A. filo} “I s .
A". .... 1...».fi,..$.
2.,
hm.
r.

aw};
”Lu.
.2

hm...
‘
spam}

5. é? ..

a
#1»... Am. +1.,
,}

ti
4!

is;

 

.

’3'

.qfih... n

 

 

 

 

 

 

 

 

 

 

 

.2
”Fun..."

.195!
(Hi. ‘
fimuwwaffie
..

 

,.....~.7,u.ras..s.l
1 l.
.:l...u.fl|

L3
F! . é
.53.... 6.. Hawks}.
1t. 5.. (X.

59

$4. . 1..

- J I». L: J?
. r . . a

.2: :3... 1-3:...» i..n
.. ti...» 31...}:

3:5 . ...u_..f.n T
....:.I.:....$. :. a

. .
:2 ,. :2.

C, 3&1!

 

LIDHAHY

' mars . .
T; Michigan State
2. not Universu‘y

 

This is to certify that the
dissertation entitled

HORMONAL ASSOCIATIONS WITH CHILDHOOD ADHD AND
ASSOCIATED TRAIT AND NEUROPSYCHOLOGICAL
MECHANISMS

presented by

Michelle M. Martel

has been accepted towards fulfillment .
of the requirements for the

Ph.D. degree in Psycholgy

 

 

A

 

Ma' r Proféfigops Signature

Date

MSU is an aflinnative-action, equal-opportunity employer

. .II-I-o-u-o-v-l—I-O-a-I-I-D- -.-.--o-----I-u-o--o-o--u-o-o-u-o-n-o-

0--.-o-u-v-n-o-o-u--.—.—

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Acc&Pres/ClRC/DateDue.indd

HORMONAL ASSOCIATIONS WITH CHILDHOOD ADHD AND ASSOCIATED
TRAIT AND NEUROPSYCHOLOGICAL MECHANISMS
By
Michelle M. Martel

A DISSERTATION
Submitted to
Michi an State University
in partial ful lllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY

Department of Psychology
2007

ABSTRACT

HORMONAL ASSOCIATIONS WITH CHILDHOOD ADHD AND ASSOCIATED
TRAIT AND NEUROPSYCHOLOGICAL MECHANISMS

By
Michelle M. Martel

Attention-Deficit/Hyperactivity Disorder (ADHD) is a childhood disorder that has a sex-
biased prevalence rate of between 2:1 and 9:1 (malezfemale). Gonadal hormones such as
testosterone and estradiol are important, yet under-explored, candidates to help explain
this male-biased prevalence rate. Testosterone and estradiol may act via psychological
mechanisms related to ADHD. Here, these are conceived as neuropsychological
executive function and two personality traits: conscientiousness and reactive control.
Participants in this study were 312 children (168 with ADHD; 178 boys), who completed
a multi-stage clinical diagnostic process, hand scans, venipuncture blood draws, and a
neuropsychological assessment battery. Primary caregivers completed trait rating scales
and teachers completed ratings of ADHD behaviors. Girls had higher levels of
conscientiousness than boys within the control group, whereas girls and boys in the
ADHD group had comparable, low levels of conscientiousness. Within the ADHD group,
girls had higher levels of reactive control than boys. However, within the control group,
girls and boys did not differ in reactive control. Finger-length ratio, serving as an index of
prenatal testosterone exposure, was related to inattentive symptoms. The personality trait
of lower conscientiousness, or thoughtful regulation, statistically mediated the
relationship between finger-length ratios and inattentive ADHD symptoms. Contrary to
hypotheses, circulating estradiol was not related to ADHD symptoms in girls. In contrast,

high levels of circulating testosterone were related to fewer ADHD symptoms in girls. It

is concluded that prenatal testosterone increases risk for ADHD in boys, perhaps through
organizational effects on dopamine brain systems. Higher levels of circulating
testosterone may be protective for girls. Differences in gonadal hormone levels may help

to explain the sex-biased prevalence rate and sex differences in correlates observed in

childhood ADHD.

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. vii
LIST OF FIGURES ................................................................................ viii
BACKGROUND ............................................................................................................. 1
Overview of Introduction ............................................................................................. 1

I. Background on ADHD ............................................................................................. 2
Overview ................................................................................................................. 2
What is ADHD? ...................................................................................................... 3

Sex Differences in ADHD Expression ..................................................................... 5
Neural Structures Implicated in ADHD .................................................................... 6

11. Executive Function (EF) and Temperament Traits as Mechanisms of ADHD .......... 8
Overview ................................................................................................................. 8
Neuropsychological Executive Function .................................................................. 8
Temperament ........................................................................................................... 9
Integration of Executive and Temperamental Views of Regulation ........................ 12
Summary ............................................................................................................... 14

III. Hormonal Influences on Development and on DA circuitry ................................. 15
Overview ............................................................................................................... 15
Organizational Effects of Sex Hormones ............................................................... 15
Effects of Circulating Levels of Hormones ............................................................ 18

iv

Summary ............................................................................................................... 21
IV. Hormonal Influences on ADHD Symptoms, Mechanisms, and Correlates ............ 22

Organizational Effects of Hormones on ADHD Symptoms, Mechanisms, and
Correlates .............................................................................................................. 22

Effects of Circulating Levels of Hormones on ADHD Symptoms, Mechanisms, and

Correlates .............................................................................................................. 25
Conclusion ................................................................................................................ 32
METHOD ..................................................................................................................... 34
Participants ................................................................................................................ 34
Multistage recruitment and diagnostic assignment process ..................................... 36
Measures ................................................................................................................... 38
Executive Function (EF) ........................................................................................ 38
Temperament Traits ............................................................................................... 40

Sex Hormones ....................................................................................................... 40
Data Analysis ........................................................................................................ 44
RESULTS ..................................................................................................................... 48
Descriptive Statistics and Preliminary Data Checks ................................................... 48

Preliminary Data Check 1: Subtype differences in finger-length ratios,
neuropsychological EF, or temperament traits in children with ADHD and controls .. 50

Preliminary Data Check II: Sex and diagnostic differences in finger-length ratios,
neuropsychological EF, or temperament traits in children with ADHD and controls ..50

Question 1: Do levels of prenatal testosterone as inferred from right 2D:4D predict (a)
more inattentive symptoms and (b) more impaired EF or conscientiousness in boys?. 55

Question 2: Does low conscientiousness mediate the relationship between prenatal
levels of testosterone and inattentive ADHD symptoms in boys? ............................... 5 6

Question 3: Does (a) more adaptive reactive control predict higher or more optimal
levels of circulating estradiol and (b) do higher levels of circulating estradiol predict
less hyperactive-impulsive ADHD symptoms in girls? ............................................... 58

Question 4: Does circulating estradiol mediate the relationship between reactive
control and hyperactive-impulsive ADHD symptoms in girls? ................................... 59

Question 5: Do lower ratios of testosterone to estradiol predict (a) less inattentive or
hyperactive-impulsive ADHD symptoms and (b) better EF/conscientiousness or

reactive control in boys and girls? .............................................................................. 59
DISCUSSION ............................................................................................................... 65
Sex Differences in Traits and EF ............................................................................... 66
Prenatal Testosterone Effects and ADHD .................................................................. 69
Circulating Hormonal Effects and ADHD .................................................................. 70
Limitations ................................................................................................................ 73
Summary ................................................................................................................... 74
REFERENCES .............................................................................................................. 76

vi

LIST OF TABLES

Table 1 Sample Demographics ..................................................................... 35
Table 2 Descriptive Statistics by Diagnostic Group. ....................................................... 49
Table 3 Descriptive Statistics by Sex Within Diagnostic Group .............................. 51

Table 4 Correlations among Hormone Measures, Clinical Symptoms, and Traits and
Cognitive Control .................................................................................... 54

vii

LIST OF FIGURES

Figure 1 Mutual Influence Model of Behavioral Control in ADHD-C ...................... 14
Figure 2 Mediator Model for Prenatal Testosterone Effects on ADHD ..................... 30
Figure 3 Mediator Model for Circulating Estradiol Effects on ADHD ...................... 32
Figure 4 Conscientiousness Mediates the Relationship Between Prenatal Testosterone
and Inattentive ADHD Symptoms in Boys ...................................................... 57
Figure 5 Higher Levels of Testosterone are Related to Fewer Inattentive ADHD
Symptoms in Girls .................................................................................. 60
Figure 6 Higher Levels of Testosterone are Related to Less Response Variability in Girls
......................................................................................................... 60
Figure 7 High Testosterone: Estradiol Ratios are Related to Fewer Inattentive ADHD
Symptoms in Girls .................................................................................. 62

Figure 8 High Testosterone: Estradiol Ratios are Related to Fewer Hyperactive-Impulsive
ADHD Symptoms in Girls ......................................................................... 62

Figure 9 High Testosterone: Estradiol Ratios are Related to Increased Conscientiousness
in Girls ................................................................................................ 63

viii

BACKGROUND
Overview of Introduction

The introduction for this dissertation is comprised of four sections. The first
section provides background information on ADHD including sex differences in ADHD
expression and a summary of the neural structures implicated in ADHD. The second
section examines neuropsychological executive function and temperament traits as
behavioral markers of developing neural structures relevant to ADHD. The third section
provides background on hormonal influences on early development with a focus on the
sex hormones, testosterone and estradiol. These two hormones are the focus because they
are likely candidates for sex differences in hormonal levels across children’s
development. Moreover, there is substantial empirical support for their effects on
cognitive functioning and behavior. In the fourth section of the introduction,
hypothesized hormonal influences on ADHD and its mechanisms are considered.
Hypotheses are then explicated.

Overall, this dissertation argues that ADHD symptoms, its key mechanisms (i.e.,
executive function and certain temperament traits), and the underpinning neural
structures, may be influenced by gonadal hormones. That is, testosterone and estradiol
may shape key neural structures, which in turn alter neuropsychological and affective
(temperamental) functioning, leading to ADHD symptoms. If so, then these relations
should be detectable via association of prenatal and/or circulating hormone levels to
neuropsychological executive function, traits, and symptoms. In this way, gonadal

hormones may make males more vulnerable to the development of ADHD than girls by

differentially impacting the behavior and neural mechanisms that give rise to this
childhood disorder.
I. Background on ADHD
Overview

Attention-Deficit/Hyperactivity Disorder (ADHD), a childhood behavioral
disorder characterized by extremes of inattention, hyperactivity, and impulsivity, arises
early in childhood (DSM-IV; American Psychiatric Association [APA], 2000; Hart,
Lahey, Loeber, Applegate, and F rick, 1995; Lahey et al., 2005; Nigg, 2006). One of the
most-replicated ADHD research findings is the predominant male-to-female ratio,
estimated to be between 2:1 and 9: l. The disorder cannot be fully understood without
explaining males’ increased vulnerability, yet the causal mechanisms of this male-biased
prevalence remain a mystery. Gonadal hormones are one potential source of this sex-
biased prevalence, yet little attention has been given to hormonal mechanisms. Gonadal
hormones exert dramatic and long-lasting effects on neural circuitry, cognition, and
behavior during the prenatal period and early childhood. As I outline, some of these
effects appear to be quite relevant to inattentive and hyperactive-impulsive ADHD
symptoms. They also appear related to the neuropsychological executive functions and
temperament/personalityl traits that may be involved in ADHD. Thus, it is surprising that

gonadal hormones are seldom investigated in regard to ADHD.

 

' In children, traits could arguably be termed temperament or personality since they are likely to encompass
some of each. For simplicity, temperament will be used to describe childhood character traits in this
document, although it is acknowledged that these traits doubtless incorporate both temperament and
personality components.

What is ADHD?

ADHD, as conceptualized in the DSM-IV (APA, 2000), is comprised of two basic
symptom domains (inattention-disorganization [e.g., “does not seem to listen when
spoken to directly”] and hyperactivity/impulsivity [e.g., “often fidgets or squirms” or
“often interrupts”]). These emanate in three subtypes: primarily inattentive, primarily
hyperactive-impulsive, and combined. Each subtype is defined in the DSM-IV as
requiring the presence of at least six symptoms (or at least twelve symptoms for the
combined subtype), lasting at least six months, beginning before seven years of age, and
causing marked impairment. It is important to distinguish these two symptom domains
because, as discussed later, they are hypothesized to have different neural substrates and
correlates.

Symptoms appear to follow a developmental trajectory such that problems with
impulse control arise early and problems with inattention emerge somewhat later.
Hyperactivity/impulsivity often drops off later in childhood or during adolescence
(Barkley, 2003; Olson, 1996). The mean age of onset is typically between ages three and
four, although the symptoms of ADHD vary for each individual across the lifespan (Hart
et al., 1995).

ADHD prevalence is estimated at between three and five percent for children in
the United States (APA, 2000). It has a high co-occurrence with other disruptive behavior
disorders, including Oppositional Defiant Disorder (ODD) and Conduct Disorder (CD;
Barkley, 2003). ODD is a behavioral disorder characterized by patterns of defiant and
hostile behavior (APA, 2000). CD is a behavioral disorder in which the basic rights of

others and social norms are consistently violated (APA, 2000). Thus, it is of interest

whether effects of hormones on ADHD will be specific to ADHD or will also be related
to disruptive behavior problems more generally.

The core behavioral symptom domains of inattention and hyperactivity-
impulsivity are important to understand. They are associated with a variety of
impairments, including deficits in cognition, language, adaptive functioning, motor
development, emotion, school performance, task performance, neuropsychological
executive functioning, peer status, psychological comorbidity, and medical and health
risks (Barkley, 2003; Hinshaw, 2002; Olson, 1996). The severity of these impairments
underscores the importance of understanding the etiology of this disorder.

With regard to causal mechanisms that lead to ADHD, recent theories postulate
two pathways, one feeding into inattentive and the other into hyperactive-impulsive
symptoms. The first path involves an executive function component, whereas the second
path involves a motivational/affective component (Sonuga-Barke, 2005). According to
this theory, mesocortical dopamine (DA) circuitry (i.e., the striatum and the caudate) is
implicated in executive control problems, especially faulty effortful or strategic response
inhibition. This, in turn, gives rise to the inattentive-disorganized symptoms of ADHD
(Nigg & Casey, 2005; Sagvolden, Johansen, Aase, & Russell, 2004; Sonuga—Barke,
2005). More recent work clarifies that the right lateralized striatal-frontal circuit is likely
involved in response suppression. In contrast, mesolimbic DA circuitry (i.e., nucleus
accumbens) is implicated in temperamental incentive approach problems (N igg, 2006). In
turn, it is thought to shape hyperactive-impulsive symptoms of ADHD (Sagvolden et al.,
2004; Sonuga—Barke, 2005). At the level of psychological mechanism, the pathway

related to inattentive symptoms operates through cognitive control. The pathway related

to hyperactive-impulsive symptoms Operates through affective dysregulation.

Returning to the level of the brain, a communality shared by both pathways is the
involvement of DA dysregulation, specifically hypodopaminergic extracellular
functioning (Volkow et al., 2004; Volkow, Wang, Fowler, & Ding, 2005). The present
dissertation exploits this theory to highlight the influence gonadal hormones may have on
the development of childhood ADHD via distinct pathways corresponding to these two
routes. This argument is based on the likelihood that hormones affect the development of
DA circuits in a variety of ways, discussed later.

Sex Differences in ADHD Expression

To date, most research on sex differences in relation to ADHD has focused on
illustrating the potential for gender—biased ADHD referral rates. Comparatively little
research has addressed sex differences in etiological correlates of the disorder
(Biederman et al., 2002; Hartung et al., 2002; Rutter, Caspi, & Moffitt, 2003). This
continuing relative shortage of studies is surprising since over a decade ago there was
consensus that such knowledge was needed (Arnold, 1996).

As a result, the initial data on sex differences in ADHD are intriguing but
inconclusive. Clinically, boys with ADHD are often believed to be more inattentive and
hyperactive-impulsive than girls with the disorder (Gaub & Carlson, 1997; Gershon,
2002, Hartung et al., 2002). Girls with the disorder appear to be twice as likely as boys to
evidence the inattentive type of ADHD (Biederman et al., 2002). Further, some research
suggests that there are differences in the pattern of comorbidity between boys and girls.
Specifically, girls with ADHD may experience relatively more comorbid internalizing

psychopathology and less externalizing psychopathology than boys with ADHD (Arnold,

1996; Biederrnan et al., 1999; Gaub & Carlson, 1997). Girls with ADHD also appear to
be at a higher risk for later substance abuse than boys (Biederman et al., 2002; Hinshaw,
2002). Sex differences in comorbidity may also be based on subtype, with boys with the
combined subtype more likely to experience mood disorders and girls with the inattentive
subtype more likely to experience anxiety disorders (Bauermeister et al., 2007). Further,
girls with ADHD may have higher rates of speech and language disorders than boys
(Hinshaw, 2002; James & Taylor, 1990).

Sex differences on cognitive tasks are worth noting, although their implications
for ADHD remain unclear. Some data suggest that girls with ADHD may have more
cognitive impairment than boys with the disorder (Biederman et al., 1999; Gaub &
Carlson, 1997; Newcorn et al., 2001). However, other studies have found few to no sex
differences on neuropsychological tasks (Rucklidge & Tannock, 2002; Seidman et al.,
1997). Further, in one study, boys with the combined subtype of ADHD had deficits in
motor response inhibition (an executive skill) whereas boys with the inattentive subtype
did not (N igg, Blaskey, Huang-Pollock, & Rappley, 2002). In this same study, girls with
either the combined or the inattentive subtype exhibited these same deficits. Other data
suggest that girls with ADHD are autonomically hypo-aroused and have more brain
abnormalities than boys with ADHD (Ernst et al., 1994; Herrnens et al., 2004).

Neural Structures Implicated in ADHD

Childhood ADHD is associated at the group level with several key structural and
functional brain abnormalities that may be related to hormonal action. Two of the most
well-replicated structural findings are reduced volume in (a) the right cortical hemisphere

and (b) the dorsolateral, orbital, and anterior cingulate areas of the prefrontal cortex

(Casey et al., 1997a; Castellanos et al., 2001; Durston, 2003; Hill et al., 2003; Nigg,
2006; Seidman, Vlaera, & Makris, 2005). Other studies find that children with ADHD
have reductions in the volume of the caudate in the basal ganglia, the corpus callosum,
and the cerebellar vermis (Casey et al., 1997b; Castellanos et al., 2001; Durston, 2003;
Hill et al., 2003; Seidman, Vlaera, & Makris, 2005; Willis & Weiler, 2005).

Altogether these findings suggest that there may be widespread cerebral volume
reduction in children with ADHD, potentially most evidenced in the frontal-striatal and
cerebellar regions of the brain. One commonality shared by these regions is that they
have extensive dopaminergic projections that are characterized by reduced extracellular
DA in children with ADHD (N igg & Casey, 2005; Tucker & Williamson, 1984; Volkow
et al., 2005).

Summary

ADHD is a childhood behavior disorder with a sex-biased prevalence rate
favoring males. Yet sex differences in the expression of the disorder are speculative and
under-researched. ADHD’s neural underpinnings are better-established. The frontal-
striatal and cerebellar regions of the brain, characterized by extensive dopaminergic
circuitry, appear to be altered in children with ADHD. Specifically, there are reduced
levels of extracellular DA in these neural regions. As outlined in the next sections,
gonadal hormones appear to influence the development of these neural regions. Further,
these regions are related to the causal mechanisms of ADHD at the psychological level,
namely problems with executive functions and motivational/affective deficits

(Castellanos & Tannock, 2002; Nigg, 2006; Sagvolden et al., 2004).

