialization for
speech in language-disordered children. _NfliropsycholggiaAS, 287-293.

Spreen, O. 8 Strauss, E. (1991). A_Compendium of Neuropsychological
Tests: Administration. Norms. and Commentaty. New York: Oxford University
Press.

Strauss, AA. 8 Werner, H. (1943). Comparative psychopathology of the
brain-injured child and the traumatic brain-injured adult. American Journal of
Psychiatry. 99, 835-838.

Still, GP (1902). Some abnormal psychical conditions in children.
Lancet 1, 1008-1012, 1077-1082, 1163-1168.

Swanson, J.M., Posner, M., Potkin, S., Bonforte, S., Youpa, D., Flore, C.,
Cantwell, D. 8 Crinella, F. (1991). Activating tasks for the study of visual-spatial
attention in ADHD children: A cognitive anatomic approach. Journal of Child
Neurology, 6(suppl), S1 19-S125.

Taylor, E. (1986). The Overactive Child. Philadelphia: Lippincott.

Trites, R.L. (1977). Neuropsychological Test Manual. Ottawa, Ontario,
Canada: Royal Ottawa Hospital.

Tucker, 0M. 8 Williamson, PA. (1984). Asymmetric neural control
systems in human self-regulation. Psychological Review. 91(2), 185-215.

Van der Meere, J.J. (1996). The role of attention. In S. Sandberg (Ed.),
Hyperactivity Disordersficmhood (pp. 111-148). Cambridge, England:
Cambridge University Press.

Van der Meere, J., Vreellng, H.J., 8 Sergeant, J. (1992). A motor
presetting study in hyperactive, learning disabled and control children. Journal of
Child Psychology and Psychiatg, 33, 1347-1354.

Van der Meere, J. 8 Sergeant, J. (1987). A divided attention experiment
with pervasively hyperactive children. Jo_umal of Apnormi Child lisycholo 15,
379-391.

Voeller, K.K.S. (1991). What can neurological models of attention,
intention, and arousal tell us about attention-deficit hyperactivity disorder?
Jo_utnal of Neuropsychiatrv. 3(2), 209-216.

Voeller, K.K.S 8 Heilman, KM. (1988). Attention deficit disorder in
children: A neglect syndrome?. Neurology,38, 806-808.

 

107

 

Waber, DP. 8 Holmes, J.M. (1985). Assessing children’s copy
productions of the Rey-Osterrieth complex figure. Journal of Clinical and
Experimental NeuropsychomL 7, 264-280.

Wechsler, D. (1991). Wechsler Intelligence Sgle for Chilcuenu Third Ed.
New York: Psychological Corp.

Weinberg,W.A. (1993). Vigilance and its disorders. Neurologic Clinics,
1_1 (1 ), 59-78.

Woodcock, R.W. 8 Johnson, M.B. (1989,1990). ngcock-Jghnson
Psycho-EducationalButtery-Revised. Chicago: Riverside.

108

 

HICH GRN STRTE UNIV.

I LIBRRRIES
II IIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIII
9301

312 8341986