CNTRICS Final Task Selection: Executive Control

Schizophrenia Bulletin vol. 35 no. 1 pp.
115135, 2009
doi:10.1093/schbul/sbn154
Advance Access publication on November 14, 2008
CNTRICS Final Task Selection: Executive Control
Deanna M. Barch1,2,4,8, Todd S. Braver2, Cameron

S. Carter5, Russell A. Poldrack6, and Trevor W. Robbins7
2
Department of Psychology; 3Department of Psychiatry and;

Department of Radiology, Washington University, MO;
5
Departments of Psychiatry and Psychology, University of
California, Davis, CA; 6Department of Psychology, University of
California, Los Angeles, CA; 7Department of Experimental
Psychology, University of Cambridge, Cambridge England.
4
Co-authors had equivalent input and are listed in alphabetical

order.
The third meeting of the Cognitive Neuroscience Treatment Research to Improve Cognition in Schizophrenia
(CNTRICS) was focused on selecting promising measures
for each of the cognitive constructs selected in the first
CNTRICS meeting. In the domain of executive control,
the 2 constructs of interest were rule generation and selection and dynamic adjustments in control. CNTRICS
received 4 task nominations for each of these constructs,
and the breakout group for executive control evaluated
the degree to which each of these tasks met prespecified criteria. For rule generation and selection, the breakout group
for executive control recommended the intradimensional/
extradimensional shift task and the switching Stroop for
translation for use in clinical trial contexts in schizophrenia
research. For dynamic adjustments in control, the breakout
group recommended conflict and error adaptation in the
Stroop and the stop signal task for translation for use in
clinical trials. This article describes the ways in which
each of these tasks met the criteria used by the breakout
group to recommend tasks for further development.
Key words: cognition/clinical trials/schizophrenia
Deficits in executive control have long been thought to be
one of the hallmark cognitive characteristics of schizophrenia. Executive control processes are those mechanisms that
allow individuals to flexibly and dynamically adjust their
performance in response to changing environmental
demands and changing internal goal states. Numerous empirical studies have shown that individuals with schizo1
To whom correspondence should be addressed; Department of

Psychology, Washington University, St Louis, MO 63130; tel: 314935-8729, fax: 314-935-8790, e-mail: [email protected].
The Author 2008. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved.
For permissions, please email: [email protected].
115
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012
phrenia show deficits on tasks designed to measure such

executive control processes.1 Deficits on such executive
control tasks are associated with both negative and disorganized symptoms in schizophrenia and those at risk for
schizoprenia,24 and deficits in executive control are associated with poor functional outcome.5
Given the prominence of executive control deficits in
schizophrenia, it is not surprising that this was one of
the cognitive domains upon which both the MATRICS
and Cognitive Neuroscience Treatment Research to Improve Cognition in Schizophrenia (CNTRICS) initiatives
have focused. As described by Kerns et al,6 a great deal of
basic research in cognitive neuroscience has focused on
elucidating the cognitive and neurobiological mechanisms
that provide humans with the ability to flexibly shift and
adapt their behavior over time and over different internal
and external contingencies. As part of the first CNTRICS
meeting, it was suggested that 2 components of executive
control were ready for immediate translation for development and use in clinical trials in schizophrenia: (1) rule
generation and selection and (2) dynamic adjustments
in control. Rule generation and selection was defined as
the processes involved in activating task-related goals or
rules based on endogenous or exogenous cues, actively
representing them in a highly accessible form, and maintaining this information over an interval during which
that information is needed to bias and constrain attention
and response selection.7
Dynamic adjustments in control were defined as the
processes involved in detecting the occurrence of conflict
or errors in ongoing processing, identifying the type of
control adjustments needed, and recruiting additional
control processes.7
The goal of the third CNTRICS meeting was to solicit
and evaluate nominations for promising tasks that would
measure each of the constructs selected in the first
CNTRICS meeting (see Carter and Barch8 for a full
list of all the constructs), including the 2 constructs
selected under executive control. Nominations were solicited from both basic and clinical scientists, from both
academia and industry. The nominators were asked to
provide a description of the task and to provide information on 6 domains relevant to selecting the most promising tasks: (1) cognitive construct validity; (2) neural
construct validity; (3) pharmacological sensitivity; (4) reliability; (5) other psychometric characteristics; and (6)
D. M. Barch et al.
Rule Generation and Selection

CNTRICS received 4 initial nominations for the rule generation and selection construct: (1) the 1-2 AX-CPT; (2)
the Groton Maze learning test; (3) CANTAB intradimensional/extradimensional (ID/ED) task; and (4) the
switching Stroop. The executive control breakout group
was tasked with the need to evaluate these nominations.
Following animated discussion, the group decided that
2 of these tasks (CANTAB ID/ED and the switching
Stroop) should be nominated for further development.
Table 1 provides a brief overview of these 4 tasks and
a very brief summary of their pros and cons in regards
to the selection criteria.
Tasks not Selected as Recommended for Development
In the 1-2 AX-CPT,14 stimuli are presented one at a time
in a sequence and subjects must press one button for a target and another for a nontarget. The stimuli are letters,
and subjects are told to look for specific target sequences
(eg, A followed by an X). Interleaved with the letters are
the numbers 1 or 2. These numbers instruct the subject to focus on one or another target sequence. For
example, if the most recent number was a 1, the target
sequence would be an A followed by an X, but if it
had been a 2, the target sequence would be a B followed
by a Y. This task is designed to test the subjects ability to
generate or select the appropriate rule for responding
based on the previously presented number. Performance
on this task has been modeled in a neural network model
of frontal-basal ganglia interactions and their contribution to working memory.14 The breakout group felt
116
that the 1-2 AX-CPT was an interesting and promising

task but that it needed more research at both the basic
and clinical levels before it was ready for translation.
Although there is encouraging pilot imaging work with
this task, and pilot work in patients with schizophrenia,
neither of these have been published yet.
The Groton Maze learning test1517 is one in which
subjects must learn a series of hidden mazes based on
feedback. Although the breakout group also felt that
this was an interesting task, the group felt that it was
not clear that it was a specific measure of the construct
of interest (rule generation and selection) as poor performance on the Groton Maze learning test could reflect
a number of different factors in addition to rule generation and selection, including problems learning from
feedback or spatial processing difficulties.
Tasks Recommended for Immediate Translation for Rule
Generation and Selection CANTAB ID/ED Task
Description
The CANTAB intra/extradimensional set shift is implemented on a computer-controlled touch screen and takes
approximately 7 min to administer, depending on the
level of impairment of the individual performing the
task (see also www.camcog.com for description and
screen shot figure). Two artificial visual perceptual
dimensions are used in the test: (1) color-filled shapes
and (2) white lines. The simple stimuli in the task are
made up of just one of these perceptual dimensions,
whereas compound stimuli are made up of both, namely
white lines overlying color-filled shapes (see stimuli on
www.camcog.com Web site). The subject starts by seeing
2 simple color-filled shapes (presented in 2 of 4 locations
on the screen, randomly on each trial) and must learn
which one is correct by touching it (simple discrimination
stage: SD). Subjects progress through the test by satisfying a set criterion of learning at each stage (6 consecutive
correct responses). If at any stage the subject fails to
reach this criterion after 50 trials, the test terminates.
Computer-provided feedback teaches the subject which
stimulus is correct, and after 6 correct responses for
the initial simple discrimination stage, there is a reversal
(simple discrimination reversal stage: SDR). Following
acquisition of this reversal stage (which teaches the subject that different stimuli within a perceptual dimension
are relevant to the rule), the subject is asked to complete
compound discriminations in which each stimulus now
has 2 potentially relevant perceptual features (lines and
color-filled shapes), initially with spatially separated
(C-D stage) and then superimposed (CD stage) stimuli.
The subject then must perform a reversal on the compound discrimination stimuli (compound discrimination
reversal: CDR stage) before the stimulus exemplars
change for an intradimensional shift. In the intradimensional shift stage, the same dimension of the stimulus
the availability of animal models. The overview article at