11. Executive Function (EF) and Temperament Traits as Mechanisms of ADHD
Overview

EF and temperament will be introduced and defined as hypothetical core
mechanisms of ADHD. EF and temperament are constructs at the psychological level of
analysis that are amenable to behavioral measurement. However, they also serve as
indices of the neural dysfunction implicated in the disorder. If hormones alter the
functioning of the dopaminergic circuits by making them less efficient, then this may be
detectable on behavioral measures related to those same circuits. Next, a framework for
integrating EF and temperament is explicated.
Neuropsychological Executive Function

Deficits in neuropsychological EF are one of the most commonly-studied
correlates of ADHD (Barkley, 1997; 2003; Nigg etal., 2002), although little attention has
been given to sex differences and hormonal influences. EF has most often been defined
as the ability to regulate cognition and behavior in order to attain a goal (N igg, 2000;
Pennington & Ozonoff, 1996). Barkley (1997; 2003) developed a hybrid model of ADHD
that postulates that several EFs interact with behavioral inhibition and motor control to
lead to the development of ADHD. Others have suggested that set shifting, working
memory, response inhibition, and planning are important components of EF (Pennington
& Ozonoff, 1996). New research has also examined response variability, or intra-
individual variability, as an important component of cognitive functioning in ADHD
(Russell et al., 2006). EF relies heavily on prefrontal-striatal circuitry and dopaminergic
function (Casey, 2005; Nigg, 2000, 2006; Pennington & Ozonoff, 1996), and intra-

individual variability may reflect a catecholaminergic deficiency that influences the

ability to modulate oscillations in neuronal activity (Castellanos et al., 2005). The
components of EF (i.e., motor inhibition, output speed, planning, working memory, and
response inhibition), response variability, and their biological underpinnings are less
efficient in children with ADHD as a group (Martinussen, Hayden, Hogg-Johnson, &
Tannock, 2005; Nigg et al., 2002; Seidman, Valera, & Makris, 2005; Willcutt, Doyle,
Nigg, Faraone, & Pennington, 2005).

Within the broad domain of EF, deficits in cognitive control in the form of
response inhibition have been emphasized in ADHD. Children with ADHD are impaired
in their ability to withhold a dominant response in order to perform a subordinate one
(Barkley, 1997; Schacher, Tannock, Marriott, & Logan, 1995). Primary among tests that
assess this ability is the Logan (1994) Stop Task. It asks that a child withhold the well-
learned response of pushing a key when a tone sounds. Many studies suggest that
response inhibition is slower or less efficient in children with ADHD (for a review, see
Willcutt et al., 2005). This problem is thought to stem from abnormalities in the
frontodorsal striatal circuit, including the right anterior cingulate (Alexander, Crutcher, &
DeLong, 1991). That circuit relies heavily on modulation by DA.

Temperament

Although various temperament traits have been linked to externalizing disorders
and ADHD (Huey & Weisz, 1997; Martel & Nigg, 2006; Nigg et al., 2002; White, 1999),
little research has examined the possibility that hormones may be related to sex
differences in these relations. Girls with ADHD have similar personality characteristics
and manifest similar interpersonal and emotional problems as boys with the disorder

(Befera & Barkley, 1985; Greene et al., 2001). However, sex differences in the normal

development of child temperament are established. Girls have higher levels of self-
regulation (McCabe, Cunnington, & Brooks-Gunn, 2004). Specifically, girls show more
self-regulated compliance, higher effortful control and conscientiousness, lower
impulsivity, and more competent social functioning than same age boys (Fabes et al.,
1999; Kochanska, Tjebkes, & Forman, 1998; Zahn-Waxler, Schmitz, Fulker, Robinson,
& Emde, 1996). A meta-analysis on sex differences in temperament in children suggests
that girls exhibit higher effortful control than boys (Else-Quest, Hyde, Goldsmith, & Van
Hulle, 2006). If these traits give rise to ADHD, then girls and boys with ADHD might
show sex differences in the expression of ADHD symptoms that are related to an
outgrth of normal sex differences in temperament traits. Alternatively, different traits
may contribute to the disorder for girls and boys.

Many temperament/personality constructs relate to behavioral control and thus
potentially to ADHD. Well-studied examples are effortful control, reactive control, ego
control, behavioral/emotional regulation, and conscientiousness. These constructs have a
great deal of overlap with one another. Thus, their patterns of relation to ADHD are
similar (Eisenberg et al., 2000; Nigg, 2000). However, there is a distinction between
effortful control and reactive control. Reactive control has theoretical overlap with
motivational, or affective, deficits.

Incentive responding is widely viewed as important in ADHD. For example, the
Dynamic Developmental Theory of ADHD (Sagvolden et al., 2004) suggests that altered
reward/motivation is a core deficit. Altered reward/motivation in turn is believed to be
due to dysfunction of DA system branches. These include frontoventral striatal reward

circuits, fiontoamygdala circuitry, and mesolimbic branches, including nucleus

10

accumbens (Casey, 2005; Nigg & Casey, 2005; Sonuga-Barke, 2005). This concept of
altered reward/motivation is consistent with one of Sonuga-Barke’s (2005) two pathways
to ADHD: the motivational/affective pathway.

Reactive control is conceptually related to this reward/motivation argument
because it has been defined as a form of control that is reactively modulated (Murray &
Kochanska, 2002; Valiente et al., 2003). Reactive responding is directly dependent on
immediate incentives (or cues that there is an immediate incentive available) rather than
mentally represented future incentives. Traits like effortful control/conscientiousness, in
contrast, are defined as flexibly managed control (Murray & Kochanska, 2002; Valiente
et al., 2003). Effortful traits are related to behavioral patterns that facilitate the attainment
of a later goal that may be held in working memory.

Following on that distinction, distinctions can be drawn between two kinds of
control: top down effortful control (i.e., most related to executive control) and bottom up
reactive control (i.e., most related to motivational processes). Effortful control (defined as
the temperament trait of thoughtful or planful regulation) and conscientiousness (defined
as the personality trait of planfulness) are very similar trait constructs. Although some
specialists in personality see distinctions between them at the item level, for purposes of
this study, they are treated as measuring virtually the same domain. Both have been
equally linked to cognitive control. Further, they both appear, at least conceptually, to be
underpinned by similar frontal-striatal neural circuitry (Rothbart & Posner, 2006).
Finally, they both refer to the ability to suppress a dominant response in order to respond
to contextual demands and are thought of as examples of “top-down” processing

(Rothbart & Rueda, 2005). Reactive control, most related to incentive-based motivational

ll

processes and their neural underpinnings, refers to reactive processes that are relatively
automatic and therefore not under conscious control. They are more akin to “bottom up”
responses to stimuli or incentives (Murray & Kochanska, 2002; Valiente et al., 2003).

Both of these major traits, top down control and bottom up control, have been
related to children’s adaptation, social functioning, and problem behaviors (Eisenberg et
al., 2001; 2004). Valiente and colleagues (2003) found links between reactive control,
effortful control, and externalizing problems. Further, as children got older, effortful
control was a stronger predictor of future externalizing behavior than was prior
externalizing behavior, even after statistically adjusting for reactive control. A review of
the literature by White (1999) suggested associations between low levels of control, low
levels of conscientiousness, and ADHD. This supposition was confirmed in a large multi-
site study of personality traits and ADHD (N igg et al., 2002). Thus, temperament and
personality research has consistently found children and adults with ADHD to be
characterized by low levels of conscientiousness (Nigg et al., 2002; Nigg, Goldsmith, &
Sachek, 2004). A somewhat smaller literature suggests that children with ADHD have
lower levels of temperamental effortful or reactive control (Huey & Weisz, 1997).
Integration of Executive and Temperamental Views of Regulation

The dopaminergic brain circuitry implicated in ADHD can be assessed by
neuroimaging, but of central interest here are functional abilities at the psychological
level of analysis. These can be indexed by laboratory behavioral measures (e.g.,
neuropsychological tests) and by behavioral trait measures (e.g., temperament), providing
a convergent evidence approach. Based on the proceeding, EF and temperament trait

control can be reformulated into an integrated model with (a) effortful (i.e., EF and

12

effortful control/conscientiousness) and (b) reactive (i.e., motivation/delay aversion and
reactive control) components. These are then hypothesized to be underpinned by similar
neural circuitry whether assessed via cognitive measures or behavioral ratings (N igg,
2006; Nigg & Casey, 2005).

As schematized in Figure 1, executive control (a neuropsychological construct)
and effortful control/conscientiousness (a temperament construct) are thought to involve
“top-down” processing, or processing in the prefrontal regions that modulates activity in
limbic, striatal, as well as posterior cortical areas (N igg, 2006; Nigg & Casey, 2005). In
contrast, reactive temperamental response involves a more “bottom-up” processing style,
generated by activity in the limbic areas and then interrupting processing in the prefrontal
regions (N igg, 2006; Nigg & Casey, 2005). Thus, effortful control is associated with
activity in the prefrontal cortex whether it is viewed as cognitive control and EFs or as
the temperamental construct of effortful control and conscientiousness (N igg, 2006; Nigg

& Casey, 2005).

13

Mutual influence model of
behavioral control in ADHD -C

Limbic-frontal L]

 

 
   

 

Neural PFC-striatal

 

 

 

 

 

 

. - Response suppression Incentive reactivity.
Cognitive (Inhibition); executive \ 6 speed. learning
Iundion
; l
florti u control Reactive control.
TW “EBXIbIe (”ping ' > H < behaw'oral inhibition
Personality l 1
l tlenti , Impulsivity. hypemcfiviy.
PM)! 3% ax J K inpubive aggression ax

Figure 1. Mutual Influence Model of Behavioral Control in ADHD-C

In contrast, reactive response is seen as more dependent on subcortical areas such
as the nucleus accumbens (related to approach processing) and potentially the amygdala
(related to fear and withdrawal processing). Reactive control may be related to a general
approach system. Since DA is involved in the neural circuitry that underpins executive
control and temperamental control, then hormones that influence DA may influence these
mechanisms, and this may be a route to their influence on ADHD symptoms (N igg &
Casey, 2005; Sonuga-Barke, 2005).

Summm

Deficits in neuropsychological EF and two key domains of temperament,
effortful control and reactive control, are likely correlates of ADHD. The theoretical
perspective adopted here, drawn from the work of several different theorists, suggests

that the mesocortical DA circuit (including striatum) is implicated in (a) executive control

14

problems, such as response inhibition, (b) the temperament trait of effortful
control/conscientiousness (Rothbart & Bates, 1998), and (c) the ADHD symptom domain
of inattentive symptoms. In contrast, the nucleus accumbens and the mesolimbic DA
circuit are implicated in (a) temperamental reactive response, especially approach, and
(b) hyperactive-impulsive symptoms of ADHD (Nigg, 2006; Sagvolden etal., 2004;
Sonuga-Barke, 2005). Both circuitries are influenced by prenatal and circulating levels of
gonadal hormones, as discussed in the following sections.
III. Hormonal Influences on Development and on DA circuitry
Overview

Alterations in hormones seem like appealing candidates for explaining sex
differences in the prevalence and expression of certain childhood disorders, including
ADHD, although such relations are seldom examined and remain unclear. The reasons
for this apparent connection are many, deriving from the many ways in which hormones
influence the neural circuitry implicated in ADHD. Gonadal hormones exert
organizational effects and the activational effects. Although these mechanisms are
discussed as separate for our purposes, it is acknowledged that they are likely interactive
and dynamic in nature. I discuss each of these mechanisms in turn.
Organizational Effects of Sex Hormones

Since ADHD emerges in childhood, gonadal hormones might exert their effects
on ADHD through organizational effects on brain development. Organizational effects on
the brain usually occur during a critical period of development. During this period,
hormones permanently affect the developing brain, thus indirectly altering behavior even

into adulthood (Breedlove & Harnpson, 2002). In animals, organizational effects occur in

15

part through the aromatization of androgens (i.e., a large class of hormones that includes
testosterone) into estradiol. However, in humans, most of these effects appear to occur
directly through the effects of androgens, especially testosterone (Morris, Jordan, &
Breedlove, 2004). This difference in organizational effects between animals and humans
highlights the need for human studies of hormonal influences on behavior in childhood.

Sexual differentiation, a term used to describe how organisms develop into males
and females, is a process that is integrally tied with organizational effects that occur
during prenatal development. Thus, it is important to understand relevant theories of this
developmental process. There are three theories of sexual differentiation: classical, active
feminization, and gradient. These three models are not necessarily mutually exclusive,
and thus all three can potentially be considered to address phenomenon such as ADHD.

The classical theory of sexual differentiation states that androgens cause
masculine development, while the absence of androgen is linked to feminine
development (Hines, 2002). This model of sexual differentiation has been well-studied in
relation to prenatal animal development. This line of research suggests that testosterone,
acting through the modulation of cell death rate, is the most likely source of
organizational effects during the prenatal period (Morris, Jordan, & Breedlove, 2004).
Testosterone, by differentially modulating the rate of cell death in various regions of the
brain, may lead to the development of relatively larger or smaller size brain regions in
males as compared to females. It can be hypothesized that hormonal effects of prenatal
testosterone may make boys more vulnerable to the development of ADHD in early
childhood though the modulation of cell death rate.

The second, gradient model of sexual differentiation states that hormones

16

influence not just behavioral differences (e.g., in cognition, childhood play, and
aggression) between sexes, but also behavioral differences within sex (Collaer & Hines,
1995; Hines, 2002). Thus, levels of prenatal and childhood testosterone and estrogen may
jointly influence the development of DA neural circuitry in boys and girls, and thus
indirectly affect cognitive profiles and behavioral symptoms.

The third model of sexual differentiation, the active feminization model, states
that ovarian hormones may play an active role in the feminization of neural circuitry and
behavior (Collaer & Hines, 1995; Hines, 2002). Estrogen (i.e., a large class of hormone
that includes estradiol) may influence and provide some neuroprotection for the
development of DA circuitry in particular. Thus, it can be hypothesized that estrogen
protects girls fiom the development of ADHD by protecting DA circuitry from insult.
Prenatal Testosterone and Classical, Organizational Effects

Animal, particularly rat, research on prenatal development often emphasizes the
classical model of sexual differentiation and the role of androgens (especially
testosterone). As already noted, such research suggests that high levels of testosterone
early in gestation may lead to increased cell death, particularly within the right
hemisphere of the brain (Morris, Jordan, & Breedlove, 2004), as well as increasing neural
lateralization and slowing development of the brain (Goodman, 1991; Lyon & Gadisseux,
1991). The brain may be more vulnerable to insult while cell death is going on than after
it has consolidated. Therefore, increased cell proliferation and cell death, increased
lateralization, and slower development may make males more vulnerable to
environmental insult by prolonging the course of their neural development (Lyon &

Gadisseux, 1991; Morris, Jordan, & Breedlove, 2004).

17

In rats, a masculine pattern of neural deveIOpment has been linked to an extended
period of overproduction and subsequent pruning of DA receptors, a pattern that is
speculated to parallel the childhood rise and fall of hyperactive ADHD symptoms
(Andersen & Teicher, 2000). This idea is echoed in a theory of human disorder--
Geschwind’s (1987) theory of cerebral lateralization. He suggested that males are at
increased risk for learning disorder and hyperactivity because fetal testosterone slows
development in the left hemisphere of the brain, modifying cerebral lateralization
(McManus & Bryden, 1991). If males are more vulnerable to prenatal insult and to the
development of structural abnormalities in the left hemisphere due to their slower rate of
prenatal development, they may also be more likely to develop behavioral problems in
early childhood. Testosterone also appears to modulate neurotransmitter excitability
prenatally. When pregnant rats were exposed to restraint stress, their male pups had
reduced testosterone levels, as well as increased DA levels in the striatum (Gerardin et
aL,2005)

Overall, testosterone appears to exert prenatal, organizational effects through its
modulation of cell death in the brain. Additionally, there is some evidence that
testosterone may influence dopaminergic neurotransmission and excitability, especially
after exposure to prenatal stress.

Effects of Circulating Levels of Hormones
Circulating Testosterone and Gradient Eflects

Testosterone also appears to modulate neurotransmitter excitability postnatally.

After exposure to a novel environment, gonadectomized male rats treated with androgen

exhibited inhibition of DA turnover in the medial prefrontal cortex and decreased activity

18

in the open field (Handa, Hejna, & Lorens, 1997). Estrogen-treated male rats showed the
opposite pattern. Postnatally, testosterone plays a role in sexual motivation and behavior
by acting on the three dopaminergic systems: the nigrostriatal, mesolimbic, and medial
preoptic area systems (Hull, Muschamp, & Sato, 2004). According to a review of human
and animal studies (Dominquez & Hull, 2005), testosterone influences the release of DA
in at least one of these areas—the medial preoptic area. However, this is the area least
relevant to ADHD.

Circulating Estradiol and Active F eminization Eflects

In accordance with an active feminization model of sexual differentiation,
estradiol may also influence the development of the dopaminergic system. Estrogen
regulation of DA neurotransmission has been examined in rodent hypothalamus, striatum,
nucleus accumbens, and frontal cortex—all regions important in human ADHD—with
potential estradiol moderation effects (elucidated below) on both the nigrostriatal (motor
system) and mesolimbic (higher order cognition and motivation) dopaminergic systems
(Etgen, 2002). Further, levels of DA in the brain change over the course of the estrous
cycle and following estrogen injections in rats (Etgen, 2002).

Research reviewing the effects of hormones on the development of disorders
related to DA dysfunction like ADHD and Parkinson’s disease suggests that estradiol
may have a neuroprotective effect in midbrain dopaminergic neurons postnatally by
regulating gene transcription, acting as an antioxidant, or though crosstalk with
neurotrophic factors (Sawada & Shimohama, 2000). For example, treatment of
gonadectomized female mice with estradiol decreased DA evocation and depletion of

striatal DA in response to methamphetamine neurotoxicity (Dluzen & McDermott, 2002;

19

Myers, Anderson, & Dluzen, 2003). When ovarectomized monkeys were deprived of
estrogen for 30 days, there was permanent loss of over 30% of DA cells in the substantia
nigra (Leranth et al., 2000). Estrogen replacement restores these cells after ten days, but
not 30 days, of deprivation. This neuroprotection has also been seen specifically in the
nigrostriatal dopaminergic system, potentially by promoting the survival of some neurons
at the expense of others (Dluzen & Mickley, 2005; Dohanich, 2002). Estrogen can also
protect dopaminergic neurons following exposure to toxins (Etgen, 2002).

In a series of studies (Becker 1999; Becker & Rudick, 1999; Hu, Crombag,
Robinson, & Becker, 2004; Zhou, Cunningham, & Thomas, 2003), Becker and
colleagues found normal sex differences in dopaminergic function in the striatum and
nucleus accumbens in male and female rats that may be due to hormonal mechanisms.
Specifically, acute injections of estradiol early in development led to increased DA
release in the striatum and nucleus accumbens and an enhanced response to amphetamine
injection in the form of hyperactivity and stereotyped behaviors. Estrogen may also
influence performance on learning and memory tasks that are dependent on the striatum.
When ovariectomized rats were injected with estradiol, they became more sensitive to the
effects of a D2 DA receptor antagonist and made more errors on a learned, elevated plus
maze with a food reward (Daniel, Sulzer, & Hulst, 2005).

