Cognitive Control and Decision Making
The studies described in this thesis were performed at the Rudolf Magnus Institute of Neuroscience,
Department of Child and Adolescent Psychiatry, University Medical Center Utrecht, the Netherlands
and at the Sackler Institute of Developmental Psychobiology, Weill Medical College of Cornell University,
New York, United States of America.
Research in this thesis was supported by grants from the UMC Utrecht International Office, and the
Netherlands Organisation for Scientific Research (NWO - Veni/Vidi to Dr. S. Durston):
Cover design: Martijn Mulder & Sven Willemse
Copyright © 2010 by Martijn Mulder
All rights reserved. No part of this book may be reproduced or utilised in any form or by any means,
electronic or mechanical, including photocopying and recording by any information storage and
retrieval system, without written permission of the author.
Cognitive Control and Decision Making in ADHD
cognitieve controle en beslisprocessen in ADHD
(met een samenvatting in het Nederlands)
ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag
van de rector magnificus, prof. dr. J.C. Stoof, ingevolge het besluit van het
college voor promoties in het openbaar te verdedigen op donderdag 3 juni
2010 des ochtends te 10.30 uur.
Martinus Johannes Mulder
geboren op 18 april 1974 te Ossendrecht
Prof. dr. H. van Engeland
Dr. S. Durston
voor Pa en Ma
Activation in ventral prefrontal cortex is sensitive to genetic vulnerability
for Attention-Deficit Hyperactivity Disorder.
Familial vulnerability to ADHD affects activity in the cerebellum in
addition to the prefrontal Systems
Functional connectivity in cognitive control networks is
sensitive to familial risk for ADHD
Basic impairments in regulating the speed-accuracy tradeoff
predict symptoms in ADHD
BOLD correlates of reward-related decision bias on a visual
Dankwoord / Acknowledgements
Attention-deficit hyperactivity disorder (ADHD) has a heterogeneous phenotype. The disorder is
defined by inappropriate, impulsive or inattentive behavior, often within social contexts. For example,
subjects with ADHD may have trouble staying in their seat in the classroom, or may blurt out
comments while at the movies, or may have difficulty waiting their turn in a conversation. In addition,
subjects with ADHD often are distracted easily, making it hard to stay focused on a task at hand.
However, to some extent impulsive and inattentive behavior is a normal part of childhood, making
it sometimes difficult to distinguish between symptoms of the disorder and typical development. In
other words: how impulsive or how distractible does a child have to be in order to receive a diagnosis
of ADHD? Criteria have been drawn up to define the disorder (see Table 1; American Psychiatric
Association, 1994 & ICD-10). In addition to classifying symptoms, they provide a measure of the
severity of impairment: a child can only be diagnosed when a minimum level of the impairment is met.
Such criteria are useful in the clinic, as they objectify the disorder. However, they also emphasize the
phenotypic heterogeneity of ADHD. Such heterogeneity makes it hard to identify the neurobiological
mechanisms underlying the disorder. Nevertheless, research has shown that ADHD is heritable and
that genetic factors account for a substantial part of the phenotypic variance. Studies have suggested
that different biological pathways may underlie the disorder, where, e.g., different genotypes and
environmental factors result in similar phenotypes that are covered by the ADHD diagnosis.
Prevalence & Treatment
ADHD is a neurodevelopmental disorder that is characterized by disorganized, inattentive behavior
as well as inappropriate levels of impulsivity and hyperactivity (American Psychiatric Association,
2000; Kaplan, et al., 1994). It is the most commonly diagnosed psychiatric disorder in childhood
(Brown, et al., 2001; Faraone, et al., 2003; Levy, et al., 1997), with a worldwide prevalence of 5.3%
(Polanczyk, et al., 2007). The age of onset is usually around age of 6 or 7 years (American Psychiatric
Association, 1994; WHO, 1992) and the disorder can persist into adulthood (Asherson, 2009; Wilens,
et al., 2002).
Treatment for the disorder includes pharmacotherapy and non-pharmacological therapy such as
education remediation and psychotherapy. Stimulants, such as methylphenidate and amphetamine
are effective, but have sometimes been questioned due to putative adverse effects. Alternative
therapies, such as food supplementation with poly-unsaturated fatty acids are gaining popularity,
but to date, the effects are not convincing enough to replace psycho or drug therapy (Dopheide &
ADHD is a heritable psychiatric disorder (Dopheide & Pliszka, 2009). The disorder has a tendency to
cluster in families and up to 75-80% of the variance in ADHD symptoms can be explained by genetic
effects (Albayrak, et al., 2008; Faraone, et al., 2005; Rasmussen, et al., 2004; Thapar, et al., 1999).