the beginning of this special section outlines why and how
these specific criteria were selected and discusses the ways
in which these criteria were used to select among tasks. In
addition, the overview article describes additional criteria
that could be used to adjudicate between tasks that
appeared similar on the properties described above,
such as evidence for impairment in schizophrenia. However, importantly, evidence for impairment in schizophrenia was not considered a requisite feature so as to facilitate
the inclusion of potentially promising paradigms that have
yet to be studied in schizophrenia. The goal of the current
article is to provide a brief summary of the tasks that were
nominated, the reasons as to why some tasks were not selected as recommended for immediate translation and to
provide an overview of the data used to select the tasks.
This overview is not meant to be an exhaustive review
of the literature on each task, but rather to give the reader
a starting point for understanding what each task is, how it
is administered, why it was chosen, what the evidence is for
construct and neural validity, what challenges the task
faces in the translation process, and pointers to the primary
literature for further consideration.
CNTRICS Final Task Selection
Table 1. Description of Tasks Nominated for Rule Generation and Selection Construct Definition: The Processes Involved in Activating
Task-Related Goals or Rules Based on Endogenous or Exogenous Cues, Actively Representing Them in a Highly Accessible Form, and
Maintaining This Information Over an Interval During Which That Information Is Needed to Bias and Constrain Attention and Response
Selection7
Tasks Recommended For Translation
CANTAB ID/ED task
Construct validity
Supported by both human and animal research
Designed to decompose the WCST
Neural systems
Evidence for role of ventro- and dorslolateral prefrontal cortex in extradimensional shifting (rule selection) from human and animal
research
Evidence for role of dopamine and norepinephrine in extradimensional shifting (rule selection) from human and animal research
Animal models
Animal models available for rhesus monkey, marmoset, and rat
Performance in schizophrenia
Consistent evidence for impairments in extradimensional shifting
Mixed evidence for impairments in intradimensional shifting
Psychometric data
Low test-retest reliability
Practice effects influenced by insight and learning on task over time
Potential confounds of generalized deficits with differences in discriminating power across conditions
Future directions
Modifications needed to improve test-retest reliability and to reduce practice effects
Evaluation of discriminating power across conditions is needed. If not equal, modifications will be needed to improve ability to
detect differential deficits
Switching stroop
Construct validity
Supported by computational modeling and individual differences research
Is a good measure of rule selection, but not of rule generation
Neural systems
Evidence from functional neuroimaging on the importance of DLPFC and ACC
Evidence from lesions studies for importance of DLPFC
Pharmacological sensitivity
Evidence that dopamine augmentation can improve Stroop performance in schizophrenia, but may not be specific to rule selection
Animal models
There are rat and primate models of paradigms described as measuring similar constructs
The degree to which the animal models (especially the rat models) are homologous to the human paradigm is an open question
Evidence for deficits in schizophrenia in the ability to select and apply the appropriate cue on the switching Stroop
Evidence for reduced DLPFC and ACC activity in schizophrenia during the performance of incongruent Stroop trials
Psychometric data
Little psychometric data available on test-rest reliability, practice effects, or floor/ceiling effects
Future directions
Research needed on assessing and improving the psychometric characteristics of the switching Stroop
Research needed on ways to make the use of the Switch Stroop feasible in clinical trials contexts (eg, task length, standardized
administration, enhanced sensitivity to detect deficits, etc.)
Evaluation of the homology of putative animal models and development of better animal models if needed
Other nominated tasks
1-2 AX-CPT
Construct validity
Task modeled in a computational model of prefrontal-basal ganglia contributions to working memory
Neural systems
No published data
117
Numerous studies in humans and animals showing that ID/ED performance is sensitive to a variety of pharmacological
manipulations
Evidence that modafinal can improve ID/ED performance in schizophrenia
D. M. Barch et al.
Table 1. Continued
remains relevant (eg, lines or shapes), but the specific feature within the dimension that is the target changes (eg,
which colored shape). This stage is designed to reinforce
perceptual dimension learning and to provide a control
for the crucial extradimensional shift stage. For the initial
intradimensional shift (IDS stage), color-filled shapes remain the only relevant dimension, then later after intradimensional shift reversal (IDR stage), again there are
new stimuli/exemplars for the extradimensional shift
(EDS state) in which the white lines become the only relevant dimension. This is followed by the extradimensional reversal stage (EDR), in which subjects must shift
between dimensions. There are thus 9 stages in all, designated SD, SDR, C-D, CD, CDR, IDS, IDR, EDS, and
EDR. The task has 2 modes: a clinical mode for testing
only once and 7 parallel modes for repeated testing. The
task has 18 outcome measures, assessing errors, and numbers of trials and stages completed.
From extensive experience with the task, the developers
have a number of recommendations for optimal task use.
First, although it is feasible to begin the task with lines and
shift to shapes, it is recommended that the initial learning
be fixed at shape. Second, they recommend, time permitting, that the subjects are screened with a short instructional test (Big circle/little circle discrimination) which
tests the subjects ability to understand what a rule is
and to be able to reverse it on command. Third, there is
a fixed rubric for the verbal instructions for this test which
are minimal but which should be followed closely.
Construct Validity
The theoretical validity of the test was established over 30
years ago and stems from early animal and human learn118
ing theory on discrimination learning, reversal, and dimensional shifting. More recent computational
approaches have confirmed for example that reversal
learning and extradimensional shifting necessarily engage
different processes and have also considered the paradigm in the context of a theory of frontal lobe function
based on working memory.18 The CANTAB test was
designed partly to decompose the well-known Wisconsin
Card Sort test (WCST) into its constituent parts including a critical extradimensional shift stage that is probably
equivalent to a category shift on the WCST. The initial
validation study demonstrating cross-species advantage
for IDS over EDS performance in humans and marmosets was by Roberts et al.19 There is an extensive and
growing bibliography of the ID/ED task, and the reader
is referred to www.camcog.com for references and
abstracts. The following constitutes a brief overview.
Neural Systems
The neural substrates of the test (or similar variants of it,
with changes in the stimuli but not the basic principles of
test design) have been defined on the basis of human neuropsychological data in brain damaged patients, functional imaging data (both functional magnetic
resonance imaging [fMRI] and positron emission tomography),20,21 and animal (monkey, rat, mouse) lesion studies. There is converging evidence that reversal learning
implicates orbitofrontal cortex function in humans, monkeys, and rats.20,2227 There is also additional evidence for
ventromedial striatal and posterior parietal involvement.20,21,28 There is considerable evidence for a role for
the primate ventrolateral prefrontal cortex in extradimensional shift learning20,22 as well as other posterior cortical
No published data
Animal models
No published data
No published data
Psychometric data
No published data
Groton maze learning task
Construct validity
Not clear that the task selectively measures rule selection or generation as performance may be strongly influenced by the ability
to learn from feedback and visual spatial processing abilities
Neural systems
Evidence that performance can elicit feedback-related error negatives in ERP studies9
Evidence that speed and efficiency of performance can be improved10
Animal models
Cincinnati Water Maze test in rodents may be an animal homologue11
Evidence for excessive rule-breaking errors in schizophrenia12
Psychometric data
Preliminary evidence of good psychometric characteristics13
(posterior parietal, temporal) areas. The neuroanatomical