Several studies have shown that injections of DA into the nucleus accumbens of
ovariectomized/estrogen-primed rats (vs. ovariectomized rats) resulted in a reduction in
DA uptake and an increase in DA clearance time at the synapse (Pandaranandaka,
Poonyachoti, & Kalandakanond-Thongsong, 2006; Thompson, 1999). Several

mechanisms of estrogenic action on DA neurotransmission in the striatum and nucleus

20

accumbens have been suggested. Treatment of ovariectomized and estrogen-primed rats
with an agent of the D2 DA receptor suggests that steroid modulation of dopaminergic
transport in the striatum and nucleus accumbens is due to steroid-mediated action at the
DA autoreceptor/DA transporter coupling and independent action of estrogen on the DA
transporter (Thompson, Bridges, & Weirs, 2001). Estrogen administration in
ovariectomized rats selectively reduced the expression of the regulators of G-protein
signaling proteins (RGS) in the nucleus accumbens shell, suggesting that estrogen may
facilitate DA signaling by affecting RGS mRNA expression (Sharifi, Brady, & Koenig,
2004)
Summm

Overall, prenatal testosterone and circulating levels of testosterone and estradiol
may influence the development of ADHD and its corresponding mechanisms via their
influence on relevant DA circuitry. Prenatal levels of testosterone may exert
organizational effects on the DA circuits implicated in ADHD, for example, through
modulation of cell death in the brain. Testosterone may be a prenatal risk factor for
ADHD by making males more vulnerable to the development of abnormalities in DA
circuitry. As to the nature of these abnormalities, one possibility is decreased levels of
extracellular DA, resulting in less efficient DA modulation of responding. Additionally,
there is some evidence that testosterone may influence postnatal dopaminergic
neurotransmission and excitability, particularly in response to prenatal stress. Circulating
levels of estradiol may have a neuroprotective impact on the DA system during
childhood, in contrast to the effects of testosterone. Estradiol may also have protective,

postnatal effects through its modulation of DA neurotransmission, perhaps by preventing

21

over-transmission. Further, in correspondence with a gradient model of sexual
differentiation, relative levels or ratios between testosterone and estradiol may be related
to ADHD symptoms. Of course, it may be that not all of these effects are linear. For
example, hormones may interact with genetic vulnerability in order to influence ADHD.
However, such gene by hormone effects are beyond the scope of this dissertation.

IV. Hormonal Influences on ADHD Symptoms, Mechanisms, and Correlates

Organizational Effects of Hormones on ADHD Smptoms, Mechanisms, and Correlat_es_

 

Circulating levels of testosterone and estradiol are potentially able to alter
functioning and development in cognitive control and motivational/affective DA
circuitry, two of the key circuits implicated in ADHD. Recent research addresses the
influence of prenatal hormones on psychological processes relevant to this discussion,
such as ADHD symptoms and their related mechanisms of postnatal EF and
approach/affective systems. Other research has focused on other correlates of ADHD
such as sensation-seeking, attention, motor movement, and arousal. The literature
includes several human studies and suggestive animal studies.

In the six studies examining the relationship between prenatal hormones and
childhood ADHD symptoms in human clinical samples, finger-length ratios served as an
indirect measure of prenatal androgen exposure. These ratios show reliable sex
differences (Manning, 2002). They serve as a promising index of prenatal exposure to
androgens since finger-length ratios are thought to be diminished by exposure to high
levels of androgens early in development (Lutchmaya et al., 2004; Malas, Dogan, Evcil,
& Desdicioglu, 2006; Okten, Kalyoncu, & Taris, 2002). Girls with congenital adrenal

hyperplasia (CAH; exposed to high prental levels of adrenal androgens) show more

22

masculinized ratios (Brown, Hines, Fane, & Breedlove, 2002). Further, although finger-
length ratios show some change across development, the rank order, or serial stability, of
finger-length ratios remained largely the same over the course of childhood (McIntyre,
Cohn, & Ellison, 2006; Trivers, Manning, & Jacobson, 2006).2

In three prior studies with well-characterized clinical samples, children with
ADHD-PI were characterized by smaller (i.e., more masculinized) finger-length ratios
measured by photocopies or scanner (McFadden, Westhafer, Pasanen, Carlson, &
Tucker, 2005). Boys with autism/Asperger and ADHD/ODD had more masculinized
ratios, measured by calipers, than boys with anxiety disorders (De Bruin, Verheij,
Wiegman, & Ferdinand, 2006). A recent study by the author (Martel, Gobrogge,
Breedlove, & Nigg, in press) examined associations between finger-length ratios (right
index finger to ring finger ratio [right 2D:4D]) and ADHD in a larger sample of children
and adolescents with ADHD than utilized in the current study (which is more limited in
sample size due to its focus on executive and trait measures that were not available in as
many children). Right hand finger-length ratios showed significant mean differences by
sex, as well as associations with ADHD diagnosis. Boys with ADHD had more
masculinized finger-length ratios. More masculine right 2D:4D and 3D:4D ratios were
correlated with parent and teacher-rated inattentive and hyperactive-impulsive ADHD
symptoms in boys, but not in girls. Masculinized finger lengths appeared to be a general
risk factor for the development of inattentive, hyperactive-impulsive, and oppositional-

defiant symptoms with most risk for inattention.

 

2 One important caveat here is that there may be differences in finger length measurements based the
method of measurement. For example, photocopies yield lower finger-length ratios than direct finger
measurements (Manning, Fink, Neave, & Caswell, 2005). For this reason, in the current study, photocopies
were obtained on all children. A subset of children was also measured using calipers in order to examine
what differences exist due to measurement method.

23

In another study (Williams, Greenhalgh, & Manning, 2003), using a non-clinical
sample, high hyperactivity in childhood was related to lower right 2D:4D ratios,
measured by scanner and calipers-- indicative of high levels of prenatal androgen
exposure. In two samples of school-age children from the UK and Austria (Fink,
Manning, Williams, & Podmore-Nappin, 2007), more masculine right and left 2D:4D,
measured by photocopies and calipers, were associated with total difficulties and poor
conduct in boys and girls. Specific to the UK sample, more masculine 2D:4D was related
to more hyperactivity and more conduct problems, in boys but not in girls (Fink,
Manning, Williams, & Podmore-Nappin, 2007). In a college sample mostly comprised of
females (Stevenson etal., 2007), more masculinized 2D:4D on the left hand, measured by
calipers, was related to more inattentive and hyperactive-impulsive symptoms, but only
in the females.

Overall, in clinical samples, more masculine finger-length ratios were linked with
increased ADHD symptoms and, in particular, with increased inattentive ADHD
symptoms (De Bruin et al., 2006; Martel et al., in press; McFadden et al., 2005). In
population samples, more masculine finger-length ratios were related to increased
hyperactivity, although these relations were occasionally only found in girls or only
found in boys (Fink et al., 2007; Stevenson et al., 2007; Williams et al., 2003).

With regard to studies of related behavior problems, graduate and undergraduate
students with higher sensation-seeking evidenced more masculine finger-length ratios,
measured by photocopies and calipers (Austin, Manning, Mclnroy, & Mathews, 2002;
Fink, Neave, Laughton, & Manning, 2006). With regard to cognitive abilities, a review

concluded that moderate to high levels of prenatal and early postnatal levels of androgens

24

facilitated the development of spatial ability by adulthood (evident both in normal human
adults and in human adults who had lifelong hormonal abnormalities; Berenbaum,
Korrnan, & Leveroni, 1995).

I now turn to animal studies. King and colleagues (2000) suggested that early
androgen treatment decreased spatial memory in spontaneously hypertensive rats
(SHR)—a strain commonly used as an animal model of ADHD. Thus, androgen levels
appeared to facilitate spatial ability, but hinder spatial memory. Androgen levels also
decreased DA innervation in the same animals (King, Kelly, & Delville, 2000). Androgen
treatment during the early postnatal period in the male SHR was also associated with
increased path lengths in the open field, a behavior viewed as an animal analogue of
activity level (Li & Huang, 2006). One intriguing but unreplicated animal study is of
interest as well. In that study, both male and female SHRs were hyperactive and exhibited
discrimination problems, but their behavior differed markedly and was attributed to
different mechanisms (Berger & Sagvolden, 1998). Females were postulated to have a
deficit in attentional mechanisms, whereas males’ behavior was attributed to a shorter
delay-of-reinforcement gradient.

Effects of Circulati_ng Levels of Hormones on ADHD Symptoms. Mechanisms. and
Correlates

There is no research to date on the relationship of circulating hormones and
ADHD symptoms in childhood; the present study is the first. Addressing the influence of
estrogen levels on concurrent cognition and attention, a review of the organization of
estrogen receptors (ER) in the human brain suggests that ERB may play a role in general

cognition, memory, and motor speed, all functions that are at least subtly implicated in

25

ADHD (Ostlund, Keller, & Hurd, 2003). Research examining postmenopausal women
taking estrogen has found contradictory results such that higher levels of estradiol have
been concurrently associated with better oral reading and verbal working memory and
also with a higher risk of cognitive decline (Kang, Weuve, & Grodstein, 2004; Shaywitz
etaL,2003)

Janowsky and colleagues (1998) found that circulating hormone levels in normal
human adults were related to performance on motor speed tasks that showed sex
differences. In addition, higher levels of estradiol were related to better performance on
measures of spatial cognition in women, while higher levels of testosterone were related
to better performance on verbal recall for men (Janowsky et al., 1998). Jennings and
colleagues (1998) reported that levels of circulating estradiol were positively associated
with women’s performance on a sequential motor task, believed by those authors to
involve the basal ganglia. fMRI research further suggested that orbitofrontal reactivity to
emotional stimuli varied across the menstrual cycle in humans (Protopopescu et al.,
2005). Estrogen may weaken limbic arousal by increasing cortical control over the
amygdala (Goldstein et al., 2005), which would serve to dampen affective and more
general cognitive reactivity.

Thus, circulating estradiol appears to be related to some neuropsychological
functions, in particular to spatial ability and to motor movement, although it is unclear
whether these results would generalize to children. It also appears related to activity in
the limbic-fiontal circuit, at least in adult women, which suggests it may be relevant to
ADHD’s hyperactive-impulsive symptoms and reactive/affective processes.

Several animal studies have also been conducted examining circulating levels of

26

estrogen and its effect on relevant animal analogues. Shansky and colleagues (2004)
found that female rats appeared to have worse prefrontal cortical dysfunction (as
indicated by their worsened performance on a spatial working memory task, delayed
alternation) during times of their cycle that were characterized by higher levels of
estradiol. Thus, optimal, but not high, levels of estrogen may be linked to lower levels of
hyperactive-impulsive ADHD symptoms.

However, animal research also suggests that gonadal hormones can make stimuli
more or less salient, affecting perception (Maney, 2006) and potentially attention.
Research in primates indicates that levels of estradiol are positively related to visuospatial
attention in females (Shively & Bethea, 2004). Jentsch and Taylor (2003) found that
circulating hormone levels influence attention and impulsivity in normal and
gonadectomized male and female Sprague-Dawley rats (i.e., rats often used as controls in
comparison to the SHR animal model of ADHD). Intact male rats were more impulsive
(i.e., made more premature errors) than intact female rats. However, ovariectomies
worsened motor impulsiveness and orchidectomies worsened attentional performance.
Thus, estrogen and testosterone may be related to distinct aspects of ADHD, with optimal
levels of estradiol relevant to the hyperactive-impulsive aspects of ADHD and Optimal
levels of testosterone to the inattentive aspects. However, few other data like this are
available with which to evaluate that idea.

Overall, little research has examined EF and reactive/affective processes. Existing
studies shed some light on attention, general intellectual ability, motor movement, and
arousal. Direct evaluation of hormonal effects on relevant mechanisms or correlates

drawn from theories of ADHD is entirely lacking. Initial data on the relations among

27

hormones, cognition (i.e., reading, etc.), attention, motor movement, and arousal is
intriguing, yet far from definitive in regard to ADHD. As it currently stands, sex
hormones may be related to some forms of cognition, prefrontal arousal, impulsive
behaviors, inattention, neuropsychological performance, and motor movement.
Interesting yet preliminary sex differences have been noted in these domains in some
animal and human studies (e.g., King, Kelly, & Delville, 2000; Shively & Bethea, 2004).

The absence of research on the ADHD mechanisms of EF and reactive/affective
temperamental processes remains a huge gap. Irrespective of the general scarcity and
limitations of the current research on hormonal influence on human development, several
hypotheses might be generated regarding potential hormonal influence on the
development of neural circuitry and processes implicated in ADHD. I turn to these next.
Summary and Hypothesized Models of Hormonal Influence on ADHD

Overall, the research reviewed herein on humans and animals suggests that
hormones may have organizational effects on two key DA neural circuits associated with
ADHD: the frontal-striatal circuit (via the striatum) and the frontal-limbic circuit (via the
nucleus accumbens). Yet, links between hormones and ADHD, as well as with the
cognitive and behavioral profile associated with ADHD, have hardly begun to be
examined in human children. If hormones influence the neural structures implicated in
ADHD, then association should be able to be detected via alterations in
neuropsychological and/or temperamental measures that are associated with ADHD and
are also correlated with integrity of these same neural circuits.

I next explain in more detail two main routes: a prenatal route involving

testosterone and a childhood route involving circulating levels of testosterone or

28

estradiol.

1. Prenatal route. First, I argue that prenatal testosterone may influence the
relationship between the development of the striatum and DA fimction in those regions,
as evidenced by EF. The striatum is potentially central to the development of inattentive
symptoms of ADHD. High levels of testosterone may influence the pattern of
development of striatal DA receptors, leading to overproduction of these receptors in
childhood (Andersen & Teicher, 2000). An overproduction of DA receptors may lead to
increased uptake of neurotransmitters from the synaptic cleft and lower levels of
extracellular DA. High levels of testosterone may also decrease the production and
availability of DA within the frontal-striatum and/or modulate circulating levels of DA
within the brain (Gerardin et al., 2005) through neurosteriod by DA interactions.3

Alterations in dopaminergic function and frontal-striatal neural circuitry appear to
underpin top down control by decreasing the efficiency of arousal and other regulatory
processes needed to sustain attention. In turn, breakdowns in such control are theorized to
lead to inattentive ADHD symptoms. In fact, previous work indicates that children with
ADHD are characterized by alterations in or decreased levels of dopaminergic activity
(e.g., Volkow et al., 2005). In addition, abnormalities in dopaminergic structure and
function in the fiontal-striatal region of the brain has been implicated in the deficits in top
down control exhibited by children with ADHD (e. g., Nigg & Casey, 2005).

High levels of prenatal testosterone may also be more subtly related to reactive

 

3 This description is not exhaustive. Prenatal levels of testosterone may have additional effects on
neurobiology that are not examined in the current study. For example, testosterone may moderate the
relationship between genes and the development of neurotransmission/neurobiology by interacting with
genetic risk for the disorder. It may also moderate the relationship between prenatal risk factors (e.g.,
exposure to environmental contaminants such as PCBs and DDT) and dopaminergic neurotransmission
(Miller, Gainetdinov, & Caron, 2000; Nigg, 2006). Those are potential topics for subsequent studies.

29

control via alterations in mesolimbic circuitry, particularly via influences on the nucleus
accumbens. Since the alterations in the nucleus accumbens and the limbic system are
implicated in reactive processes (Whittle, Allen, Lubman, & Yucal, 2006), hormonal
influences on this neural circuitry may impact both reactive processes and corresponding
hyperactive-impulsive ADHD symptoms.

One key hypothesis therefore was that high levels of prenatal testosterone,
measured via finger-length ratios, are related to more impaired top down control,
evidenced by weaker EF and lower effortful control/conscientiousness, and more
inattentive ADHD symptoms in boys. If so, EF and effortful traits should mediate the
relationship between finger-length ratios and inattentive ADHD symptoms in childhood,

as shown in Figure 2.

 

 

Striatum Conscientious-
ness/ EF

 

 

 

 

Prenatal W Inattentive
Testosterone J Symptoms

 

 

 

Figure 2. Mediator Model for Prenatal Testosterone Effects on ADHD

2. Childhood effect. Second, circulating testosterone or estradiol levels may

30

influence motor symptoms and hyperactivity during childhood. Circulating estradiol may
modulate DA activity in both the striatum and nucleus accumbens (i.e., both of the
ADHD-relevant pathways). One way it can do so is by providing neuroprotection against
oxidative stress in the nigrostriatal dopaminergic system (Becker, 1999; Dluzen &
Mickley, 2005; Sawada & Shimohama, 2000). Thus, circulating levels of estradiol may
be protective against ADHD in childhood by mitigating neural responses to illness and
injury.

Secondly, recent research on animals and humans suggests that sensory
perception, cognition, arousal, attention, and affective processes can be influenced by
circulating sex hormones through their effect on the amygdala (Goldstein et al., 2005;
Maney, 2006; Protopopescu et al., 2005; Shively & Bethea, 2004), a part of the reactive
control circuitry. Thus, circulating levels of estradiol may be positively related to reactive
control and negatively related to ADHD symptoms of hyperactivity/impulsivity,
providing some protection against the development of ADHD. As shown in Figure 3,
circulating levels of estradiol may mediate the relationship between reactive
temperamental control and current hyperactive-impulsive ADHD symptoms in girls. The
relationship between circulating levels of testosterone and childhood ADHD symptoms is

less clear and is examined in this study in exploratory fashion.

31

 

 

Nucleus

 

 

 

 

 

Circulating Accumbens/

Estradiol Amygdala
Reactive W Hyperactive-
Control impulsive
J Symptoms

 

 

Figure 3. Mediator Model for Circulating Estradiol Effects on ADHD

As a third route, circulating gonadal hormones may influence ADHD symptoms
through a balance between circulating levels of testosterone and estradiol or an optimal
level of gonadal hormones. That is, a higher ratio of testosterone to estradiol may be
more characteristic of ADHD and moderate, but not high, levels of estradiol may
facilitate adaptive affective regulation and fewer ADHD symptoms.

Conclusion

Sex hormones apparently influence the development of some of the same neural
circuitry that is implicated in ADHD. In particular, testosterone and estradiol appear to
influence the pre- and postnatal development of the prefrontal cortex, striatum, and
nucleus accumbens, as well as influencing dopaminergic neurotransmission in these
regions. Prenatal testosterone seems to reduce DA transmission and so would likely result
in less efficient operation of effortful control systems. Circulating estradiol appears to

protect the integrity of dopaminergic function and so would lead to more efficient

32

reactive control operations. Consistent with this picture, the following was hypothesized.
(1) Prenatal testosterone is related to inattentive ADHD symptoms (in boys), mediated by
impaired EF and weak effortful control/conscientiousness. (2) Circulating levels of
estradiol mediate the relationship between reactive temperamental traits and current
hyperactive-impulsive ADHD symptoms (in girls). Alternatively, in both boys and girls,
an optimal balance of circulating testosterone to estradiol or an optimal level of estradiol
may lower risk. If so, it would emerge that a larger testosterone to estradiol ratio predicts
more inattentive ADHD symptoms and more impaired executive and effortful
control/conscientiousness. At the same time, if this is true, then estradiol will exhibit a
curvilinear relationship with hyperactive-impulsive ADHD symptoms and reactive

control.

33

METHOD

Participants

The current study reports on 312 participants who were drawn from an adolescent
sample (ages 14-17) and a new child sample (ages 8-13). The only difference in project
protocol entails slight variation in the neuropsychological battery administered. The 312
children came from 274 families. Thus, 38 siblings are included in these analyses. 22
sibling pairs were disconcordant for ADHD (i.e., one was diagnosed with ADHD and the
other was a non-ADHD control). Five sibling pairs were both diagnosed with ADHD,
and 8 siblings pairs were non-ADHD controls.

Through the use of a multistage screening process, these participants were
recruited into two groups, ADHD and control, with 168 children with ADHD (89
combined subtype, 78 inattentive subtype, and 1 hyperactive-impulsive subtype) and 144
controls. Within the whole group, there were 178 boys and 134 girls, with 106 boys and
62 girls in the ADHD group and 72 boys and 72 girls in the control group. Most
adolescents and children had ADHD, cognitive, and temperament measures available.
However, not all adolescents and children had circulating hormone measures available, as
explained later.