Furthermore, twin and adoption studies suggest a substantial genetic component for ADHD, with
50-80% concordance for monozygotic twins and 30% for dizygotic twins (Bradley & Golden, 2001;
Hechtman, 1994; Thapar, et al., 1999).
Genetic studies of ADHD have been conducted. Two popular methods are linkage and association
studies. In linkage studies, the whole genome is scanned for each subject, to look for genetic markers
that all subjects share. It is then suggested that the shared behavior (or disorder) across subjects
is linked to the area around that marker. However, as these markers often represent large parts
of a chromosome, it is hard to pinpoint the exact genes that are associated with the disorder. In
association studies, two discordant groups are compared to each other, to investigate whether a
certain genotype (or allele) is present more often in the affected than in the unaffected group. Often,
candidate genes are selected on theoretical grounds and used in association studies to investigate
if they are involved in the disorder.
Several candidate genes have been implicated in ADHD. These include genes related to dopamine
and serotonin systems (Faraone, et al., 2005). Both neurotransmitter systems have been implicated
in ADHD in a variety of studies (for review see Durston, et al., 2009). For example, it has been shown
that a hypodopaminergic state of the striatum is related to ADHD symptoms. Stimulant medication
normalizes these decreased dopamine levels by blocking dopamine transporter (DAT) receptors
(Shafritz, et al., 2004; Spencer, et al., 2007; Vaidya, et al., 1998). This has been confirmed in animal
studies using DAT1 knock-out mice, where the initial hyperactive behavior improved when stimulant
Table 1. Diagnostic Criteria for ADHD, according to the DSM-IV (APA, 1994)
A. Either 1 or 2
1. Six (or more) of the following symptoms of inattention have persisted for at least 6 months to a degree that
is maladaptive and inconsistent with developmental level:
a. Often fails to give close attention to details or makes careless mistakes in schoolwork, work, or other
b. Often has difficulty sustaining attention in tasks or play activities
c. Often does not seem to listen when spoken to directly
d. Often does not follow through on instructions and fails to finish schoolwork, chores, or duties in the
workplace (not due to oppositional behavior or failure to understand instructions)
e. Often has difficulty organizing tasks and activities
f. Often avoids, dislikes, or is reluctant to engage in tasks that require sustained mental effort (such as
schoolwork or homework)
g. Often loses things necessary for tasks or activities (e.g. toys, school assignments, pencils, books, or
h. Is often easily distracted by extraneous stimuli
i. Is often forgetful in daily activities
2. Six (or more) of the following symptoms of hyperactivity-impulsivity have persisted for at least 6 months to
a degree that is maladaptive and inconsistent with developmental level:
a. Often fidgets with hands or feet or squirms in seat
b. Often leaves seat in classroom or in other situations in which remaining seated is expected
c. Often runs about or climbs excessively in situations in which it is inappropriate (in adolescents or adults,
may be limited to subjective feelings of restlessness)
d. Often has difficulty playing or engaging in leisure activities quietly
e. Is often “on the go” or often acts as if “driven by a motor”
f. Often talks excessively
g. Often blurts out answers before questions have been completed
h. Often has difficulty awaiting turn
i. Often interrupts or intrudes on others (e.g. butts into conversations or games)
B. Some hyperactive-impulsive or inattentive symptoms that caused impairment were present before 7 years of age.
C. Some impairment from the symptoms is present in 2 or more settings (e.g. at school [or work] or at home).
D. There must be clear evidence of clinically significant impairment in social, academic, or occupational functioning.
E. The symptoms do not occur exclusively during the course of a pervasive developmental disorder, schizophrenia, or
other psychotic disorder and are not better accounted for by another mental disorder (e.g. mood disorder, anxiety
disorder, dissociative disorder, or personality disorder).