dissociation between EDS and reversal learning in marmoset also holds for the rat.29 The precise substrates of discrimination learning and intradimensional shift learning are less
clear, but see Rogers et al21 for some initial observations.
Pharmacological and Behavioral Manipulation of Task
Performance
Performance in Schizophrenia
There have been over 20 studies so far listed in the CANTAB bibliography on the ID/ED test in schizophrenia,
beginning with Elliott et al.44,45 These studies have
been not only in chronic schizophrenia46,47 but also
in first-episode psychosis.48,49 Comparisons with other
disorders have included comparisons with bipolar disorder and depression, Alzheimers, Parkinsons, and
Huntingtons diseases, OCD and ADHD, neurosurgical
patients with damage to the frontal and temporal lobes, as
well as amygdalo-hippocampectomy (see CANTAB Web
site). An explicit comparison of chronic schizophrenia
with frontal and temporal lobe lesions was published
by Pantelis et al.46 That study showed that people
with chronic schizophrenia were severely impaired
at both the intradimensional and extradimensional
shifting stages, compared with both patients with lateral
PFC damage (who were mainly impaired at the extradimensional shifting stage) and temporal lobe lesioned
patients (who did not differ from age- and IQ-matched
controls). Other studies have shown that people with
chronic schizophrenia can also fail at the simple reversal
stage or compound discrimination stages.44,50 Jazbec
et al provided one of the first direct comparisons of ID/
ED performance and the WCST in schizophrenia, with significant correlations at the extradimensional shift stage.50
In terms of pharmacological treatment, Turner et al47
showed, in a double-blind placebo-controlled proof of
concept study, that acutely administered modafinil added
to conventional antipsychotic medication in schizophrenia improved several aspects of performance including
the ID/ED test. This proof-of-concept study demonstrated that it was indeed plausible to improve aspects
of cognitive function in schizophrenia and was subsequently back translated to show improvement in the
rat plus PCP model.43
Animal Models
There are number of different animal models of ID/ED
task performance that have been developed. For example, there is a paradigm for rhesus monkeys,42 and there
have been a number of studies of ID/ED performance in
Psychometric Data
Standardized data on 341 elderly subjects were published
in Robbins et al.51 Performance on the ID/ED test declined quiet sharply in the eldest cohort (>70) especially
119
In humans, there is evidence for double dissociation in

effects of indoleamine (serotonin, 5-HT) and catecholamine (dopamine, DA), noradrenaline (NA) manipulations on reversal learning, and extradimensional
shifting, respectively, in the ID/ED task. Thus, tryptophan depletion (implicating 5-HT) selectively impairs reversal learning in humans, whereas the catecholamine
agent methylphenidate improves extradimensional shifting.30,31 Noradrenergic manipulations appear particularly sensitive to human extradimensional shifting
performance.32 There is only weak evidence of effects
on extradimensional shifting of DA receptor antagonism33 or L-Dopa medication (in Parkinsons disease).34
In animals, as in humans, there is considerable evidence for an involvement of 5-HT in reversal learning.
In marmosets, selective depletion of 5-HT (and not
DA) in the orbitofrontal cortex impairs reversal learning.2426 Orbitofrontal 5-HT is also implicated in reversal
learning in the rat.35 In monkeys, DA depletion impairs
intradimensional shifting but not reversal or extradimensional shifting (which may even be artifactually improved
in certain instances).36,37 In rats, there is evidence that the
noradrenergic system is implicated specifically in the
extradimensional shifting.38,39 However, there is some
evidence in the rat and mouse of involvement of prefrontal cortex (PFC) DA in extradimensional shifting (based
on genetic studies with COMT polymorphisms40).
In addition, there are, approximately 30 studies of
pharmacological effects of drugs, including many candidates for the medication of cognitive deficits in schizophrenia on intradimensional and extradimensional
shifting in the rat model to be described below. These
drugs include the following: 5-HT6 receptor antagonist,
NA agonist, alpha-7 subunit nicotinic agonist, D1 receptor agonist, PDE inhibitors, and modafinil. Many of
these compounds selectively improve extradimensional
shifting performance, which may be most closely tied
to rule generation and selection. There are also selective
demonstrations of effects of stress on extradimensional
shifting performance and neuronal morphology in the
medial PFC in rats.41
marmosets.19,22 In addition, Birrell and Brown established a rat model based on textures and odors rather
than visual stimuli.29 This has been used to establish
the neuroanatomical substrates of the task in rats (see
above) and for pharmacological studies. It has also
been used for models of schizophrenia (eg, chronic
PCP treatment, amphetamine sensitization, isolation
rearing, MAM pretreatment). A typical study has shown
benefit of a PDE inhibitor or modafinil against deficits
produced by PCP at the extradimensional shift stage
that are not remediated by conventional antipsychotic
DA receptor antagonists.43
D. M. Barch et al.
Future Directions
More analysis of the neural and neurochemical substrates
of performance is needed in both humans and animal
subjects. Several specific issues require resolution or consolidation, as indicated above. These can be pursued using functional neuroimaging approaches in humans and
lesions or neuropharmacological manipulations in experimental animals. Longitudinal studies are also required
for schizophrenia, and other studies of pharmacological
effects, though proof of concept that ID/ED performance
can be enhanced has already been demonstrated. In psychometric terms, more data are required. The test may
have to be modified to provide good test-retest reliability.
For example, the fixed sequence of testing of the component processes ID shifting and ED shifting may need to
be replaced by a version of the test that interleaves the 2
forms of shifting. In addition, research is needed to determine whether deficits on specific stages of the task reflect differential deficits or generalized deficits that reflect
the greater discriminating power of the ED compared
with the ID, CD, or SD conditions. If ED conditions indeed have greater discriminating power, the test will need
to be modified to either match the ID and ED measures
on discriminating power or introduce another control
(eg, one in which set shifting deficits convey a performance advantage).
120
The Switching Stroop