Demographic statistics on the sample, broken into ADHD, control, and total
groups, can be seen in Table 1. There were significantly more boys in the ADHD group
than in the control group, and children in the ADHD group were significantly younger
than children in the control group (p<.05). In addition, and as expected, children with
ADHD had more ADHD symptoms, higher Conners’ t-scores, and more ODD and CD

symptoms than children in the non-ADHD control group (p<.01).

34

Table 1

Sample Demographics

 

 

ADHD Controls Total

n=168 n=144 N=312
Males 106 (63.1%) 72 (50%) 178 (57.1%)*
Ethnic Minority 40 (23.81%) 38 (26.39%) 78 (25%)
Age 12.71(3.01) 14.01(2.76) 13.31(2.96)**
Family Income 71 I73.9l(83045.01) 77306.52(5553 1 .69) 74051 .52(71400.45)
Full Scale IQ 103.09(12.66) 111.55(14.34) 106.98(14.08)
Inattention Symptoms' 8.54(.72) 1.860.91) 5.42(3.62)**
Hyperactive Symptomsl 6.19(2.90) 1.17(1.44) 3.85(3.43)**
ODD Symptoms 2.41(2.48) .51(1.21) 1.59(2.23)**
CD Symptoms .35(.73) .05(.22) .23(.58)**

Conners’ Cognitive Problems (P)
Conners’ Hyperactivity (P)
Connors’ ADHD T-score (P)
Conners’ Cognitive Problems (T)
Conners’ Hyperactivity (T)

Connors’ ADHD T-score (T)

71.08(10.64)
67.42(15.80)
72.27(9.33)

60.15(11.47)
61.84(l4.80)

66.43(12.78)

47.66(6.97) 60.37(14.83)**
49.17057) 58.93(15.58)**
48.11(7.05) 61.18(l4.67)**
48.63(7.39) 54.58(11.28)**
50.60(10.53) 56.37(14.05)**

50.03(9.58) 58.42(13.98)**

 

Note. *p<.05. **p<.01. (P)=parent-report. (T)=teacher-report. 1= parent+teacher OR
algorithm ADHD symptoms based on K-SADS. Significant group differences tested by t-

tests or chi-square statistics.

35

Multistage recruitment and diagnostic assignment process

All participants were recruited from mid-Michigan schools, clinics, and
community through public advertisements and announcements, mass mailings, and
outreach to schools and clinics in order to recruit as broad and representative of a sample
as possible. The first stage was a screen that precluded any children who took slow-acting
psychotropic medication (i.e., prednisone, antidepressant, anti-psychotic, anticonvulsant
medications), had major medical or neurological conditions, including genetic, gonadal,
or hormonal abnormalities as reported by the parent, and who did not speak English as
their first language.

At the second stage, parents and teachers completed normative behavior rating
scales: School and Home versions of the ADHD Rating Scales (DuPaul, Power,
Anastopolous, & Reid, 1998), the Parent and Teacher Behavior Assessment System for
Children (BASC; Reynolds & Kamphaus, 1992), and the Conners (1997) Parent and
Teacher Rating Scales-Revised-Short Forms. These scales were chosen based on their
empirically-validated cut-offs, thorough coverage of symptoms, and adequate reliability
and validity. Further, these scales are among the most widely-accepted rating scales
currently used and were judged to have the best out score data for screening (Ostrander,
Weinfurt, Yarnold, & August, 1998; Power et al., 1998). Test-retest reliability for these
scales range from .62 to .95. Internal reliability in this study ranged from .84 to .96, and
validity coefficients range from .78 to .98 (Conners, 1997; DuPaul et al., 1998; Reynolds
& Kamphaus, 1992).

For the third stage, participants completed an individual, semi-structured, clinical

interview (i.e., KSADS-E; Puig-Antich and Ryan, 1986). The KSADS-E was chosen

36

because its structure is clinically sensitive, and it provides a thorough coverage of
childhood disorders (Biederman et al., 1990; Biederman et al., 1992). For the child
sample, girls were asked if they had started menstruation and, in the case of the
affirmative, given a premenstrual syndrome (PMS) and premenstrual dysphoric disorder
(PMDD) symptom checklist (i.e., Daily Record of Severity of Problems; Yonkers et al.,
1997) to complete as part of their questionnaire packets. This information was not
available on the archival adolescent sample, however.

The data from the interviews and parent and teacher rating scales were then
presented to a clinical diagnostic team consisting of a social worker, board certified child
psychiatrist, and licensed clinical child psychologist. A “best estimate” diagnostic process
was implemented, in which the psychiatrist and psychologist independently arrived at a
clinical decision regarding ADHD diagnosis, subtype, and presence of comorbid
disorders. Their agreement rates were acceptable for ADHD, its subtypes, and current
Oppositional Defiant Disorder and Conduct Disorder (all kappas _>_ .89). In the event of a
disagreement, they conferred and easily arrived at consensus; if they could not, the case
was omitted.

C omorbid psychiatric and learning problems

For both samples, comorbid symptoms and diagnoses (i.e., anxiety disorders,
mood disorders, Post-Traumatic Stress Disorder, Obsessive-Compulsive Disorder, Tic
disorders, ODD, and CD) were assessed via modules of the relevant parent semi-
structured or structured interview and behavioral rating scales. IQ and learning disability
were assessed with a reliable and valid three-subtest form of the Wechsler Intelligence

Scale for Children (WISC; Wechsler, 2003a) and the Wechsler Individual Achievement

37

Test (WIAT; PsychCorp, 2005) Basic Reading and Mathematical Reasoning subtests.
Exclusionary Criteria

Autistic Disorder, Mental Retardation, Schizophrenia, Bipolar Disorder, F etal
Alcohol syndrome, known neurological disorder, Cerebral Palsy, Tourette’s disorder,
genetic, gonadal, and hormonal abnormalities, PMS (when data available), and PMDD
(when data available) were exclusionary criteria.

Medication rule out and wash out

Children and adolescents with ADHD attended neurocognitive assessment visits
after an appropriate washout fi'om psychostimulant and other ADHD medication (for
short-acting stimulants, 24 hours; for slow release formulas, 48 hours). Individuals taking
long-lasting psychotrophic medications were necessarily excluded when possible for
purposes of the neuropsychological testing and blood draw. Although taking of hormones
such as prednisone or birth control pills was not formally screened out in the initial years
of data collection, information on medications was obtained. In the current study, 115
participants had a prescription for stimulant medication, and four participants had a
prescription for birth control pills.

Measures
Executive Function (EF)

The neuropsychological EF measures examined components of EF derived from
the model suggested by Pennington (1996). These tests were administered in a fixed
order so as to minimize children’s fatigue.

Response Inhibition

The tracking version of the Stop Task is a well-validated computer task (Band,

38

van der Molen, & Logan, 2003; Logan, 1994) during which the participant is instructed
to press the “X” key when s/he sees the letter X on the computer screen and the “O” key
when s/he sees the letter O on the screen. When the child hears a tone before the letter
flashes, s/he tries to refrain from responding. The Stop Task, with alpha reliability higher
than .8, measures the ability to suppress a motor response on cue. A measure of
inhibitory control, stop signal reaction time, was computed by subtracting mean stop
signal latency from mean go response time.
Response Variability

A response variability measure was obtained from the Stop Task, defined as the
within-child variability of the reaction time on the “go” trials. This measure served as an
index of response variability, conceptualized as reflecting degree of interference between
. “hot” and “cool” EF processes, energetic deficits, and temporal instability of responding
(Russell et al., 2006).
Set-Shifting

The Trailmaking Task, a subtest of the Halstead-Reitan Neuropsychological
Battery, requires the participant to trace a path between numbers and letters as quickly as
possible without errors. Widely used in neuropsychological research measuring frontal
lobe function, the outcome measure of the task was set-shifting, measured as the time
taken to complete the more difficult Form B minus the time to complete Form A (Reitan
& Wolfson, 1985; Spreen & Strauss, 1991).
[Q and Academic Achievement

Child I_Q was assessed with the widely-used and extensively-researched Wechsler

Intelligence Scale for Children-Fourth Edition (WISC-IV) or the Wechsler Intelligence

39

Scale for Children-Third Edition (WISC—III). A reliable and valid 3-subtest short form
was used to estimate the Full Scale Intelligence Quotient (Wechsler, 2003a; 2003b).
While the WISC-III was used for the adolescent sample, the WISC-IV was used for the
new sample. In instances where this variable is used as a covariate in the analyses,
standardized scores were used.

Temperament Trafi

The current study examined two key traits: conscientiousness and reactive
control. They were hypothesized to be relevant to the development of ADHD via distinct
pathways (see Figure 1). To measure these traits, the mother (or primary caregiver)
completed the common language version of the California Child Q-Sort (CCQ; Caspi et
aL,l992)

The QC_Qis a typical Q-Sort consisting of 100 cards which must be placed in a
forced-choice, 9-category, rectangular distribution. The rater describes the child by
placing descriptive cards in one of the categories, ranging from 1 (least descriptive) to 9
(most descriptive). Items in the middle are thus uninformative, and those on the two ends
are most informative. The items for the reactive control scale examined in the current
dissertation were selected to follow the suggestions of Eisenberg and colleagues (1996),
consisting of 14 items with internal reliability of .74. To measure conscientiousness, a
scale developed by John and colleagues (1994) was used, consisting of nine items with
internal reliability of .87.

Sex Hormones
Circulating Levels of Estradiol and Testosterone

Venipuncture blood draws were performed at the MSU Clinical Center at 5:00 in

40

the evening, for half of the adolescent sample (i.e., those obtained in the latter half of that
data collection period after funding for blood draws had been obtained; thus, this was
expected to be a random sample from that group) and the entirety of the child sample.
Immediately after the blood draw, the child had another rest period along with snack and
drink, and then-- at 5:30—-- neuropsychological testing in the laboratory began. The blood
draws were preceded by informed parent consent and child assent. For girls who had
begun menstruation, data regarding the first date of their menstrual cycle, the end date of
their last two menstrual cycles, and a history of missed periods was collected by a female
staff member in private after the establishment of the confidentiality policy and rapport.
These data were checked against the relative levels of hormones in their blood in order to
validate their self-report to the extent possible, although such validation was limited in
girls of low gynecological age (i.e., with low hormone levels).

Samples were stored in a -20 freezer in the Psychology Building, although they
were shortly moved to a -80 freezer, and storage duration was minimized. Estradiol was
analyzed in Dr. Fisher’s laboratory using standard sermn/plasma radioimmunoassay kits.
Because the manufacturer failed to supply enough kits, estradiol data were missing for 18
cases. These cases were considered missing at random.

The clinical laboratory at Michigan State University is certified by both the
College of American Pathology and Clinical Lab Improvement Act. Duplicate analyses
of the same samples were run in the laboratory using separate kits in order to assure the
reliability of the results. The coefficient of variation used in the laboratory within runs
was less than ten percent variation, meaning that duplicate samples that differed by more

than ten percent were deemed unacceptable. Reliability between runs was also checked

41

by obtaining the average mean and standard deviation of each run for between six to
eight runs. Then, values from other runs were rejected if they fell more than one or two
standard deviations outside of these average values. In addition, the sensitivity of the
estradiol assay in measuring low levels of estradiol in prepubertal girls was a critical
issue that was monitored in an ongoing fashion. Less than 5% of samples fell outside the
reliably-detectable range.

All samples were assayed for salivary testosterone in duplicate using a highly-
sensitive enzyme immunoassay (Cat. No. 1-2402/1-2412, Salimetrics LLC, State College
PA). Although Salimetrics’ EIA kits were designed for use with salivary samples, they
were successfully validated on low levels of serum testosterone before use with the
current sample using test samples with known concentrations of testosterone. 25 uL of a
1:10 serum dilution was used per test. The test used 25 ul of saliva per determination, has
a lower limit of sensitivity of 1.0 pg/mL, standard curve range from 6.1 pg/mL to 600
pg/mL, an average intra-assay coefficient of variation of 4.6%, and an average inter-assay
coefficient of variation of 8.25%. Method accuracy determined by spike recovery
averaged 104.4%, and linearity determined by serial dilution averaged 99.9%. The
correlations between serum and saliva was highly significant with r (14) = 0.91, p<0.001
for males and r (12) = 0.61, p<0.001 for females. Testosterone, originally measured in
nanograrns per milliliter, was converted to picograms per milliliter so that it would be
comparable to estradiol measurements when discussing or graphing absolute levels.
Then, a ratio was computed between testosterone and estradiol (i.e.,
testosterone/estradiol).

Self-ratings of pubertal morphological changes for the children and adolescents

42

was assessed with the Pubertal Development Scale (Angold et al., 1999; Petersen,
Richards, & Boxer, 1988), using a self-report rating scale completed by the child and by
the parent about the child. A categorical and continuous pubertal status variable was
generated from this rating scale. Although this report obviously is not as valid as a
physical exam, the correlation between self-report and physician rating of pubertal
development is .70 (Petersen et al., 1988), suggesting that this approach had some
validity for the present study.

One hundred eight-one children in the current study had Pubertal Development
Scale data. In these 181 participants, maternal and child report on the Pubertal
Development Scale correlated at .80 (p<.01) for categorical stage and .81 (p<.01) for
continuous pubertal stage. Thus, these indices were averaged to form a composite
categorical and continuous measure of pubertal stage. Of the 60 girls who had circulating
estradiol data available, 28 had not started menstruating and 32 had started menstruating.
Fifteen girls reported the dates of their previous two cycles and had hormone data
available. The combination of both forms of data suggested that nine girls were in the
follicular phase, 1 was in the ovulatory phase, and 5 were in the luteal phase. The
remaining girls who had started menstruation either did not report their cycle
information, their phase was unable to be determined because they had irregular cycles,
or their reported cycle information did not agree with their hormone levels, as compared
to tables with normative levels per menstrual cycle phase for their age.
Prenatal Levels of Testosterone via Finger length Ratios

Finger length ratio was used to index prenatal testosterone (for discussion of

validity of this procedure, see McFadden et al., 2005). This approach is a relatively

43

established procedure for measuring prenatal androgen exposure; however, it remains
debated (Cohen-Bendahan, van de Beek, & Berenbaum, 2005). Child finger-length
images were obtained using a photocopier, 4 and measurements were taken using a ruler.
For the photocopier measurements, children were asked to remove any jewelry and place
their two hands side-by—side, fingers closed, onto a clear template marked like a ruler. A
one-pound bag of rice was then placed over their hands to provide a contrast, and their
hands were photocopied. The examiner then asked the parent for information regarding
the number of older and younger siblings of the child, the birth order of the child, and the
presence of any miscarriages (and gender if known) before the birth of the child. Two
raters then independently recorded measurements of the finger-length of index, middle,
ring, and little fingers. The measurements between raters were reliable (n=20; all r above
.91; all p<.01) with an average correlation between the two raters’ measurement of
finger-lengths of .97. Ratios were then computed between each of the pairs of fingers on
both hands. The ratios exhibiting large sex differences and most often used in prior
research were the right 2D:4D and the right 2D:5D (Fink et al., 2007; Martel et al., in
press; Stevenson et al., 2007). It should be noted that finger-length ratios show ethnic, as

well as sex, differences (Cohen-Bendahan et al., 2005).

Data Analysis

 

4 Photocopies of finger length may introduce some distortion into the calculation of finger-length ratios
(Manning, Fink, Neave, & Caswell, 2005). Such effects were not a major focus because they presumably
would not be able to explain ADHD-specific effects. However, as a precaution, calipers were also used in
addition to photocopies and rulers for a subset of children to enable comparison between these two
different measures of finger length. For the subset of the sample (n=57) who had photocopied image and
caliper measurements available, correlations between raw finger lengths were large and significant (average
r=.84, p<.01). However, correlations between finger-length ratios, computed from photocopied images and
caliper measurements, were relatively low and marginally significant (average r=.26, average p=.13)
consistent with the literature on lower reliability of ratio scores than raw scores (Manning et al., 2005). For
main study analyses, photocopies of finger-length ratios were analyzed, and this potential unreliability is
noted as a limitation in the discussion.

44

Initial descriptive data analysis, such as t-tests, correlations, and MANOVAs,
were handled using SPSS. Subsequent data analyses, such as regressions and mediation
analyses, were handled using Mplus.

Handling of Outliers and Skewed, Nonnormal Data

For variables with outliers, trimmed means were used (Wilcox, 2002). That is,
outliers that were greater than 3 standard scores from the mean were trimmed to be
within three standard scores of the mean. An exception to this rule was made for clinical
(i.e., hormonal) data. In this case, severe outliers were deleted since they appeared to be
inaccurate measurements, and the deleted data points were handled with the F IML
procedure described below. This was true for less than 5% of data points. For variables
that were significantly skewed or nonnorrnal (as indicated by the Kolmogorov-Smimov
and Shapiro-Wilk tests), data transformations were attempted. However, data
transformations were not successful in correcting the nonnormality of the data. Thus,
subsequent univan'ate and multivariate regressions were conducted using the Mplus
software package (Muthen & Muthen, 2007) which allows for the statistical control of
non-normality and outliers through the use of robust maximum likelihood estimation
(Curran, West, & Finch, 1996).

Handling of Missing Data

The current sample was comprised of all adolescents and children who had
diagnostic, trait, EF, and finger length data available (N=312). Of these, approximately,
200 participants had circulating hormone data available. Missingness for the hormone
data (and for all other variables) was handled using Mplus since MPlus provides full

information maximum likelihood estimation under MCAR (missing completely at

45

random) and MAR (missing at random; Muthen & Muthen, 2007). Full information
likelihood estimation (i.e., F IML or direct fitting) is a method of directly fitting models to
raw data without imputing values (McCartney, Burchinal, & Bub, 2006). A state-of-the-
art missing data technique, FIML is seen as having all the strengths of single and multiple
data imputation, while also being easy to interpret since only one parameter (rather than
multiple parameters) is produced by the analysis. In addition, FIML can be used when the
amount of missing data is moderate or large (i.e., up to thirty percent missingness;
McCartney, Burchinal, & Bub, 2006).
Handling of Non-independence of Data

The presence of siblings and the resulting non-independence of data points were
handled using the clustering feature of MPlus. This clustering feature takes into account
the non-independence of the data when computing test statistics and significance tests.
Choice of Covariates

Covariates varied for each type of analysis, but the choice of covariates followed
pre-set rules. For any analyses utilizing all children (i.e., both boys and girls), sex was
covaried in order to control for the greater number of boys in the ADHD group, as
compared to the control group. For analyses examining finger-length ratio, ethnicity was
also covaried due to the fact that finger-length ratios show ethnic differences. For
analyses examining circulating hormone levels, age was covaried in order to control for
age differences between the ADHD and control groups.
Correction for Type 1 Error

Due to the number of statistical tests conducted in the current study, a correction

for Type I error was implemented. A Bonferroni correction for all tests was viewed as

46

excessively stringent as it would result in unacceptably low power, particularly due to the
fact that several tests were run as “control tests” (i.e., not expected to result in positive
findings). Therefore, results in the current study were considered significant if they had a
p-value at or below .01. Results with a p-value at or below .05 were considered

marginally significant or at trend-level.