Code based on type:
314.01 Attention-Deficit/Hyperactivity Disorder, Combined Type: if both criteria A1 and A2 are met for the past 6
314.00 Attention-Deficit/Hyperactivity Disorder, Predominantly Inattentive Type: if criterion A1 is met but criterion
A2 is not met for the past 6 months
314.01 Attention-Deficit/Hyperactivity Disorder, Predominantly Hyperactive, Impulsive Type: if criterion A2 is
met but criterion A1 is not met for the past 6 months
314.9 Attention-Deficit/Hyperactivity Disorder Not Otherwise Specified
medication was administered. Interestingly, this effect was mediated by serotonin systems, showing
the involvement of both dopamine and serotonin systems in these improvements (Gainetdinov, et
Genes that have most often been associated with ADHD are the dopamine D4 receptor gene
(DRD4) and dopamine transporter gene (DAT1) (see for review Durston, et al., 2009). Furthermore,
across studies that used the less detailed method of linkage mapping, the link with region 5p13 was
replicated in two independent samples; the region that includes the candidate DAT1 gene (Friedel,
et al., 2007; Hebebrand, et al., 2006). A second region that is replicated is 17p11, which lies close to
17q11, a region that covers the serotonine transporter gene (Ogdie, et al., 2003).
Despite these convincing results, findings of association and linkage studies are not consistent,
suggesting that ADHD may be related to combinations of genes, each of which have a limited effect
(Faraone, et al., 2005). Furthermore, even for the best-replicated candidate genes (DAT1 and DRD4)
nearly as many negative as positive associations have been found (Durston, et al., 2009).
It has been suggested that gene x environment interactions may explain some of the inconsistent
findings in genetic studies of ADHD (Plomp, et al., 2009; Rutter & Silberg, 2002). For example, an
increased risk for ADHD has been reported for carriers of the polymorphism of DAT1 that is most
often associated with ADHD, but only when they were exposed to psychosocial adversity (Laucht,
et al., 2007), maternal smoking during pregnancy (Becker, et al., 2008; Kahn, et al., 2003; Todd &
Neuman, 2007) or maternal use of alcohol during pregnancy (Brookes, et al., 2006). Similar findings
have been reported for carriers of the DRD4 polymorphism (Neuman, et al., 2007; Todd & Neuman,
In sum, although etiology data suggest an underlying heritable biological basis for ADHD, only a
few genes have been confirmed repeatedly, both with and without an interaction with environmental
factors. A possible explanation for these inconsistent findings is that the heterogeneous symptoms in
ADHD do not map onto the neurobiological effects of gene variations. The use of endophenotypes
may help to address this problem.
Diagnostic classification criteria are often extensive and diverse. As such, investigating the
neurobiology of the disorder at this level is extremely difficult since links between the behavioral
data and the underlying biological mechanisms are diverse and noisy (Gottesman & Gould, 2003).
An endophenotype is a quantifiable component between biology and behavior that reduces the
heterogeneity of psychiatric symptoms (Gottesman & Gould, 2003; Gould & Gottesman, 2006).
Measures of such quantity allows to identify biological pathways that are less susceptible for noisy
variations of the phenotype (see Figure 1). For example, with functional magnetic resonance
imaging (fMRI) one can measure brain activity during tasks that probe disabilities associated with
Such a measure will likely be closer to genetic influences, producing a more homogeneous measure
for investigating causative agents. As such, it may be useful to identify measures that can serve as
endophenotype in investigating the neurobiology of ADHD. There are a number of criteria that a
measure should meet in order to serve as an endophenotype (see Table 2; for review see Durston, et
al., 2009). One such criterion is that the endophenotype should also be found in non-affected family
members, at a higher rate than in the general population.
Figure 1. Links between the disorder and the underlying biological
mechanisms are diverse and noisy (A). Use of an endophenotype
reduces the heterogeneity in the phenotype of interest allowing research
to identify biological pathways that are shorter and thus less susceptible
to variations in the phenotype (B).
Cognitive control in ADHD
A sensible starting point to elucidate gene-behavior pathways is to reduce the phenotype to
measurable units. This is necessary to link neurobiological mechanisms to the heterogeneous
symptoms of ADHD. A measure that has been used successfully for this purpose is cognitive control.
Cognitive control is the ability to control behavior in response to contextual and temporal cues
and adjust behavior accordingly (Nigg & Casey, 2005). According to this model, the brain makes
predictions about what is going to happen and when it is going to happen. When these predictions
are violated, behavior needs to be adjusted in order to act toward a specific goal (see Figure 2). It
has been suggested that two important brain circuits are involved in regulating these processes:
The fronto-striatal circuit, involved in the detection of novel events (what events; Berns, et al., 1997;
Schultz, et al., 1997), and the fronto-cerebellar circuit, involved in monitoring and detecting violations
in the timing of events (when events; Ivry, et al., 2002; R. M. Spencer, et al., 2003). Both circuits are
believed to play an important role in the development of the ability to adjust behavior in response to
contextual cues (Casey, 2005).