Description
The switching Stroop is a variation on the classic
Stroop task. It was nominated as a measure of the rule
generation and selection construct as an example of an
asymmetric task-switching paradigm. An asymmetric
paradigm is one in which one of the task involves
a more prepotent dimensions than another. In such
task-switching paradigms, subjects must use the cue to
generate and/or select the appropriate rule to govern response selection for the upcoming stimulus. Thus, although this description will focus on the Stroop
specifically, it is possible that other asymmetric taskswitching paradigms may be just as promising. In the
classic Stroop,53 participants are presented with stimuli
that vary on 2 dimensions (eg, color words written in different colors). Participants are told to respond to one of
the dimensions and to ignore the other. Typically, one of
the dimensions is more prepotent (ie, word reading),
making it more difficult for participants to ignore that
dimension and leading to more interference from the prepotent dimension when trying to attend to and respond
based on the less dominant dimension (eg, color naming).
This typically leads to slower responding and more errors
when participants are told to respond on the basis of the
less dominant dimension.54 To perform the task correctly, especially when responding to the less prepotent
dimension, participants need to use task instructions to
select the appropriate rule for which stimulus dimension
to use for responding. However, standard versions of this
task do not sensitively probe the selection or maintenance
of rule information. This is because the task is typically
kept constant across blocks of trials, so that each trial
reinforces the correct response dimension and thus the
correct rule. To make the task more sensitive to the ability
to select and maintain the appropriate rule, one can vary
the task (ie, the dimension to be responded toword or
color) on a trial-by-trial basis55 (see figure 1). In such
switching versions of the Stroop, the subject is cued
to the relevant dimension (rule) prior to each trial.
This can be done either via an auditory cue or a visual
cue. In addition, one can vary the delay (1 vs 5 s) between
the task cue and the stimulus. This allows one to assess
whether rule maintenance is problematic by comparing
performance over a longer rule maintenance delay vs
a shorter rule maintenance delay.
In most versions of the Stroop, including the switching
Stroop, experimenters typically include 3 types of conditions (see figure 1): (1) congruent, in which the 2 dimensions point to the same response (eg, the word red
written in red); (2) neutral, in which response relevant information is present in only one dimension (eg, a row of
Xs printed in redsee Barch and Carter for discussion of
factors influencing choice of neutral condition56); and (3)
incongruent, in which the 2 dimensions point to different
at the extradimensional shift stage.51 There are limited

data on test-retest reliability; Lowe and Rabbitt found
quite low (r = 0.4) test-retest correlations for some
variables of the ID/ED test in normal volunteers.52
This is not surprising in view of the fact that the
WCST has very low test-retest reliability. Thus, significant developmental work is clearly needed to determine
how reliability on this test might be improved and what
reliability is like in patient groups. In general, this class of
tests does depend on insight and/or procedural learning
or other forms of practice effects derived from previous
experience of the tests, and between-group comparisons
are recommended for pharmacological studies to alleviate this problem (as compared with within subject designs
that require multiple testing sessions per person). The use
of parallel versions of the task may help with this concern
but may not entirely reduce the potentially long-acting
influence of insight into the basic nature and structure
of the task. In addition, the various conditions on the
task are not matched for discriminating power, with
the conditions that are often most impaired in schizophrenia likely those with the greatest discriminating
power (eg, extradimensional shift stages). Thus, work
is needed to determine whether greater impairments at
the extradimensional shift stage in schizophrenia reflect
a specific impairment in generating and/or selecting rules
or a generalized deficit that interacts with the psychometric characteristics of this stage of the task.
responses (eg, the word red written in blue). In addition,

performance on most versions of the Stroop can be measured in terms of either reaction times or errors. Experimenters typically look at one of 3 variables: (1) total
Stroop effect (performance in the incongruent conditionperformance in the congruent condition); (2) interference
(performance
in
the
incongruent
conditionperformance in the neutral condition); or
(3) facilitation (performance in the congruent conditionperformance in the neutral condition).
Construct Validity
There is a long history of research on standard versions of
the Stroop task that provides evidence for the construct
validity of this task as a measure of the ability of an individual to select and apply the appropriate rule for response selection. For example, there is a good deal of
computational modeling work demonstrating that performance on the Stroop task across many experimental
manipulations, including switching manipulations, can
be accounted for by mechanisms that provide for the active selection and maintenance of the appropriate task
rule.57,58 Such mechanisms have been implemented in
connectionist interactive activation models and have
been used to make predictions about specific patterns
of fMRI data.58,59 These models have typically assumed
that such task rule representations are stored in prefrontal cortex, maintained through recurrent connections
that provide top-down support for lateral inhibition
mechanisms in relevant processing pathways that allow
the less dominant pathway (eg, color naming) to compete
successfully with the more dominant pathway (eg, word
naming). In addition, there are data suggesting that indi-
vidual differences in the ability to actively maintain representations in working memory predicts individual
differences in the ability to select and effect use the appropriate task rule during Stroop task performance.60
As with the computational modeling work, these data
have been interpreted as reflecting an important role
for prefrontally maintained representations in successful
Stroop performance. One thing to note with the switching
Stroop is that while it is a good measure of the ability to
select task rules, it is not a good measure of the ability
to generate rules as the experimenter provides the rule.
In addition, there may be some confounding of taskswitching demands with rule selection demands in the
switching Stroop, given that RTs could also reflect the
time it takes to switch from one rule to another (eg, color
vs word), as well as the ability to maintain and apply the
appropriate rule. Thus, care would need to be taken to
dissociate these effects in future work with the switching
Stroop.
Neural Systems
There have been only a handful of studies using functional imaging to examine the neural substrates of performance on the switching Stroop. However, these studies
have been particularly informative as to the specific roles
played by dorsolateral prefrontal cortex (DLPFC) vs anterior cingulate cortex (ACC) in task performance.
Specifically, these studies have demonstrated that cuerelated activity in DLPFC was enhanced on trials
in which the participants were cued to color name as
compared with word read, consistent with the hypothesis
that greater top-down control to support rule representation is needed on color-naming trials.61,62 In contrast,
121
Fig. 1. (A) Illustration of the Structure of the switching Stroop Task. The cue is either to color or word read and then the stimulus is
congruent, neutral, or incongruent. (B) An illustration of trial structure for computing conflict adaptation effects in the trial-by-trial Stroop.
D. M. Barch et al.
Pharmacological and Behavioral Manipulation of Task