47

RESULTS

Descriptive Statistics and Preliminam Data Checks

Descriptive statistics. Descriptive statistics on the sample, broken into ADHD,
control, and total groups, are shown in Table 2. As shown in Table 2, the right 2D:4D and
3D:4D were marginally more masculine (i.e., smaller) in the ADHD group as compared
to the control group (p<.05). In addition, children with ADHD, as compared to controls,
exhibited significantly more impaired response inhibition (p<.01), increased response
variability (p<.01), less conscientiousness (p<.01), and less reactive control (p<.01).
Children with ADHD also had marginally decreased set-shifting ability (p<.05). These

results were as expected.

48

Table 2

Descriptive Statistics by Diagnostic Group

 

 

ADHD Control Total

n=168 n=144 N=312
R 2D:4D .94(.06) .96(07) .95(.06)*
(z-score) -.14(.92) . 1 8(1 .07) 0(1)
R 2D:5D l.50(.19) 1.54(.21) 1.52(.20)
(trimmed z-score) -.08(.94) .09(1) -.01(.97)
R 3D:4D l.12(.04) 1.14(.05) 1.13(.05)
(trimmed z-score) -.15(.91) .l9(1.02) -.00(.97)*
Pubertal Stage (cat.)' 2.84(l.08) 3.310.45) 3.030.13)*
Pubertal Stage (cont)l 10.27(4.45) 12.06(4.50) 10.98(4.54)*
Testosterone (ng/mL) .95(1.13) l.15(1.11) l.02(1.12)
Estradiol (pg/mL) 35.43(39.35) 31 .66(26.99) 34.00(35. 13)
Response Inhibition 284.10(125.00) 233.72(78.50) 259.64(107.81)**
(trimmed z-score) .20(1.08) -.24(.73) -.01(.95)
Variability 153.22(47.14) 130.21(39.62) 142.10(45.07)**
(trimmed z-score) .21(.91) -.25(.85) -.02(.91)
Set-shifting 55.90(24.80) 50.60(26.8l) 53.45(25.84)*
(trimmed z-score) .08(.90) -.15(.82) -.02(.87)
Conscientiousness 4.05(1.25) 6.40(1 .19) 5. 13(1 .69)* *
Reactive Control 4.29( 1.12) 5.25(.99) 4.73(l.l7)**

 

Note. *p<.05. **p<.01. Significant group differences tested by t-tests. ' = categorical and
continuous measures of pubertal stage are assessed via an average of child and maternal ratings
on the Pubertal Development Scale.

49

Preliminary Data Check 1: Subggpe differences in finger-length ratios,

 

neuropsychological EFJI’ temperament traits in children with ADHD and controls

A MANCOVA, covarying sex and ethnicity, was conducted to examine subtype
differences in three finger-length ratios (i.e., right 2D:4D, 2D:5D, and 3D:4D). There
were no significant differences between subtypes in finger-length ratios (F[3, 119]=1.71,
772:.04, p>.05). A MANCOVA, covarying sex, examined subtype differences in cognitive
control (i.e., response inhibition, variability, and set-shifting) and was also nonsignificant
(F[3, 119]=.95, n’=.02, p>.05). A MANCOVA, covarying sex, was conducted to examine
subtype differences in traits. The combined subtype had lower conscientiousness and
reactive control than did the inattentive subtype (F[2,163]=14. 17, n’=.15, p<.01). Since
no subtype differences were noted in hormone measures and in order to maximize
statistical power, the two subtypes were pooled in subsequent analyses.
PreliminarvfiData Checlgl: Sex and diagnostic differences in finger-length ratios,
neuropsychological EF. or temperament traits in children with ADHD and controls

Descriptive statistics, broken into four groups by sex and diagnostic group, are
shown in Table 3. Three MANCOVAs were conducted to examine sex and diagnostic

differences in 1) finger finger-length ratios, 2) EF, and 3) temperament traits.

50

Table 3

Descriptive Statistics by Sex within Diagnostic Group

 

ADHD(n=168)

Control (n=144)

 

Boys Girls Boys Girls

n=106 n=62 n=72 n=72
R 2D:4D .93(.06) .97(.05)** .95(.07) .98(.06)+
(z-scores) -.38(.89) .26(.84) -.00(1.09) .3 8(1.02)
R 2D:5D 1.46(. 18) 1.57(.17) l.51(.20) 1.57(.22)
(trimmed z-scores) -.30(.92) .27(.87)** -.06(.94) .26(1.04)
R 3D:4D 1.11(.05) 1.14(.03) 1.14(.05) 1.14(.04)
(trimmed z-scores) -.38(.94) .21(.73)** .10(1.10) .29(.93)
Pubertal Stage (cat.)l 2.72(.96) 3.05(1.23) 3.23(.99) 3.43(1.32)

Pubertal Stage (cont.)' 8.83(3.27)

Testosterone (ng/mL) l .27(1.34)

Estradiol (pg/mL)

207702.29)

Response Inhibition 293.20(137.87)

(trimmed z-scores)

Variability
(trimmed z-scores)

Set-shifting
(trimmed z-scores)

Conscientiousness

Reactive Control

270.17)

154.78(46.14)
.24(.86)

58.14(26.61)
.16(.94)

4.020.13)

4,130.09)

12.83(5.11)** 10570.52)
.43(.17)** 1.66(l.31)

59.29(54.3)** 23.04(8.65)

13.85(4.93)**
.63(.48)**

43.l6(37.42)*

266.34(94.01) 243.90(78.44) 222.63(77.74)

-.34(.72)

-.l7(.83)

54.32(32.00) 46.88(l9.89)

-.25(.77)

6.68(1.04)**

.06(.87) -.15(.73)
150140947) 127270933) 133400004)
150.00) -.33(.87)

52.06(21.00)

-.05(.81) -.O4(.86)

4.110.44) 6.120.26)

4.550.14)* 5.290.06)

5.22(.93)

 

Note. +p<.10. *p<.05. "p<.01. Significant group differences for boys vs. girls within diagnosis
were tested by t-tests. ' = categorical and continuous measures of pubertal stage are assessed via
an average of child and maternal ratings on the Pubertal Development Scale.

51

The first MANCOVA examined sex and diagnostic differences, controlling for
ethnicity, in right 2D:4D, 2D:5D, and 3D:4D entered together in the MANCOVA. There
was a significant multivariate main effect for sex (F[3, 214]=4.99, A=.94, 772:.07, p<.01)
in the expected direction (i.e., boys had more masculine finger-length ratios), but not for
diagnostic status (F[3, 214]=1.80, A=.98, n’=.03, p>.05). The interaction between sex and
diagnostic status was not significant (F[3, 214]=.74, n‘=.01,p>.05).

AN COVA was conducted to examine sex and diagnostic differences in right
2D:4D (i.e., the most widely validated finger-length ratio), controlling for ethnicity. Sex
and ADHD main effects were significant in the expected direction: boys and children
with ADHD had more masculine finger-length ratios (F=13.96, n2=.06, p<.01 for sex;
F=4.13, 772:.02, p<.05 for ADHD). The interaction between sex and diagnostic status was
not significant (p=.34). The ADHD effect held in a single-factor ANCOVA with sex and
ethnicity covaried (F=4.79, 772:.02, p<.05).

The second MANOVA examined sex and diagnostic differences in EF. There was
a significant multivariate main effect of ADI-[D diagnostic status (F[3, 235]=6.26, A=.93,
772:.07, p<.01), but no significant sex differences (F[3, 235]=2.08, A=.97, n’=.03, p>.05)
or interactions (F[3, 235]=.32, A=1.00, n’=.004, p>.05). Diagnostic differences in EF
have been well-validated (e.g., Ni gg et al., 2002). Sex differences in EF are more
controversial. However, the current study’s results are in line with some work on sex
differences in EF which have failed to find sex effects (Seidman et al., 2005).

The third MAN OVA examined sex and diagnostic differences in the two
personality traits, conscientiousness and reactive control, that had been selected as

behavioral probes of top down and bottom up control respectively. The interaction

52

between sex and ADHD diagnosis was significant (F[2, 307]=4.76, A=.97, 172:.03,
p<.01). Girls had higher levels of conscientiousness than boys in the non-ADHD group.
Girls had higher levels of reactive control than boys within the ADI-ID group (Table 3).

Bivariate correlations were conducted between hormone measures and symptoms,
cognitive control, and traits in the full sample. These correlations, conducted for
descriptive purposes, are shown in Table 4. As can be seen, right 2D:4D was significantly
related to inattention, hyperactivity-impulsivity, and conscientiousness, in the expected
direction (i.e., more masculine finger-length ratios related to more symptoms and less
conscientiousness). In addition, circulating estradiol was significantly and positively
related to reactive control, but only marginally, inversely related to hyperactivity-

impulsivity.

53

Table 4

Correlations among Hormone Measures, Clinical Symptoms, and Trait and Cognitive
Control

 

 

R 2D:4D 2D:5D 3D:4D Estradiol Testosterone
Inatt Sx -.21** -.11 -.19** -.02 -.07
Hyper Sx -.20** -.09 -.19** -.16+ -.16*
Cog Probs (T) -.08 .03 -.09 .14 -.02
Hyper (T) -.09 -.01 -.06 .16+ -.17+
Cog Probs (P) -.11 -.05 -.12 .13 -.12
Hyper (P) -.16* -.11 -.14* -.01 .08
ODD sx -.14 -.08 -.08 .01 .03
CD sx .03 .03 -.04 .11 -.06
Consc .22“ .07 .17* —.01 .05
Rx Cont .15* .03 .23" .26** .14+
SSRT -.02 -.13 -.13 .09 .13
SD -.10 -.01 -.10 .07 -.09
Trails B -.02 .01 -.08 .07 .05

 

13193,. +p<.10. *p<.05. **p<.01. Inatt Sx= Inattentive symptoms. Hyper Sx=Hyperactive-
impulsive symptoms. Cog Probs (T )fieacher-rated Cognitive Problems. Hyper (T )=teacher-rated
Hyperactivity. Cog Probs (P)=parent-rated cognitive problems. Hyper (P)=parent-rated
hyperactivity. ODD Sx=Oppositional-Defiant symptoms. CD Sx=Conduct Disorder symptoms.
Consc=Conscientiousness. Rx Cont=Reactive Control. SSRT=Response Inhibition.
SD=Response Variability. Trails B=Set-Shifting.

Overall, finger-length ratios showed expected sex differences, supporting the
validity of the supposition that the ratios are indexing prenatal testosterone exposure and

providing a minimal necessary condition for proceeding to the main analyses. Significant

54

sex differences were apparent for trait, but not EF, measures, also consistent with the
literature. In addition, correlational analyses in the combined group of boys and girls
provide preliminary support and justification for formally testing the hypothesized
mediation analyses, which now follow.

Question 1: Do levels of prenatal testosterone as inferred from rigl_rt 2D:4D predict (a)
more mittentive symptoms and 02) more impaired EF or conscientiousness in bovs?

In boys, lower (i.e., more masculine) right 2D:4D predicted more inattentive
symptoms (B=-.23, z=-2.70, R’=.05, p<.01), controlling for ethnicity (covaried in all
subsequent analyses involving finger-length ratios). Lower right 2D:4D also significantly
predicted less conscientiousness (5:.25, z=3.20, R’=.08, p<.01). Right 2D:4D was not
significantly related to response inhibition (B=—.06, z=-.44, R2=.004, p>.05). Low
conscientiousness significantly predicted inattentive symptoms (fi=-.70, z=-11.74,
R’=.48, p<.01), while poor response inhibition only marginally predicted inattentive
symptoms ([3=.23, z=3.64, R’=.05, p<.05).

Right 2D:4D was marginally related to reactive control ([3=.19, z=-2.46, R’=.O6,
p<.05) and hyperactive-impulsive ADHD symptoms (B=—.20, z=-2.29, R’=.06, p<.05).
Right 2D:4D was not significantly related to ODD or CD symptoms (t3=-.16, z=-1.66,
R’=.04 for ODD symptoms; [3:04, z=.64, R’=.Ol; both p>.05).

For completeness, results were checked in the full sample (boys and girls
combined; without covarying sex), with the same results, as follows. Lower right 2D:4D

again predicted more inattentive symptoms ( )8=-.21, z=-3.26, R2=.05, p<.01) and less

55

conscientiousness (0:22, z=3.76, R’=.05, p<.01).5 Right 2D:4D was again not
significantly related to response inhibition ([3=-.04, z=-.43, R2=.002, p>.05). Low
conscientiousness significantly predicted inattentive symptoms ([3=-.74, z=-19.73,
R2=.55, p<.01), while poor response inhibition was a marginal predictor of inattentive
symptoms (5:.25, z=4.43, R’=.06, p<.05).

In the full group, lower right 2D:4D was marginally related to less reactive
control ([3=.16, z=2.46, R’=.02, p<.05) and significantly related to more hyperactive-
impulsive symptoms (B=-.l9, z=-2.82, R’=.04, p<.01). Right 2D:4D was not significantly
related to ODD (B=-.15, z=-1.79, R’=.03, p>.05) or CD symptoms (B=.02, z=.29, R2=.01,
p>.05).

In summary, conscientiousness, but not EF, emerged as a possible mediator of
prenatal testosterone effects on inattentive ADHD symptoms, because it was related to
both ADHD symptom and right 2D:4D. This was next tested formally.

Question 2: Does low conscientiousness mflme—mm
of testosterone 1nd inattentive ADI-ID syr_nptoms in boys?

Since response inhibition was not significantly related to right 2D:4D it could not
be a mediator based on the traditional Baron and Kenny (1986) prerequisites. Therefore,
only conscientiousness was examined as a mediator between right 2D:4D and ADHD.
Within boys, when right 2D:4D and conscientiousness were both entered as predictors of
inattentive symptoms, right 2D:4D no longer significantly predicted inattentive

symptoms (B=-.06, z=-.84, p>.05), but conscientiousness remained a significant predictor

 

5 Finger-length ratios, measured by calipers, were not significantly related to inattentive ADHD symptoms
(t3=-.26, p>.05) or conscientiousness (B=.16, p>.05) in boys and girls, although the direction of effects was
the same as for hand scan measurements.

56

(B=-.69, z=-10.86, R’ for full model=.49, p<.01). The mediation model was significant
(z=2.67, p<.01), according to the Sobel test developed by Preacher and Hayes (2004) and
the bootstrapping confidence interval did not contain zero (95% c.i.: -15.35 to -3.01),
supporting the mediation model in that the indirect path was significant. Thus, low
conscientiousness mediated the relationship between right 2D:4D and inattentive ADHD

symptoms in boys (Figure 4).

 

.25” Conscientiousness -.69**

 

 

 

 

 

Mediated path -,06
Right 2D:4D Inattentive
Symptoms

 

 

 

 

 

 

 

 

Direct path -.23**

Note. All paths are identified using standardized coefficients. Conscientiousness was
measured by maternal report. Inattentive symptoms were measured using a
parent+teacher OR algorithm using DSM-IV symptom criteria. 95% Ci. of mediation
effect: -15.35 to -3.01.
Figure 4. Conscientiousness Mediates the Relationship between Prenatal Testosterone
and Inattentive ADHD Symptoms in Boys

Again for completeness, the same effects were tested in boys and girls combined.
Once again, the same result held: When right 2D:4D and conscientiousness were both
entered as predictors of inattentive symptoms, right 2D:4D no longer significantly

predicted inattentive symptoms ([3=-.05, z=~1 . 13, p>.05), but conscientiousness remained

a significant predictor ([3=-.74, z=-l9.73, R2 for full model=.55, p<.01). The mediation

57

model was significant (z=3.21, p<.01), according to the Sobel test developed by Preacher
and Hayes (2004) and the bootstrapping confidence interval did not contain zero (95%
c.i.: -13.56 to -3.83), supporting the mediation model shown in Figure 4.

Qpestion 3: Doe__s__fgl more We regctive control predict higher or more optimal levels
of circulating estradiol and 0;) do higher levels of circulating estradiol predict less

hyperactive-impulsive ADHD smptoms in g'rls?

Within girls, higher reactive control did not significantly predict higher
circulating levels of estradiol (B=.09, z=.80, R’=.28, p>.05), and circulating estradiol did
not significantly predict hyperactive-impulsive ADHD symptoms ([3=.17, z=1.23, R’=.08,
p>.05), after controlling for ADHD versus control group differences in age. Lower
reactive control significantly predicted more hyperactive-impulsive symptoms ([3=-.36,
z=-4.20, R’=.18, p<.01).

In order to explore a possible curvilinear effect between estradiol levels and
outcome, a quartile approach was utilized to examine mean differences in hyperactive-
impulsive symptoms based on level of circulating estradiol. The first (25%) quartile was
comprised of girls with estradiol levels of less than or equal to 16.60 pc/mL. The second
(50%) quartile was comprised of girls with between 16.60 and 31.99 pc/mL of circulating
estradiol. The third (75%) quartile was comprised of girls with between 31.99 and 78.00
pc/mL of circulating estradiol. The fourth quartile was comprised of girls with over 78.00
pc/mL of circulating estradiol. These four groups were compared via MANCOVA with a
quadratic polynomial contrast. Age was covaried, and hyperactive-impulsive symptoms
and reactive control were the dependent variables. The MANCOVA was not significant

(F[6, 104]=1.36, ”2:07, p>.05).

58

Question 4: Does circulating estradiol mediate the relationship between reactive control
and hyperactive-impulsive ADHD smptoms in girls?

Reactive control did not significantly predict circulating estradiol, and circulating
estradiol did not significantly predict hyperactivity-unpulsivity. Thus, mediation did not
occur as evidenced by these patterns of relations.

Question 5: Do lowrmtios of testosterone to estradiol predict (a) less inattentive or

hyperactive-impulsive ADHD smptoms and 0;) better EF/conscientiousness or reactive
control in boys and girls?

In the full group, testosterone and estradiol were correlated weakly and
marginally at -.06 (p<.05). However, within sex, the correlation was substantial. In boys,
testosterone and estradiol were significantly correlated at .51 (p<.01). In girls,
testosterone and estradiol were significantly correlated at .35 (p<.01). Unsurprisingly,
testosterone level was substantially correlated with testosteronezestradiol ratio (r=.88,
p<.01). As expected, boys had higher levels of circulating levels of testosterone than
girls (t[163]=6.12, p<.01) and higher ratios of testosterone to estradiol (t[130]=8.48,
p<.01 ).

Because of the sex differences in circulating testosterone, interactions between
sex and testosterone and between sex and the ratio of testosterone to estradiol were
examined in regression analyses. Both testosterone and ratios of testosterone to estradiol
were examined in order to examine any potential testosterone effects that might be
independent of estradiol levels. In these regression analyses, ADHD symptoms, EF, and
the two personality traits were each separately regressed on sex, hormones, and the

interaction term, controlling for age. A significant interaction between sex and

59

testosterone was found in predicting inattentive symptoms ([3=-.80, p<.01), and a

marginal interaction between sex and testosterone was found in predicting response

variability (B=-1.89, p<.05). There was a stronger relationship between higher circulating

testosterone and fewer inattentive ADHD symptoms in girls (B=-.22, z=-2.80, R’=. 16,

p<.01) than in boys (where the relationship was essentially zero; B=.02, F. l 3, R2=.03,

p>.05 [shown in Figure 5]). There was also a stronger relationship between higher
circulating testosterone and decreased response variability in girls ( B=-.5 3, z=-1.95,
R’=.27, p<.10) than in boys (where the relationship was nearly zero, B=-.08, z=-.42,

R2=.04, p>.05 [shown in Figure 6]). Other interactions were not significant (p>.05).