It has been suggested that impairments in cognitive control may explain a substantial part of the
ADHD phenotype (Barkley, 1997; Durston, 2003; Durston, et al., 2009; Nigg & Casey, 2005; Sergeant,
et al., 2002; Willcutt, et al., 2005). Subjects with ADHD often perform worse than typically developing
children on tasks that probe cognitive control. Furthermore, neuroimaging studies have shown
deficits in brain functioning underlying these behavioral differences. Specifically, during tasks where
subjects are required to inhibit a prepotent response, the ventro-lateral prefrontal cortex, anterior
cingulate cortex and striatum showed decreased activity for subjects with ADHD compared to control
subjects (for review see Bush, et al., 2005; Castellanos & Tannock, 2002; Dickstein, et al., 2006;
Durston, et al., 2009). Furthermore, structural neuroimaging in ADHD has shown volume reductions
Figure 2. Schematic representation of aspects of cognitive control and their relationship to ADHD. The brain makes
predictions about what is going to happen (fronto-striatal system) and when it is going to happen (fronto-cereblellar
system). When these predictions are violated, behavior needs to be adjusted in order to act toward a specific goal
(prefrontal cortex). Impairments in these systems can lead to impulsive and dysregulated behavior as seen in the ADHD
phenotype. Modified with permission from Nigg & Casey, 2005.
“impulsive and dysregulated
“violations of predictions... ? ”
“Adjust behavior ! ”
15 Download full-text
in prefrontal cortex (gray matter), striatum and cerebellum (for review see Durston, 2003; for meta
analyses Valera, et al., 2007). As these regions are involved in cognitive control (Aron & Poldrack,
2005; Botvinick, et al., 2004; Nigg & Casey, 2005), it is not surprising that impairments in activity of
these regions are associated with ADHD.
These findings suggest that neuroimaging measures of cognitive control could be a candidate
endophenotype, as they meet a number of the criteria (see Table 2): behavioral and neuroimaging
measures of cognitive control are continuous quantities (criterion 1); deficits are stable and well
established in the ADHD phenotype (criteria 2 & 4) (Lijffijt, et al., 2005; Nigg, et al., 2005); behavioral
changes in cognitive control are heritable (criteria 6 & 7) (Rasmussen, et al., 2004; Slaats-Willemse,
et al., 2003) and deficits in cognitive control are grounded in neuroscience (criterion 8) (for review
see Durston, et al., 2009).
Perceptual Decision Making
As noted, theoretical accounts have often linked the ADHD phenotype to deficits in cognitive control,
where children with ADHD have problems responding to contextual and temporal cues and adjusting
their behavior accordingly. Tasks that tap cognitive control are often designed to manipulate various
aspects of cognition, but at the trial level they have often one thing in common: they require the
subject to make a perceptual decision before choosing the most appropriate course of action. It is an
open question whether these basic processes are also affected by the disorder.
Mathematical models have been developed to describe the process of perceptual decision making.
One class of these models is the drift-diffusion models (DDM; see for review Bogacz, 2007; Gold &
Shadlen, 2007). These models describe the decision process as an accumulation of noisy sensory
information towards a decision threshold. For example, imagine a task that requires a subject to
choose whether the perceived motion of a ‘cloud’ of moving dots is directed to the right or to the
left. Here, the sensory information is determined by the number of dots moving coherently towards
Table 2. Criteria for endophenotypes for investigating gene effects in psychiatry
Criteria for intermediate phenotypes in psychiatric research
(1) Continuously quantifiable
(2) Stable (a trait as opposed to a state measure)
(3) Closer to the causative agent (e.g., genes and gene expression) than the disorder
(4) Associated with disorder
(5) Probabilistically predictive of the disorder
(6) Cluster in families where the disorder is found
(7) Found in unaffected relatives of affected individuals
(8) Grounded in neuroscience
(adapted from Almasy & Blangero, 2001; Castellanos & Tannock, 2002; Durston, et al., 2009; Gottesman
& Gould, 2003)