Performance
A number of different variations on the Stroop task have
been used in a range of pharmacological and behavioral
manipulation studies. Of most relevance to the current
topic are those that involved individuals with schizophrenia. For example, work by Barch and Carter demonstrated that low doses of amphetamine in patients
taking stable doses of typical antipsychotics improve reaction times on a standard Stroop task,64 though this improvement was found in all conditions, not just the
incongruent condition. In addition, research has shown
that nicotine administration can improve Stroop interference, with evidence of greater improvements in individuals with schizophrenia than in controls.65 Thus, prior
research indicates that Stroop performance is amenable
to change with pharmacological interventions in individuals with schizophrenia. However, to our knowledge, research has not yet examined the amenability of switching
Stroop performance specifically to change via either
pharmacological or behavioral manipulations.
Animal Models
A number of paradigms for use in animals have been developed that are designed to capture the type of response
competition that requires the type of appropriate rule selection required by the switching Stroop in humans. For
example, work in rats has focused on biconditional discrimination tasks that require animals to use contextual
information to modify responses to specific stimuli.6668
In addition, paradigms have been developed for monkeys
that are designed to mimic critical features of the Stroop
task, including the need to selectively attend to one dimension of stimulus and the inclusion of different feature
dimensions that provide conflicting response information.65,69,70 The degree of homology between the cognitive and neural mechanisms engaged by these
paradigms in animals vs humans needs further investiga122
tion. However, these studies provide encouraging evidence of the ability to model Stroop-like task
performance in animals.
At least one study has specifically examined switching
Stroop performance in patients with schiophrenia.55
This work used a version of the switching Stroop in
which: (1) an auditory cue signaled whether participants
should word read or color name, (2) there were 2 different
delays between the cue and the onset of the stimulus (1 vs
5 s); and (3) the neutral stimuli for color naming were
rows of Xs. The schizophrenia patients showed disproportionate interference in errors compared with controls
for color naming as well as a disproportionate total
Stroop effect in reaction times compared with controls
for color naming (see figure 2). These effects did not interact with delays, suggesting that the magnitude of the
impairments was similar for both short and long delays
(see figure 2). Functional imaging work with the Stroop
has consistently revealed deficits in the activation of
DLPFC and ACC in schizophrenia. For example, Kerns
et al71 found reduced anterior cingulate activity during
incongruent Stroop trials in schizophrenia, Weiss
et al72 found underactivation of DLPFC and anterior cingulate in schizophrenia during performance of a Stroop
task, and Yucel et al73 found underactivation of paracingulate cortex in schizophrenia. These studies did not use
switching Stroop tasks specifically and instead used standard Stroop versions. However, standard Stroop versions do still assess the ability to select and apply the
appropriate rule, though they may just not be as sensitive
to this process as the switching Stroop. Deficits on the
Stroop task are also present in the relatives of individuals
with schizophrenia, suggesting that they are associated
with vulnerability to the disorder.74,75 In addition, recent
work has shown reduced DLPFC activity during performance of the Stroop in unaffected first-degree relatives of
patients, though intact ACC activity.76 However, to our
knowledge, the switching Stroop has not been used with
relatives of individuals with schizophrenia.
Psychometric Data
There is very little information about any psychometric
characteristics of the switching Stroop. Cohen et al did
report an internal consistently of 0.9 for incongruent
color-naming errors at a 5-s cue-probe delay and an internal consistency of 0.53 for color-naming facilitation at
a 5-s delay.
Future Directions
There are a number of different important future directions for research on the switching Stroop for use in clinical trial contexts in schizophrenia. As noted above, little
is known about the psychometrics of this task, and thus,
cue-related anterior cingulate (ACC) activity did not vary

as a function of task, but ACC activity was greater on
incongruent relative to congruent trials, consistent with
a role of for ACC in monitoring for competition during
responding.61
Research has also examined the degree to which lesions
to the frontal cortex impair performance on standard versions of the Stroop task, though to our knowledge no
such studies have been conducted with switching Stroop
tasks. Although the evidence in this regard is mixed, some
meta-analytic work suggests that frontal lesions are reliably associated with impaired performance on the Stroop
task,63 though it is not clear that such deficits are isolated
to conditions in which rule selection and maintenance are
most critical (eg, incongruent conditions or color naming
specifically).63
Accuracy
0.4
Proportion Errors
0.3
0.2
0.1
0
N
Long ISI
Short ISI
Long ISI
Short ISI
Color Naming
Word Reading
Reaction Times
1500
Millseconds
1300
1100
900
700
C
Short ISI
Long ISI
Color Naming
Patients with Schizophrenia
Short ISI
Long ISI
Word Reading
Patient Controls
Healthy Controls
Fig. 2. Stroop Task Accuracy and Reaction Times from Cohen et al.55 Reprinted with permission from the American Psychological
Association. C 5 congruent, N 5 neutral, I 5 incongruent.
studies examining test-retest reliability, practice effects,

floor, and ceiling effect, etc are needed. In addition, research is needed to establish the most sensitive neutral
condition to use in studies of schizophrenia. In addition,
the ways in which the task has been presented in prior
studies involve many trials and a lengthy presentation.
This may not be practically feasible in clinical trial contexts. As such, work on ways in which to shorten and enhance the sensitivity of the task is needed. Although
standard versions of the Stroop task have been shown
to be amenable to pharmacological manipulation,
work is needed to demonstrate that performance on
the switching Stroop can be enhanced either pharmacologically or behaviorally. In addition, work is needed to
establish that the greater deficit on color-naming trial
references a differential deficit rather than a generalized

deficit given that the color-naming condition may have
greater discriminating power.
Dynamic Adjustments in Control
CNTRICS received 4 initial nominations for the dynamic
adjustments in control construct: (1) the attention networks task (ANT); (2) the Simon task; (3) the trialby-trial Stroop task; and (4) the stop signal task. In
the stop signal task, as described below, dynamic adjustments in control are examined by determining how individuals modulate their behavior when given a cue to
interpret an ongoing response. In the ANT, the Simon
task, and the trial-by-trial Stroop, dynamic adjustments
123
D. M. Barch et al.
Tasks Recommended for Immediate Translation for

Dynamic Adjustments of Control Conflict and Error
Adaptation in the Stroop Task
Description
The basic design of a Stroop task was described above
under the switching Stroop description. In addition,
the basic approach to assessing dynamic adjustments
in control in the context of conflict tasks was described
above. To clarify, in the Stroop task, a high conflict trial
for color naming is an incongruent trial in which the word
and the color are different (eg, RED written in blue). A
low conflict trial is a congruent trial in which the word
and the color are the same (RED written in red). In previous studies, dynamic adjustments in control via conflict
adaptation have been studied in the Stroop task by comparing 2 sets of trials (see figure 1B): (1) incongruent
124
following an incongruent (iI) vs incongruent following