20.00

 

 

 

 

 

.011
180° Inattentive Response 190
160° Symptoms Variability 170'“
14.00 150.00
12-00 __ ___ _. 130.00
10.00 " - 110.00
:3: 90.00
4:00 70-00
_ .. [ .2
1128:; 5°-°° gar:
0.0 - - 30. .
Low High Low High
Testosterone Testosterone
Note. Inattentive
symptoms=teacher+parent rated
symptoms defined using OR algorithm.
Figure 5. Higher Levels of Testosterone Figure 6. Higher Levels of Testosterone
are Related to Fewer Inattentive ADHD are Related to Less Response Variabrlrty
Symptoms in Girls 1n Gll‘lS.

With regard to the ratio of testosterone to estradiol, marginal interactions by sex

60

were found for inattentive symptoms, hyperactive-impulsive symptoms, and
conscientiousness (all p<.05). For girls, there was a significant relationship between
testosterone:estradiol (T:E) ratios and inattentive symptoms (B=-.43, z=-4.36, R2=.28,
p<.01) such that more testosterone relative to estradiol (i.e., larger ratios) predicted fewer
symptoms (shown in Figure 7); for boys, this relationship was slightly negative, but not
nearly significant ([3=-.05, z=-.3 7, R’=.03, p>.05). Girls also exhibited a stronger
relationship between T:E ratios and hyperactive-impulsive symptoms (B=-.40, z=-4.22,
R’=.20, p<.01) than did boys (where the relation was slightly negative but not significant;
l3=-. 13, z=-.93, R’=.12, p>.05) such that more testosterone relative to estradiol (i.e.,

larger ratios) was related to decreased hyperactive-impulsive symptoms (shown in Figure

8).

61

 

 

20.00 Inattentlve Impulsive 20.00
18.00 Symptoms Symptoms 18.00
16.00 16.00
14.00 14.06
12.00 12.00
10.00 10-00
8.00 8.00
6_oq 6.00
4.00 4-00
2m +Boys 2.00
0.‘ 07

 

Low ' High
Testosterone: Estradiol Ratio

N913. Inattentive
symptoms=teacher+parent rated
symptoms defined using OR algorithm.
Low ratios=more estradiol relative to
testosterone. High ratios=more
testosterone relative to estradiol.

Figure 7. High TestosteronezEstradiol
Ratios are Related to Fewer Inattentive
ADI-ID Symptoms in Girls

Hyperactive-

 

 

 

Low ' High
Testosterone: Estradiol Ratio

No_te. Hyperactive-impulsive
symptoms=tea¢her+parent rated
symptoms defined using OR
algorithm. Low ratios=more
estradiol relative to testosterone.
High ratios=more testosterone
relative to estradiol.

Figure 8. High Testosterone:Estradiol
Ratios are Related to Fewer
Hyperactive-Impulsive ADHD
Symptoms in Girls

Likewise, girls exhibited a stronger relationship between high T:E ratios and

conscientiousness (which is inversely related to inattention; B=.26, z=2.22, R’=.15,

p<.05) than did boys ([3=.01, z=.07, R’=.04, p>.05) such that more testosterone relative to

estradiol (i.e., larger ratios) was related to increased conscientiousness (shown in Figure

9). No other interactions (in regard to ADI-ID symptom domains, ODD symptoms, CD

symptoms, or other traits or EF) were significant (all p>.05). Overall, it appeared that

high levels of testosterone and high levels of testosterone relative to estradiol were

protective for girls in regard to ADI-II) symptoms, response variability, and

conscientiousness.

62

conscientiousness.

900' Conscien-
8.00- trousness

7.00'
6.00‘
5.00‘
4.001
3.00‘
2.00'

1.00- +B°Ys I
+ Girls

0.00 .
Low High
Testosterone: Estradiol Ratio

 

 

 

 

Note. Low ratios=more estradiol relative to testosterone. High ratios=more
testosterone relative to estradiol.

Figure 9. High TestosteronezEstradiol Ratios are Related to Increased Conscientiousness
in Girls

Therefore, three moderated mediation models were tested. Response variability
was examined as a mediator of the relationship between circulating testosterone levels
and inattentive symptoms in girls. Conscientiousness was examined as a mediator of the
relationship between T:E ratios and inattentive symptoms in girls, and conscientiousness
was examined as a mediator of the relationship between T:E ratios and hyperactive-
impulsive symptoms in girls. Age was covaried in all moderated mediation analyses. In
none of the three models did all paths reach significance. Moderated mediation effects
were not significant for the relations among circulating testosterone, response variability,

and inattentive symptoms (95% c.i.: -4.98 to .002; p=. 1 3). Moderated mediation effects

63

were also not significant for the relations among T:E ratios, conscientiousness, and
inattentive symptoms (95% c.i.: -.17 to .07; p=.25) or for the relations among T:E ratios,
conscientiousness, and hyperactive-impulsive symptoms (95% c.i.: -.10 to .02; p=.35).
Thus, in none of the models was moderated mediation supported. However, it should be
noted that power was low to detect these effects, particularly because they occurred in

girls.

64

DISCUSSION

Hormonal effects are often-touted, but seldom empirically examined, as key
factors in childhood psychopathology and in ADHD specifically. The current study
examined sex differences in ADHD in relation to mechanisms of top down (i.e.,
conscientiousness and EF) and bottom up (i.e., reactive) control. It also examined
whether these behavioral probes might mediate relations between prenatal as well as
circulating hormones and current childhood ADHD symptoms.

Summary of findings

The main four findings were as follows. First, normative sex differences in
conscientiousness found in the control group were not evident in children with ADHD,
and sex differences in reactive control seen in the ADHD group were not evident in the
control group. Second, more masculine finger-length ratios, which are indicative of
higher exposure to prenatal testosterone, were related to lower levels of conscientiousness
and more inattentive ADHD symptoms in boys. Third, lower conscientiousness mediated
the relationship between right 2D:4D and inattentive ADHD symptoms. Fourth, levels of
circulating estradiol were not related to reactive control or hyperactivity, after controlling
for age. However, increased circulating testosterone was related to less inattentive and
hyperactive-impulsive ADHD symptoms in girls.

Overall, results are consistent with study hypotheses that high levels of prenatal
testosterone lead to decreased conscientiousness, which in turn increases risk for
inattentive symptoms of ADHD. Study hypotheses about circulating estradiol in girls
were not supported. Instead, results were consistent with the idea that increased

circulating levels of testosterone may provide some protection against ADHD symptoms

65

in girls. That effect was not hypothesized and so must be replicated. Key findings are
now discussed in turn.
Sex Diflerences in Traits and EF

This was the first study to examine sex differences in traits (i.e.,
conscientiousness and reactive control) and EF, comparing across control and ADHD
groups. Trait sex differences varied between the ADHD and control groups, manifested
in significant interactions. While girls had higher levels of conscientiousness than boys
within the control group, girls and boys in the ADHD group had comparable, low levels
of conscientiousness. The sex difference in conscientiousness found in the control group
is in keeping with past research. In childhood, girls typically exhibit higher levels of
conscientiousness (i.e., are more planful) than boys (e.g., Else-Quest et al., 2006). These
kinds of sex differences may not be due entirely to sex, but rather may be due to prenatal
hormonal exposure that occurs in sex-specific patterns.

Sex and diagnostic differences were also noted for reactive control (i.e., reflexive,
rather than effortful, regulation). Girls with ADHD had higher levels of reactive control
than boys with the disorder, whereas girls and boys within the control group had
comparable levels of reactive control. One potential interpretation of this finding is that
reactive control serves as a sex-specific phenotype for childhood ADHD. As noted in
longitudinal research by the author, boys and girls may arrive at ADHD through different
routes (Martel, Lucia, Nigg, & Breslau, 2007). Boys may be at increased risk for ADHD
symptoms due to poor motor coordination and increased motor restlessness (Martel et al.,
2007). However, higher levels of reactive control may increase risk for ADHD symptoms

in girls, potentially interacting with other types of reactive control such as negative

66

emotionality (Martel & Nigg, 2006). This interpretation would be consistent with the
constitutional variability model of ADHD, which states that different causal factors are
important for girls and boys (Rhee, Waldman, Hay, & Levy, 2001). Higher levels of
reactive control may predispose girls, but not boys, to ADHD.

Although sex differences in traits were apparent, there were no sex differences in
EF (i.e., response inhibition, response variability, or set-shifting) across the sample, or
within the ADHD or control groups. Furthermore, hormones were not related to EF. This
finding raises challenges for the theory of hormonal associations with ADHD
mechanisms that was proposed in the current dissertation. One possible revision of the
theory might be to acknowledge that EF may not be functioning as a true endophenotype
for biological risk factors of ADHD in childhood. A second revision of the theory might
propose that hormones may be related to more broad-based alterations of neural function
rather than to alterations in the more localized neural circuitry implicated in EF.

EF may be more heavily influenced by socialization and experience than some
recent theories have suggested. For example, some aspects of EF, such as response
inhibition, may be more sensitive to the influence of parenting and classroom influences
than others (e.g., set-shifting). EFs are, in fact, influenced by context characteristics such
as time of day and structure of the environment (Antrop, Roeyers, & De Baecke, 2005).
Recent research suggests that EF relations with genes, another biological marker, are
relatively weak for individual EF measures, but stronger for the shared variance between
EF measures (Friedman et al., in press). Another consideration is that EF may be heavily
reliant upon frontal-striatal neural circuitry early in development (e.g., between the ages

of three and six), but become more routinized and experiential-based during mid-to-late

67

childhood (e.g., between the ages of six and twelve). Thus, studies examining older
children and adolescents may fail to find associations between EF and biological risk
factors due to increased environmental influences on EF at later ages.

Hormones may be influencing general indices of brain development, such as rate
of fetal growth, cerebral lateralization, and effects on the prefrontal cortex. In this way,
they may have an impact on broader measures of behavior such as psychopathology and
traits which capture a range of behaviors across numerous settings over a period of time,
rather than specific measures of EF which capture a single behavior in a single setting at
a single point in time. EF measures may be limited in their ability to tap into more broad-
based brain alterations since they measure more localized neural activity. In this way,
there may be components of conscientiousness (e.g., broad-based effortful control),
unrelated to response inhibition, that are related both to hormones and ADHD.
Alternatively, EF may be protected from hormonal influences on the striatum via its
interaction with other neural systems that are not as DA-dependent, such as the
cerebellum.

Traits and ADHD may be more related than EF and ADHD since measurement of
behavioral traits and symptoms are more similar than the measurement of EF and
symptoms. Traits may fall on a continuum with psychopathology, as suggested by a
spectrum model of trait-psychopathology relations (Watson, Kotov, & Gamez, 2006). If
extreme levels of maladaptive trait control predispose individuals to behavioral
symptoms of ADHD or entirely overlap with ADHD symptoms, this phenomenon may
account for the sex differences found in both domains. Temperament traits and ADHD

symptoms may also be related because low levels of trait control predispose individuals

68

to the development of ADHD. This idea is consistent with a vulnerability model of trait-
psychopathology relations (Watson et al., 2006). If traits do, in fact, predispose
individuals to the development of psychopathology, traits may be a more proximal index
of prenatal hormones than EF. Similarity of measurement of traits and symptoms may
also conflate their mutual associations with sex and hormones.
Prenatal Testosterone Efi'ects and ADHD

Lower (i.e., more masculine) finger-length ratios, particularly the right 2D:4D,
were related to more inattentive ADHD symptoms. This finding replicates other research
studies that have found that more masculine finger-length ratios are related to more
externalizing problems and more ADHD symptoms (de Bruin et al., 2006; Fink et al.,
2007; McFadden et al., 2005; Stevenson et al., 2007; Williams et al., 2003). Increased
prenatal testosterone was related to trait dysregulation more generally, in that it was
associated with lower levels of conscientiousness and increased ADHD symptoms. If
ADHD is seen as emanating from a trait, rather than influencing the development of a
trait, then the data were consistent with the idea that high levels of prenatal testosterone
increase risk for low conscientiousness and the subsequent development of inattentive
ADHD symptoms. In the current study, more masculine finger-length ratios were related
to lower conscientiousness, which mediated the relationship between those ratios and
inattentive ADHD symptoms in boys.

Exposure to prenatal testosterone, rather than male sex per se, was associated with
ADHD and some forms of trait dysregulation. Current results are consistent with the idea
that higher prenatal testosterone exposure increases inattention and does so via its

association with conscientiousness in childhood. Children (i.e., both boys and girls) with

69

ADHD had more masculine finger-length ratios and lower levels of conscientiousness
than children in the control group. In contrast, girls in the control group were
characterized by higher levels of conscientiousness than boys in the control group and
also had higher (i.e., more feminine) finger-length ratios.

Circulating Hormonal Eflects and ADHD

Circulating estradiol did not appear to be related to fewer ADHD symptoms in
girls via an association with reactive control, contrary to study hypotheses. Contrary to
the theory proposed in this dissertation, circulating estradiol does not appear to be related
to inattentive ADHD symptoms, hyperactive-impulsive ADHD symptoms or reactive
control. Estradiol does not appear to have an effect on the nucleus accumbens or on the
forms of reactive control that rely on this region of the brain. However, future research
might examine effects of circulating estradiol on ADHD symptoms without controlling
for age. It is likely that important variance in estradiol due to age and pubertal status was
eliminated when controlling for age differences between the ADHD and control groups.
Future research might examine estradiol-ADHD relations in a sample of children
matched on age so as to assess whether hyperactive-impulsive ADHD symptoms improve
over time due to age-related increases in estradiol.

Results indicated that ratios between testosterone and estradiol and/or circulating
levels of testosterone were associated with ADHD in girls. Higher levels of testosterone
were associated with fewer ADHD symptoms of inattention and hyperactivity-
impulsivity in girls. This finding was unexpected and needs replication. However, if this
finding was replicated, one possible explanation is that circulating testosterone levels

counterbalance circulating estradiol levels.

70

However, an alternative explanation is that circulating testosterone levels are
related to fewer symptoms in girls through another, currently unknown, protective
mechanism. For example, protective effects of circulating testosterone may be seen in
girls but not boys due to the fact that an optimal level of testosterone is related to
decreased ADHD symptoms. Since girls have lower levels of testosterone than boys, an
optimal effect may be more detectable in girls than in boys. A future study might address
this point by examining curvilinear effects between testosterone and ADHD symptoms in
a larger sample of girls across a wide range of ages.

Alternatively, the organizational effects of prenatal testosterone exposure in boys
may make them less sensitive to circulating testosterone levels than girls. Circulating
levels of testosterone in girls may have activational effects on the expression of ADHD
symptoms in childhood, potentially via neurosteriod interactions with dopaminergic
neurotransmission. The examination of circulating testosterone-ADHD relations in girls
with congenital adrenal hyperplasia (CAH), a condition in which girls are exposed to
high levels of prenatal androgens, would allow for the examination of organizational
effects of testosterone in girls. If boys’ insensitivity to circulating testosterone is due to
the organizational effects of androgens, circulating levels of testosterone would not be
protective for ADHD in girls with CAH. It should be noted that the relationship between
testosterone and ADHD found in girls in this study was not hypothesized and requires
replication.

Implications for Neural Development and Pathways to Developmental Psychopathology

Although these data are cross-sectional (and one could argue that ADHD

influences conscientiousness rather than the other way around), the developmental

71

implications are still interesting. Prenatal testosterone exposure may influence the
development of dopaminergic neural circuitry both during the prenatal period and
throughout childhood. Differences in such neural circuitry and decreased extracellular
DA may then predispose an individual toward certain behavioral tendencies that are first
manifest in characteristic traits, beginning around 18 months of life and consolidating
around age three or four (Aksan & Kochanska, 2004; Zuckerman, 2005). The
characteristic trait of low control may begin to seem developmentally inappropriate and
dysfunctional around six or seven years of age, since at this age children begin school and
are expected to use trait and behavioral control to remain seated and listen to their
teachers. At this stage of development, low levels of trait control may be manifest in
behavioral syndromes like ADHD. This idea could be in line with a vulnerability (i.e.,
certain traits increase the likelihood of psychopathology), common cause (i.e, traits and
psychopathology have a common cause), or spectrum (i.e., traits and psychopathology
fall on a continuum) model of personality/temperament trait to psychopathology relations
(Caspi & Shiner, in press; Watson et al., 2006).

However, as recent dual—pathway (Sonuga-Barke, 2005) or multi-pathway (N igg,
Goldsmith, & Sachek, 2004; Nigg, 2006) models suggest, there are likely multiple routes
to the development of ADHD. In the instance of ADHD, prenatal androgen may
influence the development of dopaminergic circuitry underpinning trait control, resulting
in the precursors of low conscientiousness (e.g., low effortful control), occurring more
often in boys. Low effortful control/conscientiousness may then emerge around the age
of three or four years (Aksan & Kochanska, 2004) and eventually lead to behavioral

symptoms of inattention. These symptoms may become most evident when the child

72

(most often, a boy) becomes old enough to begin attending school, although it should be
noted that recent research indicates that ADHD can be reliably diagnosed as young as
four years old (Lahey et al., 2005). Optimal levels of circulating testosterone and T:E
ratios, rather than estradiol, may provide some protection against the further development
of symptoms later in childhood in girls, potentially via neurosteriod interactions with DA
neurotransmission in the frontal-striatal region of the brain.

More generally, hormones have implications for risk and protective factors for
many disorders, including male-biased, childhood-onset, externalizing disorders and
female-biased, adolescent-onset, internalizing disorders. However the present research
suggests these effects will not be easy to find. There was no effect on EF, which weakens
our ability to use hormones to explain ADHD-related phenomena. However, there were
effects on traits, which is encouraging as a preliminary step to mapping these associations
to the development of childhood ADHD. Prenatal and circulating hormonal mechanisms
have the potential to illuminate the complex biological risk pathways that predispose
individuals to developing ADHD in early childhood. Hormonal associations with
childhood psychopathology also have the potential to illuminate mechanisms of sex
differences in trait and cognitive correlates of psychopathology, as well as sex-biased
prevalence rates favoring boys in childhood.

Limitations

The current study had several limitations. Because the design was cross-sectional
and non-experimental, causality cannot be addressed (and, in fact, as was noted in
passing, ADHD may be leading to lower conscientiousness rather than the other way

around). The study had a relatively small sample of girls with available measures of

73

circulating estradiol and testosterone. If a larger sample had been available, mediation
analyses might have been more conclusive. Additionally, it would have been informative
to examine hormone-psychopathology relations in pre and post-menses girls separately,
as well as to examine relations across the phases of the menstrual cycle. Sample size was
too small to allow this kind of analysis in the current study.

The current study did not correct for item or conceptual overlap between traits
and ADHD symptom counts since the integrity of trait and ADHD symptom scales were
judged to be essential to the current work. Further, the prior work of the author indicated
that the effects of item overlap are often modest (Martel & Nigg, 2006). In addition,
multiple measures of temperament traits were not available. Future work might examine
observational measures of traits in addition to parental report on rating forms. Finally,
finger lengths and ratios were measured using photocopies and rulers. Questions about
the reliability and validity of this measure remain. Analyses conducted on finger-length
ratios measured via calipers utilized a sample that was too small to yield reliable findings.
However, the effects of the caliper analyses were in the same direction as those
conducted using hand scan measurements, providing some support for the validity of the
hand scan approach to finger-length measurement. Future work might examine
correlations between finger-length ratios, measured via photocopies/rulers and calipers,
and actual prenatal testosterone exposure.

Summary

Overall, sex differences in key traits relevant to ADHD are apparent. Prenatal

testosterone was related to inattentive symptoms of ADHD in boys, and this relationship

was mediated statistically by low levels of conscientiousness. Circulating estradiol was

74

not related to reactive control or hyperactive-impulsive symptoms of ADHD. Higher
levels of testosterone may protect against ADHD symptoms in girls, but this unexpected
finding needs replication. Gonadal hormones appear to be related to ADHD, but the
nature of these effects will require the continuing revision of theory. Broadly, gonadal
hormones appear to increase general risk and protective factors for ADHD, potentially

acting through trait mechanisms.