a congruent (cI) and (2) congruent following a congruent
(cC) vs congruent following an incongruent (iC).9093 Dynamic adjustments in control via posterror adjustments
are examined as described above. There are several design
aspects that one needs to attend in designing a Stroop
study for use in examining dynamic adjustments in control. Most importantly is the need to ensure that the
sequences of trials are set up in a way to allow sufficient
numbers of the relevant pairings of trials, as described
above. This may require a pseudorandom trial sequence
generation that ensures a lack of predictability on the
part of the participants coupled with a sufficient number
of trials in each cell.
Construct Validity
The construct validity of conflict and error adaptations
as measures of dynamic adjustments in control has
been supported from a number of different types of studies. For example, several different computational models
have been developed that account for both error and conflict adaptation effects in terms of the detection of conflict
and the use of such conflict detection to adjust and regulate the amount of control. These models are framed in
terms of control being implemented via active representations of task rules or S-R mappings (potentially housed
in prefrontal cortex) that provide feedback over posterior
corticalprocessing areas,9497 with conflict detection occurring via anterior cingulate (ACC) mechanisms that
monitor for the degree of conflict between competing response options. One potential threat to the construct validity of conflict adaptation effects is the degree to which
the improvement in performance on iI trials compared
with cI trials or on cC vs iC trials can be accounted
for by repetition priming instead of increases in cognitive
control.98 Although it is clear that such repetition priming can facilitate performance on subsequent trials, research has also shown that conflict adaptation effects
are present even when no exact repetitions are used in
the calculation of such effects.90,99,100
Neural Systems
There is a burgeoning literature on the neural substrates
of such conflict adaptation and error adaptation effects
as measures of control adjustments, stemming from both
ERP and fMRI research in humans. In the evoked response potential (ERP) literature, components such as
the error-related negativity (ERN) and the N2 (or
N450 during the Stroop task) have been localized to
the ACC and have been frequently investigated in the
context of conflict and error monitoring and control
adjustments. In addition, the conflict-related negativity
(CRN) has also been investigated in relationship to conflict monitoring and control adjustments. Further, there
is relevant lesion work as well as recent nonhuman
of control are examined by investigating 2 effects: (1) conflict adaptation77 and (2) posterror adjustments.78,79
Conflict adaptation is thought to reflect behavioral
adjustments following trials with high conflict (eg, incongruent trials) vs those with low conflict (eg, neutral or
congruent trials). In other words, participants are faster
to response to an incongruent trial that is preceded by an
incongruent trial compared with an incongruent trial preceded by a congruent trial.77,80 The interpretation of this
result has been that the conflict engendered by the first
incongruent trial leads to an upregulation in control
that reduces the conflict induced by the next incongruent
trial.80 Posterror adjustments are thought to reflect a similar process. Participants are typically slower, but more
accurate, on trials following errors vs trials following correct responses.78,79 This is again thought to reflect an
upregulation of control engendered by the awareness
(not necessarily conscious) of having made an error.
The executive control breakout group felt that all 4
tasks were reasonable measures of the construct of interest. However, the breakout group felt that the ANT, the
Simon, and the Stroop were all variations on a theme,
with a similar deep structure in terms of assessing dynamic adjustments in control. Of the 3, the breakout
group felt that the Stroop version has the largest extant
literature in terms of the cognitive and neural bases of
dynamic adjustments in control and the largest existing
literature in schizophrenia. Thus, the breakout group recommended that the Stroop version be put forth for immediate translation for use in clinical trials in
schizophrenia. In addition, the breakout group made
a similar recommendation for the stop signal task, given
that it measured a somewhat different aspect of dynamic
adjustments in control and thus might provide complimentary information. Table 2 provides a brief overview
of these 4 tasks and a very brief summary of their pros
and cons in regards to the selection criteria.
Table 2. Description of Tasks Nominated for Dynamic Adjustments in Control Construct Definition: The Processes Involved in Detecting
the Occurrence of Conflict or Errors in Ongoing Processing, Identifying the Type of Control Adjustments Needed, and Recruiting
Additional Control Processes7
Stop signal task

Construct validity
Supported by evidence of stopping deficits in populations thought to have impulse control problems, such as ADHD
Neural systems
Evidence from human imaging and lesion studies on the role of inferior frontal gyrus in inhibition
Evidence from human imaging and neurological disorder studies on the role of the subthalamic nucleus in inhibition
Strong evidence that inhibition on the stop signal task is affected by manipulations of catecholamines, which may involve both
dopamine and norepinephrine
Animal models
There is a primate model that involves countermanding of eye movements
There is a rodent model that involves an auditory stop signal and button press
Evidence for reduced stopping accuracy and elongated stop signal reaction times
Psychometric data
The use of an adaptive stop signal reaction time estimate procedure reduces floor and ceiling effects
Preliminary positive evidence for good test-retest reliability and small practice effects
Future directions
Need to develop better adaptive methods for stop signal reaction time estimation that reduce or eliminate strategy changes on the
part of the participant
Need to directly compare different versions of the stop signal task in terms of sensitivity and psychometric characteristics
Need an evaluation of the degree of homology between the human and animal versions
Other nominated tasks
Attention networks task
Construct validity
Evidence from individual differences research and computational modeling
Neural systems
Evidence for activation of the ACC during conflict81,82
Polymorphisms in dopamine receptor 4 and monoamine oxidase A genes were found to be related to differential behavioral
conflict response83 as well as differential ACC conflict response in an fMRI study84
No published data
Animal models
Animal models available for monkey and rat, though the degree of homology needs further research
125
Conflict and Error Adaptation in the Stroop Task

Construct validity
Supported by computational modeling work and individual differences research
Neural systems
Evidence for the role of ACC in conflict detection and DLPFC in control adjustments from human neuroimaging studies
Mixed evidence for the role of DLPFC in control adjustments from lesion research
Mixed for the role of ACC in conflict detection from lesion research
Consistent evidence that basic ERN effects to errors can be modulated via pharmacological challenges
Mixed evidence on the degree to which behavioral performance adjustments can be manipulated pharmacologically
Evidence for dissociations in the effects of pharmacological manipulations on ERN indices vs behavioral indices
Animal models
Evidence for reduced conflict adaptation and reduced posterror slowing (mixed) in schizophrenia
Evidence for reduced ACC responses to errors and conflict
Psychometric data
Little to none currently available
Future directions
Research needed on assessing and improving the psychometric characteristics of the error and conflict adaptation effects
Research needed on ways to make the use of conflict and error adaptation effects feasible in clinical trials contexts (task length,
standardized administration, enhanced sensitivity to detect deficits, etc.)
Evaluation of antipsychotic effects on error and conflict adaptation effects
D. M. Barch et al.
Table 2. Continued
primate work. Much of this literature suggests that the