75

REFERENCES

Aksan, N. & Kochanska, G. (2004). Links between systems of inhibition from infancy to
preschool years. Child Development, 75(5), 1477-1490.

Alexander, L.Y., Crutcher, M.D., & DeLong, MR. (1991). Basal ganglia thalamocortical
circuits: Parallel substrates for motor, oculomotor, prefrontal and limbic
functions. Progress in Brain Research, 85, 119-145.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental
disorders (Fourth ed., text rev.). Washington, DC: American Psychiatric
Association.

Andersen, S.L. & Teicher, M.H. (2000). Sex differences in dopamine receptors and their
relevance to ADHD. Neuroscience and Biobehavioral Reviews, 24, 137-141.

Angold, A., Costello, E.J., Erkanli, A., & Worthman, CM. (1999). Pubertal changes in
hormone levels and depression in girls. Psychological Medicine, 29, 1043-1052.

Antrop, 1., Roeyers, H., & De Baecke, L. (2005). Effects of time of day on classroom
behaviour in children with ADHD. School Psychology International, 26(1), 29-
43.

Arnold, L.E. (1996). Sex differences in ADHD: Conference Summary. Journal of
Abnormal Psychology, 24(5), 555-569.

Austin, E.J., Manning, J.T., Mclnroy, K., & Mathews, E. (2002). A preliminary
investigation of the associations between personality, cognitive ability and digit
ratio. Personality and Individual Differences, 33, 1115-1124.

Band, G.P.H., van der Molen, M.W., & Logan, GD. (2003). Horse-race model
simulations of the stop-signal procedure. Acta Psychologica, 112, 105-142.

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

Barkley, RA. (2003). Attention-Deficit/Hyperactivity Disorder. In El. Mash & R.A.
Barkley (Eds) Child Psychopathology (Second Ed., pp. 75-143). The Guilford
Press: New York.

Baron, R.M. & Kenny, DA. (1986). The moderator-mediator variable distinction in

social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51 (6), 1 173-1 182.

76

Bauerrneister, J.J., Shrout, P.E., Chavez, L., Rubio-Stipec, M., Ramirez, R., Padilla, L.,
Anderson, A., Garcia, R, & Canino, G. (2007). ADHD and gender: Are risks and
sequela of ADHD the same for boys and girls? Journal of Child Psychology and
Psychiatry, 48(8), 831-839.

Becker, J. (1999). Gender differences in dopaminergic function in striatum and nucleus
accumbens. Pharmacology, Biochemistry, and Behavior, 64(4), 803-812.

Becker, J .B. & Rudick, ON. (1999). Rapid effects of estorgen or progesterone on the
amphetamine-induced increase in striatal dopamine are enhanced by estrogen
priming: A microdialysis study. Pharmacology Biochemistry and Behavior, 64(1),
53-57.

Befera, M.S. & Barkley, RA. (1985). Hyperactive and normal girls and boys: Mother-
child interaction, parent psychiatric status, and child psychopathology. Journal of
Child Psychology and Psychiatry, 26, 439-452.

Berenbaum, S.A., Korrnan, K., & Leveroni, C. (1995). Early hormones and sex
differences in cognitive abilities. Learning and Individual Differences, 7(4), 303-
321.

Berger, D.F., & Sagvolden, T. (1998). Sex differences in operant discrimination
behaviour in an animal model of attention-deficit hyperactivity disorder.
Behavioural Brain Research, 94(1), 73-82.

Biederman, J ., Faraone, S.V., Keenan, K., Benjamin, J ., Krifcher, B., Moore, C., Sprich-
Buckminster, S., Ugaglia, K., Jellinek, M.S., Steingard, R., Spencer, T., Norman,
D., Kolodny, R, Kraus, 1., Perrin, J., Keller, M.B., & Tsuang, MT. (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 Psychiatry, 49, 728-73 8.

Biederman, J., Faraone, S.V., Keenan, K., Knee, D., & Tsuang, MT. (1990). Family-
genetic and psychosocial risk factors in DSM-III attention deficit disorder.
Journal of the American Academy of Child and Adolescent Psychiatry, 29(4),
526-533.

Biederman, J., F araone, S.V., Mick, E., Williamston, S., Wilens, T.E., Spencer, T.J.,
Weber, W., Jetton, J., Kraus, I., Pert, J ., & Zallen, B. (1999). Clinical correlates of
ADHD in females: Findings from a large group of girls ascertained from pediatric
and psychiatric referral sources. Journal of the American Academy of Child and
Adolescent Psychiatry, 38(8), 966-975.

Biederman, J., Mick, E., Faraone, S.V., Braaten, E., Doyle, A., Spencer, T., Wilens, T.E.,
Frazier, E., & Johnson, MA. (2002). Influence of gender on attention deficit

77

hyperactivity disorder in children referred to a psychiatric clinic. The American
Journal of Psychiatry, 159(1), 36-42.

Breedlove, S.M. & Hampson, E. (2002). Sexual differentiation of the brain and behavior.
In J .8. Becker, S.M. Breedlove, D. Crews, & M.M. McCarthy (Eds) Behavioral
Endocrinology (Second Ed.;, pp.75-114). MIT Press: Cambridge, MA.

Brown, W.M., Hines, M., Fane, B.A., & Breedlove, SM. (2002). Masculinized finger
length patterns in human males and females with Congenital Adrenal
Hyperplasia. Hormones and Behavior, 42, 380-3 86.

Casey, B.J., Castellanos, F.X., Giedd, J.N., Marsh, W. L., Hamburger, S.D., Schubert,
A.B., Vauss, V.C., Vaituzis, A.C., Dickstein, D.P., Sarfatti, S.E., & Rapoport, J.L.
(1997a). Implication of right frontostriatal circuitry in response inhibition and
attention deficit/hyperactivity disorder. Journal of the American Academy of

Child and Adolescent Psychiatry, 36(3), 374-384.

Casey, B.J., Trainer, R., Giedd, J ., Vauss, Y., Vaituzis, C.K., Hamburger, S., Kozuch, P.,
& Rapoport, J.L. (1997b). The role of the anterior cingulated in automatic and
controlled processes: A developmental neuroanatomical study. Developmental
Psychobiology, 30, 61-69.

Casey, B]. (2005). Frontostriatal and frontocerebellar circuitry underlying cognitive
control. In U. Mayr, E. Awh, & S.W. Keele (Eds), Developing individuality in
the human brain: A tribute to Michael I. Posner (pp. 141 -l66). Washington, DC:
APA.

Caspi, A., Block, J., Block, J.H., Klopp, B., Lynarn, D., Moffitt, T.E., & Stouthamer-
Loeber, M. (1992). A “common-language” version of the California Child Q-Set
for personality assessment. Psychological Assessment, 4(4), 512-523.

Caspi, A. & Shiner, R.L. (in press). Temperament and personality. In M. Rutter, D.
Bishop, D. Pine, S. Scott, J. Stevenson, E. Taylor, & A. Thapar (Eds), Rutter ’s
Child and Adolescent Psychiatry (5th Ed.). London: Blackwell.

Castellanos, F .X., Giedd, J.N., Berguin, P.C., Wlater, J .M., Sharp, W., Tran, T., Vaituzis,
C., Blumenthal, J.D., Nelson, J ., Bastain, T.M., Zijdenbos, A., Evans, A.C., &
Rapoport, J .L. (2001). Quantitative brain magnetic resonance imaging in girls
with Attention-Deficit/Hyperactivity Disorder. Archives of General Psychiatry,
58(3), 289-295.

Castellanos, F .X., Sonuga-Barke, E.J.S., Scheres, A., Di Martino, A., Hyde, C., &

Walters, J .R. (2005). Varieties of Attention-Deficit/Hyperactivity Disorder-related
intra-individual variability. Biological Psychiatry, 57, 1416-1423.

78

Castellanos, F.X. & Tannock, R. (2002). Neuroscience of Attention-Deficit/Hyperactivity
Disorder: The search for endophenotypes. Nature Reviews Neuroscience, 3, 617-
628.

Cohen-Bendahan, C.C.C., van de Beek, C., & Berenbaum, SA. (2005). Prenatal sex
hormone effects on child and adult sex-typed behavior: Methods and findings.
Neuroscience and Biobehavioral Reviews, 29, 353-3 84.

Collaer, M.L. & Hines, M. (1995). Human behavioral sex differences: A role for gonadal
hormones during early development? Psychological Bulletin, 118(1), 55-107.

Conners, C. K. (1997). Conners Rating Scales-Revised. Toronto: Multi-Health Systems,
Inc.

Curran, S.G., West, S.G., & Finch, J .F . (1996). The robustness of test statistics to
nonnorrnality and specification error in confirmatory factor analysis. Psychological
Methods, 1(1), 16-29.

Daniel, J .M., Sulzer, J.K., & Hulst, J .L. (2006). Estrogen increases the sensitivity of
ovariectomized rats to the disruptive effects produced by the antagonism of D2 but

not D1 dopamine receptors during performance of a response learning task.
Hormones and Behavior, 49, 38-44.

De Bruin, E., Verhiej, F ., Wiegman, T., & Ferdinand, RF. (2006). Differences in finger
length ratio between males with autism, pervasive developmental disorder-not
otherwise specified, ADHD, and anxiety disorders. Developmental Medicine and
Child Neurology, 48(12), 962-965.

Dluzen, D.E. & McDermott, J .L. (2002). Estrogen, anti-estrogen, and gender: Differences
in methamphetamine neurotoxicity. In SF. Ali (Ed.) Cellular and Molecular
Mechanisms of Drugs of A buse II: Cocaine, Substituted Amphetamines, GHB, and
Opiates (pp. 136-156). New York Academy of Sciences: New York.

Dluzen, D.E. & Mickley, KR. (2005). Gender differences in modulatory effects of
tamoxifen upon the nigrostriatal dopaminergic system. Pharmacology,
Biochemistry, and Behavior, 80, 27-33.

Dohanich, G. (2002). Gonadal steroids, learning, and memory. In D.W. Pfaff, A.P.
Arnold, A.M. Etgen, S.E. Fahrbach, & R.T. Rubin (Eds), Hormones, Brain and
Behavior (Vol. 2; pp. 265-327). Boston: Academic Press.

Dominquez, J .M. & Hull, EM. (2005). Dopamine, the medial preoptic area, and male
sexual behavior. Physiology & Behavior, 86, 356-368.

DuPaul, G. J., Power, T. J., Anastopolous, A. D., & Reid, R. (1998). ADHD Rating
Scale—I V: Checklists, Norms, & Clinical Interpretation.

79

Durston, S. (2003). A review of the biological bases of ADHD: What have we learned
from imaging studies? Mental Retardation and Developmental Disabilities
Research Reviews, 9, 184-195.

Eisenberg, N., Cumberland, A., Spinrad, T.L., F abes, R.A., Shepard, S.A., Reiser, M.,
Murphy, B.C., Losoya, S.H., & Guthrie, I.K. (2001). The relations of regulation
and emotionality to children’s externalizing and internalizing problem behavior.
Child Development, 72(4), 1112-1134.

Eisenberg, N., Fabes, R.A., Guthrie, I.K., Murphy, B.C., Maszk, P., Holmgren, R., &
Suh, K. (1996). The relations of regulation and emotionality to problem behavior
in elementary school children. Development and Psychopathology, 8, 141-162.

Eisenberg, N., Fabes, F .A., Guthrie, I.K., & Reiser, M. (2000). Dispositional
emotionality and regulation: Their role in predicting quality of social functioning.
Journal of Personality and Social Psychology, 78, 136-157.

Eisenberg, N., Spinrad, T.L., Fabes, R.A., Reiser, M., Cumberland, A., Shepard, S.A.,
Valiente, C., Losoya, S.H., Guthrie, I.K., & Thompson, M. (2004). The relations
of effortful control and impulsivity to children’s resiliency and adjustment. Child
Development, 75(1), 25-46.

Else-Quest, N.M., Hyde, J.S., Goldsmith, H.H., & Van Hulle, CA. (2006). Gender
differences in temperament: A meta-analysis. Psychological Bulletin, 132(1), 33-
72.

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

Etgen, AM. (2002). Estrogen regulation of neurotransmitter and growth factor signaling
in the brain. In D.W. Pfaff, A.P. Arnold, A.M. Etgen, S.E. Fahrbach, & R.T.
Rubin (Eds), Hormones, Brain and Behavior (Vol. 3; pp. 381-440). Boston:
Academic Press.

Fabes, R.A., Eisenberg, N., Jones, 8., Smith, M., Guthrie, I., Poulin, R., Shepard, S., &
Friedman, J. (1999). Regulation, emotionality, and preschoolers’ socially
competence peer interactions. Child Development, 70(2), 432-442.

Fink, B., Manning, J .T., Williams, J .H.G., & Podmore-Nappin, C. (2007). The 2"d to 4th
digit ratio and developmental psychopathology in school-aged children.
Personality and Individual Differences, 42, 369-379.

Fink, B., Neave, N., Laughton, K., & Manning, J .T. (2006). Second to fourth digit ratio
and sensation seeking. Personality and Individual Differences, 41, 1253-1262.

80

Friedman, N.P., Miyake, A., Young, S.E., DeFries, J.C., Corley, R.P., & Hewitt, I.K. (in
press). The unity and diversity of executive functions are almost entirely genetic
in origin.

Gaub, M., & Carlson, CL. (1997). Gender differences in ADHD: A meta-analysis and
critical review. Journal of the American Academy of Child and Adolescent
Psychiatry, 36(8), 1 036- 1045.

Gerardin, D.C.C., Pereira, O.C.M., Kempinas, W.G., Florio, J.C., Moreira, E.G., &
Bemardi, M.M. (2005). Sexual behavior, neuroendocrine, and neurochemical
aspects in male rats exposed prenatally to stress. Physiology & Behavior, 84, 97-
104.

Gershon, J. (2002). A meta-analytic review of gender differences in ADHD. Journal of
Attention Disorders, 5(3), 143-154.

Goldstein, J.M., Jerram, M., Poldrack, R., Ahern, T., Kennedy, D.N., Seidman, L.J., &
Makris, N. (2005). Hormonal cycle modulates arousal circuitry in women using

functional magnetic resonance imaging. The Journal of Neuroscience, 25 (40),
9309-9316.

Goodman, R. (1991). Developmental disorders and structural brain development. In M.
Rutter & P. Casaer (Eds), Biological Risk F actors for Psychosocial Disorders
(pp. 260-310). Cambridge: Cambridge University Press.

Greene, R.A., Biederman, J., Faraone, S.V., Monuteaux, M.C., Mick, E., DuPre, E.P.,
Fine, C.S., & Goring, J. C. (2001). Social impairment in girls with ADHD:
Patterns, gender comparisons, and correlates. Journal of the American Academy
of Child and Adolescent Psychiatry, 40(6), 704-710.

Handa, R.J., Hejna, G.M., & Lorens, SA. (1997). Androgen inhibits neurotransmitter
trunover in the medial prefrontal cortex of the rat following exposure to a novel
environment. Brain Research, 751, 131-138.

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

Hartung, C.M., Willcutt, E.G., Lahey, B.B., Pelharn, W.E., Loney, J ., Stein, M.A., &
Keenan, K. (2002). Sex differences in young children who meet criteria for
attention deficit hyperactivity disorder. Journal of Clinical Child and Adolescent
Psychology, 31(4), 453-464.

Hermens, D.F., Williams, L.M., Lazzaro, I., Whitrnont, S., Melkonian, D., & Gordon, E.

(2004). Sex differences in adult ADHD: A double dissociation in brain activity
and autonomic arousal. Biological Psychology, 66(3), 221-233.

81

Hill, D.E., Yeo, R.A., Campbell, R.A., Hart, B., Vigil, J ., & Brooks, W. (2003). Magnetic
resonance imaging correlates of Attention-Deficit/Hyperactivity Disorder in
children. Neuropsychology, 17(3), 496-506.

Hines, M. (2002). Sexual differentiation of human brain and behavior. In D.W. Pfaff,
A.P. Arnold, A.M. Etgen, S.E. Fahrbach, & R.T. Rubin (Eds), Hormones, Brain
and Behavior (Vol. 4; pp.425-462). Boston: Academic Press.

Hinshaw, SP. (2002). Preadolescent girls with Attention/Deficit/Hyperactivity: I.
Background characteristics, comorbidity, cognitive and social functioning, and
parenting practices. Journal of Consulting and Clinical Psychology, 70(5), 1086-
1098.

Hu, M., Crombag, H.A., Robinson, T.E., & Becker, J .B. (2004). Biological basis of sex
differences in the propensity to self-administer cocaine.
Neuropsychopharmacology, 29, 81-85.

Huey, S.J. & Weisz, J .R. (1997). Ego control, ego resiliency, and the Five-Factor model
as predictors of behavioral and emotional problems in clinic-referred children and
adolescents. Journal of Abnormal Psychology, 106, 404-415.

Hull, E.M., Muschamp, J .W., & Sato, S. (2004). Dopamine and serotonin: Influences on
male sexual behavior. Physiology & Behavior, 83, 291-307.

James, A. & Taylor, E. (1990). Sex differences in the hyperkinetic syndrome of
childhood. Journal of Child Psychology and Psychiatry, 31, 43 7-446.

Janowsky, J.S., Chavez, B., Zamboni, B.D., & Orwoll, E. (1998). The cognitive
neuropsychology of sex hormones in men and women. Developmental
Neuropsychology, 14(2/3), 421-440.

Jennings, P.J., Janowsky, J .S., & Orwoll, E. (1998). Estrogen and sequential movement.
Behavioral Neuroscience, 112(1), 154-159.

Jentsch, J .D. & Taylor, J .R. (2003). Sex-related differences in spatial divided attention
and motor impulsivity in rats. Behavioral Neuroscience, 11 7(1), 76-83.

John, O.P., Caspi, A., Robins, R.W., Moffitt, T.E., & Stouthamer-Loeber, M. (1994).
The Little-Five: Exploring the nomological network of the five-factor model of
personality in adolescent boys. Child Development, 65, 160-178.

Kang, J.H., Weuve, J ., & Grodstein. (2003). F. Postmenopausal hormone therapy and risk

of cognitive decline in community-dwelling aging women. Neurology, 63(1), 101-
107.

82

King, J.A., Barkley, R.A., Delville, Y., & Ferris, CF. (2000). Early androgen treatment
decreases cognitive fimction and catecholamine innervation in an animal model of
ADHD. Behavioural Brain Research, 107(1-2), 35-43.

King, J .A., Kelly, T.N., & Delville, Y. (2000). Adult levels of tesosterone alter
catecholamine innervation in an animal model of Attention-Deficit Hyperactivity
Disorder. Neuropsychobiology, 42(4), 163-166.

Kochanska, G., Tjebkes, T.L., & Forman, DR. (1998). Children’s emerging regulation of
conduct: Restraint, compliance, and internalization from infancy to the second
year. Child Development, 69(5), 1378-1389.

Kuntsi, J ., Stevenson, J., Oosterlaan, J ., & Sonuga-Barke, E.J.S. (2001). Test-retest
reliability of a new delay aversion task and executive function measures. British
Journal of Developmental Psychology, 19(3), 339-348.

Lahey, B.B., Pelham, W.E., Loney, J ., Lee, S.S., & Willcutt, E. (2005). Instability of the
DSM-IV subtypes of ADHD from preschool through elementary school. Archives
of General Psychiatry, 62, 896-902.