dorsal anterior cingulate (ACC), via inputs from the midbrain dopamine system, monitors for and detects conflict.80,94,95,97,101 In the discussion below, the term
ACC will be used to refer both to activation in fMRI
studies and to ERP studies examining the ERN, the
N2, and the CRN, given the large literature linking these
components to an ACC generator. Although there is
some debate as to the precise interpretation of the role
of ACC in cognition,102 it is clear empirically that the
ACC is more active on high vs low conflict trials in
the Stroop task and that it is more active on cI than iI
trials in the Stroop task.90,92 Further, the magnitude of
ACC activity on trial N predicts the degree of behavioral
adjustment on trial N 1, with greater behavioral adjustments associated with greater ACC activity on the previous trial in the Stroop.90 In addition, the ACC is strongly
active on error trials in the Stroop, and the degree of ACC
activity errors predicts the magnitude of posterror slowing (more ACC / more slowing).90,103,104
The actual implementation of control adjustments following conflict detection has been linked to the function
of DLPFC. For example, DLPFC activity is greater on iI
than on cI trials, putatively reflecting the increased control necessary to execute conflict adaptation,92,105 and the
degree of DLPFC activity predicts the degree of conflict
adaptation.88,90,92,106 Further, the degree of ACC activity
on previous high conflict trials predicts the degree of
DLPFC activity on the subsequent trial.88,90
The lesion data on the role of ACC and DLPFC in control adjustments is mixed. For example, Fellows and Farah107 did not find evidence that ACC lesions influenced
Stroop performance of posterror slowing. However, recent work by di Pellegrino showed reductions in both
conflict adaptation and posterror slowing in ACC lesion
126
patients.108 Damage to PFC does seem to reduce the likelihood that individuals will correct their errors, but does
not appear to reduce post error slowing, which is not consistent with the hypothesis that DLPFC is necessary for
conflict-related adjustments in control.109
The above discussion perhaps implies more consistency in the literature on the neural substrates of dynamics adjustments of control than would be apparent based
on a thorough review of the literature. Although there is
a large amount of data to support important roles for
DLPFC and ACC in dynamic adjustments of control,
there is also conflicting data and data that suggest other
roles for ACC in action monitoring102,110,111 (for a review,
see Carter and van Veen112). As noted above, some of the
open questions involve the degree to which ACC and/or
DLPFC function is necessary for dynamic adjustments of
control and the degree to which posterror and conflict
adaptation effects reflect the same mechanisms. This is
still a rich and active area of investigation that is generating much interesting data about the neutral substrates
of cognitive control.
Pharmacological and Behavioral Manipulation
A number of studies have examined pharmacological
influences on the ERN and both conflict adaptation
and posterror slowing, with mixed results. For example,
benzodiazepines reduce the magnitude of the ERN113,114
and show some evidence of reducing the N2 congruency
effect on the Stroop.114 Benzodiazepines do not appear to
reduce posterror slowing but do reduce the influence of
errors on conflict adaptation114 and reduce the number of
errors that are corrected.113 Yohimbine (a selective
alpha-2 adrenergic receptor antagonist) increases the
ERN but does not change posterror slowing. However,
A selective impairment of ANT conflict processing in schizophrenia has been reported85
Psychometric data
Preliminary evidence of good test-retest reliablity86
Simon task
Construct validity
Behavioral studies have shown robust postconflict and posterror adaptation with this task87
Neural systems
Human fMRI studies have linked the ACC to conflict-related activity and subsequent adjustments in control and the DLPFC
to supporting those adjustments as for the Stroop88,89
Individuals with lesions to rostral ACC show impaired conflict adaptation on the Simon task 96
Unknown
Animal models
Unknown
Psychometric data
Unknown
Animal Models
A growing number of studies have examined conflict- and
error-related processing in nonhuman primate models.
These studies indicate that neurons in a number of medial
prefrontal regions are modulated in response to errors,
including anterior cingulate, supplementary motor
regions including the supplementary eye fields, and the
superior frontal gyrus.119121 However, it is less clear
in monkeys that these regions also signal nonerrorrelated conflict, with some studies suggesting conflictrelated activity122 and other not,121,123 and the temporal
dynamics of the error-related signal may differ from that
seen in humans. Some work has also examined the role of
DLPFC in conflict adaptation in nonhuman primates.
This work demonstrated that lesions to DLPFC reduced
postconflict behavioral adjustments, but that lesions to
the ACC did not.124 There is also recent work in a rat
model of posterror slowing. In this study, rats were
trained a simple reaction time task, in which they had
to hold a lever for a specific period of time. The rats
showed longer RTs on trials following an error vs those
trials following a correct response. Further, inactivation
of dorsal medial PFC reduced the amount of posterror
slowing shown by the rats.125
A growing number of studies have examined fMRI and
ERP measures of error and conflict processing in schizophrenia as wellasindicesofdynamicadjustmentsincontrol.
Several studies suggest that individuals with schizophrenia
show reduced error-related ACC responses71,126132 as well
as reduced posterror slowing71,126 on the Stroop task as well
as other tasks. However, there is also evidence that patients
with schizophrenia can show normal error correction
performance even in the context of reduced ACC responses
to errors127,133 and that the relationship between the
magnitude of the ERN and error-related behaviors is intact

in schizophrenia.130 Individuals with schizophrenia also
show reduced conflict-related ACC activation on the
Stroop task71,127 as well as reductions in conflict adaptation
effects.71
Psychometric Characteristics
There is little published psychometric data on either ERN
or ACC responses to error- or conflict-related processes
or behavioral indices of conflict adaptation or posterror
slowing.
Future Directions
As noted above, there are still open questions about the
specific neural circuits that underlie dynamic adjustments
in control and open questions about the degree to which
conflict adaptation and posterror changes in behavior reflect the same or dissociable mechanisms. There is also
a large need for psychometric research on the conflict adaptation and posterror change measures, as well as for
ERN and ACC responses. This includes research on
test-retest reliability, practice effects, and floor and ceiling
effects. In addition, work is needed to determine the optimal number of trials needed to obtain robust estimate of
conflict- and error-related behavioral adjustments. In addition, more information is needed on the degree to which
reductions in ACC responses (measured either by fMRI
or ERPs) or behavioral indices of dynamic adjustment of
control are negatively influenced by either typical or atypical antipsychotics in individuals with schizophrenia.
The Stop Signal Task
Description
The stop signal paradigm measures an individuals ability
to interrupt a motor response after it has been initiated.
The subject is presented with a primary task that generally involves a 2-choice response, such as deciding which
of 2 letters is presented and pressing the appropriate button. On a proportion of trials, the subject is presented
with a stop signal, in which case the subject is
instructed to withhold their response if possible (see
figure 3). The likelihood of stopping is determined by
the delay at which the stop signal is presented; if the signal
occurs soon after the stimulus, then the subject is very
likely to stop, whereas successful stopping becomes
much less likely if the signal occurs later in the trial.
The stop signal paradigm allows one to estimate the
time needed to stop a behavior. Although this is not an
observable quantity (ie, it is a parameter regarding trials
where no behavior was made), it can be estimated from
observed behavior if certain assumptions are made. These
assumptions comprise the so-called race model of the
stop signal reaction time.134 This model assumes that response production and response inhibition processes race
127
Yohimbine does lead to fewer errors following errors,

another reflection of control adaptation.115 Caffeine
also increases the ERN but does not influence posterror
slowing.116 In terms of dopaminergic modulation, haldol
(a D2 antagonist) consistently reduces the magnitude of
the ERN,117,118 while amphetamine (indirect dopamine
agonist) increases it.118 However, neither appears to influence N2 congruency effects or posterror slowing or
conflict adaptation effects.114,117,118 Olanzapine reduced
the ERN and reduced the degree of posterror slowing.118
Thus, these studies provide strong evidence that basic
ERN effects to errors can be modulated via pharmacological challenges, both in terms of enhancement and reduction. However, there is mixed evidence as to the
degree to which behavioral performance adjustments
can be manipulated pharmacologically and evidence
for dissociations in the effects of pharmacological manipulations on ERN indices vs behavioral indices.
D. M. Barch et al.
GO TRIAL
>
[The time is takes to respond is the RT]
STOP TRIAL (with a relatively short Stop Signal Delay)
>
Stop Signal Delay
STOP TRIAL (with a relatively long Stop Signal Delay)
>
Stop Signal Delay
Stop Signal
Fig. 3. An Illustration of the Structure of the Stop Signal Task. On each trial, subjects are presented with a stimulus to which they must respond.
On some trials they receive a subsequent cue that indicates that they have to stop their response (the stop signal). If the stop signal comes
quickly after the initial stimulus, it is a short stop signal delay and it is easier for participants to halt their response. If the stop signal is presented
at a longer time following the initial stimulus, it is harder for participants to halt their response.
against one another and that the first one to finish determines whether a response is executed or inhibited. Although the strong assumption of independence is
unlikely to be correct,135 it nonetheless does a good job
of characterizing behavior in the stop signal task.136
The goal of the race model is to allow estimation of the
stop signal reaction time (SSRT), which is the amount of
time needed by the stop process to successfully inhibit
a response. There are a number of ways to estimate
SSRT;134 a commonly used method does this by determining the stop signal delay at which subjects fail to
stop on 50% of trials. This value can be estimated either
by using an adaptive staircase method to estimate it directly or by evaluating performance at a range of stop
signal delay (SSD) values and interpolating to estimate
it. Once is known, then the SSRT can be estimated as,
since failing on 50% of trials means that half of the
responses are faster than SSRT and half are slower,
which is equivalent to the median RT.
A comprehensive users guide to the stop signal task
was published by Logan and Cowan,134 Once the range
of SSD values are known, then the SSRT can be estimated, since failing on 50% of trials means that half of
the responses are faster than SSRT and half are slower,
which is equivalent to the median RT. which should be
read closely by any prospective user of the task. It is im128
portant to highlight that this task is highly sensitive to the