Leranth, C., Roth, RH, Elswoth, J.D., Nafiolin, F., Horvath, T.L., & Redmond, DE.
(2000). Estrogen is essential for maintaining nigrostriatal dopamine neurons in
primates: Implications for Parkinson’s disease and memory. The Journal of
Neuroscience, 20(23), 8604-8609.

Li, J. & Huang, Y. (2006). Early androgen treatment influences the pattern and amount of
locomotion activity differently and sexually differentially in an animal model of
ADHD. Behavioural Brain Research, 1 75, 176-182.

Logan, GD. (1994). A users guide to the stop signal paradigm. In D. Dagenbach and
TH. Carr (Eds), Inhibitory processes in attention, memory, and language, (pp.
189-239). San Diego: Academic Press.

Lutchmaya, S., Baron-Cohen, S., Raggett, P., Knickmeyer, R., & Manning, J .T. (2004).
2nd to 4th digit ratios, fetal testosterone and estradiol. Early Human Development,
77, 23-28.

Lyon, G. & Gadisseux, J. (1991). Structural abnormalities of the brain in developmental
disorders. In M. Rutter & P. Casaer (Eds), Biological Risk F actors for
Psychosocial Disorders. New York: Cambridge University Press.

Malas, M.A., Dogan, S., Evcil, E.I-l., & Desdicioglu, K. (2006). Fetal development of the
hand, digits and digit ratio (2D:4D). Early Human Development, 82, 469-476.

Maney, D. (2006). Hormones excite the senses. Psychological Science Agenda, 20(3).

83

Manning, J .T. (2002). Digit Ratio: A Pointer to Fertility, Behavior, and Health. New
Brunswick, NJ: Rutgers University Press.

Manning, J .T., F ink, B., Neave, N., & Caswell, N. (2005). Photocopies yield lower digit
ratios (2D:4D) than direct finger measurements. Archives of Sexual Behavior,
34(3), 329-333.

Martel, M.M., Gobrogge, K.L., Breedlove, S.M., & Nigg, J .T. (in press). Evidence for
organization effects of steroid hormones on childhood ADHD in boys but not
girls.

Martel, M.M., Lucia, V.C., Nigg, J.T., & Breslau, N. (2007). Sex differences in the
pathway from low birth weight to inattention/hyperactivity. Journal of A bnormal
Child Psychology, 35, 87-96.

Martel, M.M. & Nigg, J .T. (2006). Child ADHD and personality/temperament traits of
reactive and effortful control, resiliency, and emotionality. Journal of Child
Psychology and Psychiatry, 4 7(1 1), 1175-1183.

Martinussen, R., Hayden, J., Hogg-Johnson, S., & Tannock, R. (2005). A meta-analysis
of working memory in children with Attention-Deficit Hyperactivity Disorder.
The American Academy of Child and Adolescent Psychiatry, 44(4), 377-384.

McCabe, L.A., Cunnington, M., & Brooks-Gunn, J. (2004). The development of self-
regulation in children: Individual characteristics and environmental contexts. In
R.F. Baumeister & K.D. Vohs (Eds), Handbook of Self-Regulation: Research,
Theory, and Applications (pp. 340-356). New York: Guilford Press.

McCartney, K., Burchinal, M.R. & Bub, KL. (2006). Best Practices in Quantitative
Methods for Developmentalists. Monographs of the Society for Research in Child
Development, 71(3).

McFadden, D., Westhafer, J.G., Pasanen, E.G., Carlson, C., & Tucker, D.M. (2005).
Physiological evidence of hypermasculinization in boys with the inattentive type

of Attention-Deficit/Hyperactivity Disorder (ADHD). Clinical Neuroscience
Research, 5, 233-245.

McIntyre, M.H., Cohn, B.A., & Ellison, RT. (2006). Sex dimorphism in digital formulae
of children. American Journal of Physical Anthropology, 129, 143-150.

McManus, I.C. & Bryden, MP. (1991). Geschwind’s theory of cerebral lateralization:
Developing a formal, causal model. Psychological Bulletin, 110(2), 237-253.

Miller, G.W., Gainetdinov, R.R., & Caron, M.G. (2000). Involvement of dopamine
transporters in psychiatric disorders. In M. Myslobodsky & I. Weiner (Eds),

84

Contemporary Issues in Modeling Psychopathology (pp. 1- l 8). Boston: Kluwer
Academic.

Morris, J.A., Jordan, C.L., & Breedlove, SM. (2004). Sexual differentiation of the
vertebrate nervous system. Nature Neuroscience, 7(10), 1034-1039.

Murray, K.T. & Kochanska, GK. (2002). Effortful control: Factor structure and relation
to externalizing and internalizing behaviors. Journal of A bnormal Child
Psychology, 30(5), 503-514.

Muthen, L.K. & Muthen, B.O. (1998-2007). Mplus User’s Guide, Fourth Edition. Los
Angeles, CA: Muthen & Muthen.

Myers, R.E., Anderson, L.I., & Dluzen, DE. (2003). Estrogen but not testosterone,
attenuates metharnphetamine-evoked dopamine output from superfused striatal
tissue of female and male mice. Neuropharmocology, 44, 624-632.

Newcom, J.H., Halperin, J.M., Jensen, P.S., Abikoff, H.A., Arnold, E., Cantwell, D.P.,
Conners, C.K., Elliott, G.R, Epstein, J .N., Greenhill, L.L., Hechtrnen, L.,
Hinshaw, S.P., Hoza, B., Kraemer, H.C., Pelham, W.E., Severe, J.B., Swanson,
J.M., Wells, K.C., Wigal, T., & Vitiello, B. (2001). Symptom profiles in children
with ADHD: Effects of comorbidity and gender. Journal of the American
Academy of Child and Adolescent Psychiatry, 40(2), 137-146.

Nigg, J .T. (2000). On inhibition/disinhibition in developmental psychopathology: Views
from cognitive and personality psychology and a working inhibition taxonomy.
Psychological Bulletin, 126(2), 220-246.

Nigg, J .T. (2006). What causes ADHD? Understanding what goes wrong and why. New
York: The Guilford Press.

Nigg, J.T., Blaskey, L.G., Huang-Pollock, C.L., & Rappley, MD. (2002).
Neuropsychological executive functions and DSM-IV ADHD subtypes. Journal
of the American Academy of Child and Adolescent Psychiatry, 41(1), 59-66.

Nigg, J .T. & Casey, B.J. (2005). An integrative theory of attention-deficit/hyperactivity
disorder based on the cognitive and affective neurosciences. Development &
Psychopathology, 1 7, 785-806.

Nigg, J.T., Goldsmith, H.H., & Sachek, J. (2004). Temperament and Attention Deficit
Hyperactivity Disorder: The development of a multiple pathway model. Journal
of Clinical Child and Adolescent Psychology, 33(1), 42-53.

Okten, A., Kalyoncu, M., & Taris, N. (2002). The ratio of second- and fourth-digit

lengths and congenital adrenal hyperplasia due to 2l-hydroxylase deficiency.
Early Human Development, 70, 47-54.

85

Olson, S. (1996). Developmental perspectives. In S. Sandberg (Ed), Hyperactivity and
attention disorders of childhood (Second Ed, pp. 242-289). New York:
Cambridge University.

Ostlund, H., Keller, E., & Hurd, Y.L. (2003). Estrogen receptor gene expression in
relation to neuropsychiatric disorders. Annuals of the New York Academy of
Science, 1007, 54-63.

Ostrander, R., Weinfurt, K.P., Yarnold, P.R., & August, G.J. (1998). Diagnosing
Attention Deficit Disorders with the Behavioral Assessment System for Children
and the Child Behavior Checklist: Test and construct validity analyses using

optimal discriminant classification trees. Journal of Consulting and Clinical
Psychology, 66(4), 660-672.

Pandaranandaka, J ., Poonyachoti, S., & Kalandakanond-Thongsong, S. (2006).
Anxiolytic property of estrogen related to the changes of the monoarnine levels in
various brain regions of ovariectomized rats. Physiology & Behavior, 87, 828-
835.

Pennington, B.F. & Ozonoff, S. (1996). Executive functions and developmental
psychopathology. Journal of Child Psychology and Psychiatry, 37(1), 51-87.

Petersen, A.C., Richards, M., & Boxer, A. (1988). A self-report measure of pubertal
status: Reliability, validity, and initial norms. Journal of Youth and Adolescence,
17(2), 117-133.

Pickles, A. & Hill, J. (2006). Developmental Pathways. In D. Cicchetti & D.J. Cohen
(Eds) Developmental Psychopathology (Second Ed., Vol. 1, pp. 211-243). John
Wiley & Sons: Hoboken, NJ.

Power, T.J., Andrews, T.J., Eiraldi, R.B., Doherty, B.J., Ikeda, M.J., DuPaul, G.J., &
Landau, S. (1998). Evaluating Attention Deficit Hyperactivity Disorder using
multiple informants: The incremental utility of combining teacher with parent
reports. Psychological Assessment, 10(3), 250-160.

Preacher, K. J ., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect
effects in simple mediation models. Behavior Research Methods, Instruments, &
Computers, 36(4), 717-73 1.

Protopopescu, X., Pan, H., Altemus, M., Tuescher, 0., Polanecsky, M., McEwen, B.,
Silbersweig, D., & Stern, E. (2005). Orbitofrontal cortex activity related to
emotional processing changes across the menstrual cycle. PNAS, 102(44), 16060-
16065.

86

PsychCorp. (2005). Wechsler Individual Achievement Test—Second Edition: Examiner ’5
manual. Harcourt Assessment: San Antonio, TX.

Puig-Antich, J. & Ryan, N. (1986). Kiddie Schedule for A flective Disorders and
Schizophrenia. Pittsburgh, PA: Western Psychiatric Institute.

Reitan, R.M. & Wolfson, D. (1985). Category test and Trailmaking test as measures of
frontal lobe functions. Clinical Neuropsychologist, 9, 50-56.

Reynolds, C.R., & Kamphaus, R.W. (1992). Behavioral assesment system for children
manual. Circle Pines, MN: American Guidance Service.

Rhee, S.H., Waldman, I.D., Hay, D.A., & Levy, F. (2001). In F. Levy & D. Hay aids),
Attention, Genes, and ADHD. Philadelphia: Brumer-Routledge.

Rothbart, M.K. & Bates, J .E. (1998). Temperament. In W. Damon (Series Ed.) & N.
Eisenberg (Vol. Ed), Handbook of child psychology: Vol. 3. Social, emotional,
and personality development (5th Ed; pp. 105-176. New York: Wiley.

Rothbart, M.K. & Posner, M.I. (2006). Motivation, self-regulation, and self-organization.
In D. Cicchetti & D.J. Cohen (Eds), Developmental Psychopathology (Second
Ed, Vol. 2 pp. 465-501). John Wiley & Sons: Hoboken, NJ.

Rothbart, M.K. & Rueda, MR. (2005). The development of effortful control. In U. Mayr,
E. Awh, & S.W. Keele (Eds), Developing individuality in the human brain: A
tribute to Michael I. Posner (pp.l4l-l66). Washington, DC: APA.

Rucklidge, J.J. & Tannock, R. (2002). Neuropsychological profiles of adolescents with
ADHD: Effects of reading difficulties and gender. Journal of Child Psychology
and Psychiatry, 43(8), 988-1003.

Russell, V.A., Oades, R.D., Tannock, R., Killeen, P.R., Auerbach, J .G., Johansen, B.B.,
& Sagvolden, T. (2006). Response variability in Attention-Defict/Hyperactivity
Disorder: A neuronal and glial energetics hypothesis. Behavioral and Brain
Functions, (2), 30.

Rutter, M., Caspi, A., & Moffitt, TE. (2003). Using sex differences in psychopathology
to study causal mechanisms: Unifying issues and research strategies. Journal of
Child Psychology and Psychiatry, 44(8), 1092-1 1 15.

Sagvolden, T., Johansen, B.B., Aase, H., & Russell, VA. (2004). A dynamic
developmental theory of attention-deficit/hyperactivity disorder (ADHD)
predominantly hyperactive-impulsive and combined subtypes. Behavioral Brain
Science, 28, 397-419.

87

Sawada, H. & Shimohama, S. (2000). Neuroprotective effects of estradiol in
mesencephalic dopaminergic neurons. Neuroscience and Biobehavioral Reviews,
24, 143-147.

Schachar, R. Tannock, R., Marriott, M., & Logan, GD. (1995). Deficient inhibitory
control and attention deficit hyperactivity disorder. Journal of A bnormal Child
Psychology, 23, 411-437.

Seidman, L.J., Biederman, J ., F araone, S.V., Weber, W., Mennin, D., & Jones, J. (1997).
A pilot study of neuropsychological function in girls with ADHD. Journal of the
American Academy of Child and Adolescent Psychiatry, 3 6(3), 366-373.

Seidman, L.J., Biederman, J .., Monuteaux, M.C., Valera, E., Doyle, A.E., & Faraone,
S.V. (2005). Impact of gender and age on executive functioning: do girls and boys
with and without attention deficit hyperactivity disorder differ
neuropsychologically in preteen and teenage years? Developmental
Neuropsycholology, 27, 79- 1 05.

Seidman, L.J., Valera, E.M., & Makris, N. (2005). Structural brain imaging of Attention-
Deficit/Hyperactivity Disorder. Biological Psychiatry, 57(11), 1263-1272.

Shansky, R.M., Glavis-Bloom, C., Lerman, D., McRae, P., Benson, C., Miller, K.,
Cosand, L., Horvath, T.L., & Amsten, A.F.T. (2004). Estrogen mediates sex
differences in stress-induced prefrontal cortex dysfunction. Molecular Psychiatry,
9, 531-538.

Sharifi, K.L., Brady, D.L., & Koenig, M. (2004). Estrogen modulates RGS9 expression
in the nucleus accumbens. Neuroreport: For Rapid Communication of
Neuroscience Research, 15(15), 2433-2436.

Shaywitz, S.E., Nafiolin, F ., Zelterman, D., Marchione, K., Holahan, J ., Palter, S., &
Shaywitz, BA. (2003). Menopause, 10(5), 420-426.

Shively, C.A. & Bethea, CL. (2004). Cognition, mood disorders, and sex hormones.
ILAR Journal, 45(2), 189-199.

Sonuga-Barke, E.J.S. (1994). Annotation: On dysfunction and function in psychological
theories of childhood disorder. Journal of Child Psychology and Psychiatry, 35,
801-815.

Sonuga-Barke, E.J.S. (2005). Causal models of Attention-Deficit/Hyperactivity Disorder:
From common simple deficits to multiple developmental pathways. Biological
Psychiatry, 57(11), 1231-1238.

Spreen, O. & Strauss, E. (1991). A compendium of neuropsychological test:
Administration, norms, and scoring. New York: Oxford University Press.

88

Stevenson, J.C., Everson, P.M., Williams, D.C., Hipskind, G., Grimes, M., & Mahoney,
ER. (2007). Attention deficit/hyperactivity disorder (ADHD) symptoms and digit
ratios in a college sample. American Journal of Human Biology, 19, 41-50.

Thompson, TL. (1999). Attenuation of dopamine uptake in vivo following priming with
estradiol benzoate. Brain Research, 834, 164-167.

Thompson, T.L., Bridges, S.R., & Weirs, W.J. (2001). Alteration of dopamine transport
in the striatum and nucleus accumbens of ovariectomized and estrogen-primed

rats following N-(p-isothiocyanatophenethyl) spiperone (N IPS) treatment. Brain
Research Bulletin, 54(6), 631-638.

Trivers, R., Manning, J ., & Jacobson, A. (2006). A longitudinal study of digit ratio

(2D:4D) and other finger ratios in Jamaican children. Hormones and Behavior,
49, 150-156.

Tucker, D.M. & Williamson, RA. (1984). Asymmetric neural control systems in human
self-regulation. Psychological Review, 91(2), 185-215.

Valiente, C., Eisenberg, N., Smith, C.L., Reiser, M., F abes, R.A., Losoya, S., Guthrie,
I.K., & Murphy, BC. (2003). The relations of effortful control and reactive

control to children’s externalizing problems: A longitudinal assessment. Journal
of Personality, 71(6), 1171-1196.

Volkow, N.D., Wang, G., Fowler, J .S., & Ding, Y. (2005). Imaging the effects of

methylphenidate on brain dopamine: New model on its therapeutic actions for
attention-deficit/hyperactivity disorder. Biological Psychiatry, 5 7, 1410-1415.

Volkow, N.D., Wang, G., Fowler, J.S., Telang, F., Maynard, L., Logan, J., Gatley, S.J.,
Pappas, N., Wong, C., Vaska, P., Zhu, W., & Swanson, J .M. (2004). Evidence
that methylphenidate enhances the saliency of a mathematical task by increasing

dopamine in the human brain. American Journal of Psychiatry, 161(7), 1173-
1180.

Watson, D., Kotov, R., & Gamez, W. (2006). Basic dimensions of temperament in
relation to personality and psychopathology. In RF. Krueger & J.L. Tackett
(Eds), Personality and Psychopathology (pp. 7-38). New York: Guildford Press.

Wechsler, D. (2003a). Wechsler Intelligence Scale for Children-Fourth Edition:
Administration and scoring manual. San Antonio: The Psychological
Corporation.

Wechsler, D. (2003b). Wechsler Intelligence Scale for Children-Fourth Edition:
Technical and interpretive manual. San Antonio: The Psychological Corporation.

89

White, J .D. (1999). Personality, temperament, and ADHD: A review of the literature.
Personality and Individual Diflerences (2 7), 589-598.

Whittle, S., Allen, N.B., Lubman, D.I., & Yucel, M. (2006). The neurobiological basis of
temperament: Towards a better understanding of psychopathology. Neuroscience
and Biobehavioral Reviews, 30, 511-525.

Wilcox, RR. (2002). Understanding the practical advantages of modern ANOVA.
Journal of Clinical Child & Adolescent Psychology, 31: 399-412.

Willcutt, E.G., Doyle, A.B., Nigg, J.T., Faraone, S.V., & Pennington, B. (2005). Validity
of the executive function theory of Attention-Deficit/Hyperactivity Disorder: A
meta-analytic review. Biological Psychiatry, 57, 1336-1346.

Williams, J .H.G., Greenhalgh, K.D., & Manning, J.T. (2003). Second to fourth finger
ratio and possible precursors of developmental psychopathology in preschool
children. Early Human Development, 72(1), 57-65.

Willis, W.G. & Welier, MD. (2005). Neural substrates of childhood Attention-
Deficit/Hyperactivity Disorder: Electroencephalographic and magnetic resonance
imaging evidence. Developmental Neuropsychology, 27(1), 135-182.

Yonkers, R.A., Halbreich, U., Freeman, E., Brown, C., Endicott, J., Frank, E., Parry, B.,
Pearlstein, T., Severino, S., Stout, A., Stone, A., & Harrison, W. (1997).
Symptomatic improvement of Premenstrual Dysphoric Disorder with Sertraline
treatment: A randomized, controlled trial. Journal of the American Medical
Association, 278(12), 983-988.

Zahn-Waxler, C., Schmitz, S., F ulker, D., Robinson, J ., & Emde, R. (1996). Behavior
problems in 5-year-old monozygotic and dizygotic twins: Genetic and
environmental influences, patterns of regulation, and internalization of control.
Development and Psychopathology, 8, 103-122.

Zhou, W., Cunningham, K.A., & Thomas, ML. (2003). Estrogen effects on the
hyperactivity induced by (+)-MDMA and cocaine in female rats. Behavioral
Neuroscience, 11 7(1), 84-94.

Zuckerman, M. (2005). Psychobiology of Personality (2"d Ed). Cambridge, NY:
Cambridge University Press.

90