instructions and feedback that are given to the subject. In
particular, many subjects will have the tendency to delay
their responses in order to wait for the stop signal. If an
adaptive staircase method is used, then this can result in
SSD values that do not converge but rather continue increasing across the entire task. When this occurs, it is impossible to obtain an accurate estimate of SSRT because
there is no SSD for which the proportion of successful
stops is known accurately. These strategies can be
reduced by carefully instructing the subjects, and also
by providing feedback on go task response time after every block, to ensure that subjects do not slow down on go
trials. It is also critical to examine the resulting staircases
when adaptive methods are used, to ensure that they have
converged to a stable estimate of. It is particularly useful
to use multiple staircases, so that reliability can be
assessed across them and so that subjects cannot easily
predict changes in SSDs.
One particular issue that has arisen in the literature is
whether SSRT should be estimated using adaptive methods or using a set of fixed SSDs. Although more complicated to implement, the adaptive staircase method has
a number of benefits. Psychologically, it ensures that subjects with very different levels of inhibitory function
exhibit the same proportion of failed inhibitory trials,
Stop Signal
which eliminates confounds regarding frustration or error likelihood prediction. Psychometrically, it ensures
roughly equal numbers of successful and unsuccessful
stop trials, which provides increased power to contrast
these conditions; contrast this with the go/no-go paradigm, in which there is generally a very low rate
(;10%) of commission errors. Simulation studies136
also show that SSRT is more reliably estimated using
an adaptive tracking method than using a set of fixed
SSDs. The one caution, as noted above, is that adaptive
methods may be more likely to lead subjects to adopt
waiting strategies; careful instruction and continuous
feedback throughout testing can help avoid this problem.
The stop signal task measures the ability to reactively inhibit a response once it has been initiated. Its construct
validity as a measure of control has been demonstrated
primarily in the context of impulsivity and behavioral control. In the normal population, SSRT is correlated with
real-world impulsivity as measured using paper-and-pencil
impulsivity scales.137 A large body of research has linked
performance on the stop signal task to disorders of impulse
control, including attention deficit hyperactivity disorder
(ADHD) (reviewed by Lijffijt et al138), obsessive-compulsive disorder and trichotillomania,139 methamphetamine
abuse,140 and cocaine use.141,142
Neural Systems
Evidence regarding the neural systems supporting stop
signal performance in humans comes from both lesion
and imaging studies. A study of frontal lesion patients143
provided evidence that the right inferior frontal gyrus is
critical for stopping in this task; patients with lesions to
this area were showed lengthened SSRT values, and
SSRT was correlated with the amount of lesion in the region. However, another study found that the medial
superior prefrontal cortex (supplementary and presupplementary motor areas) is also important for response inhibition.144,145 Neuroimaging studies have
also implicated both of these prefrontal regions, which
show activity on stop trials in comparison to go trials
that are correlated across subjects with SSRT.146149
These studies have not generally found activity in right
inferior prefrontal or medial superior prefrontal cortex
for the comparison of successful vs unsuccessful inhibition; this is predicted by the race model because that
model proposes that successful inhibition is primarily
driven by the speed of the go process rather than differences in the effectiveness of the stop process.
There is also evidence that basal ganglia regions may be
involved in stop signal inhibition, specifically the subthalamic nucleus (STN). Stimulation of the STN in individuals
with Parkinsons disease resulted inimproved stop signal reaction time.150 Neuroimaging studies have also found that
Pharmacological or Behavioral Manipulation

There is strong evidence that inhibition on the stop signal
task is affected by manipulations of catecholamines.
SSRT is improved by methylphenidate in both children153 and adults154 with ADHD. It is likely that this
reflects both dopaminergic and noradrenergic effects:
SSRT is improved by low doses of cocaine155 and by specific inhibition of noradrenergic reuptake using atomoxetine156 and desipramine.157 There is also evidence for
association between stop signal reaction time and genetic
polymorphisms in the D4 dopamine receptor and dopamine transporter genes.158
Animal Models
One of the strong aspects of the stop signal task for translational research is the existence of animal models in both
nonhuman primates and rodents. The primate model
involves countermanding of eye movements; after fixation on a central point, a lateral stimulus is presented
and the animal makes an eye movement to that stimulus.
On a small proportion of stop trials, the central fixation
point reappears, and the animal is trained to countermand its eye movement on those trials. Extensive work
has characterized the neural systems involved in this process (reviewed by Schall et al120). This model has
provided a great deal of knowledge about how inhibition
occurs in eye movements, but there are reasons to believe
that the countermanding model may not be applicable to
other forms of the stop signal task.
A rodent model of the stop signal paradigm has been
developed by Eagle et al.159,160 In this paradigm, the rat
first presses a button to start the trial and then has to
press a second button as the go response; on trials
when an auditory stop signal is presented, the rat must
withhold response in order to obtain reward. Studies using this paradigm have shown that stopping is affected by
lesions to the orbitofrontal cortex and STN161 and that
it is modulated by d-amphetamine,159 modafinil, and
methylphenidate.162 These results are largely consistent
with the human literature and provide strong support
for this paradigm as a model system for translational research on response inhibition.
There are relatively limited data on stop signal performance in schizophrenia. Badcock et al163 found that
129
Construct Validity
STN is activated for successful stop vs go trials and that its

activation is also correlated with SSRT across subjects.146,147 The role of the STN in stopping is consistent
with its excitation of pallidal neurons, which results in thalamic inhibition and cortical deactivation.151 There is also
evidence that STN receives direct projections from both
of the prefrontal regions implicated in stopping,147 as
part of the hyperdirect pathway into the basal ganglia.152
D. M. Barch et al.
Psychometric Characteristics
As noted above, the use of adaptive SSRT estimation
methods imbues the stop signal task with desirable psychometric characteristics. By adapting the SSD to each
subject to ensure 50% successful inhibition, it prevents
ceiling and floor effects (which are problematic in the
go/no-go task). In unpublished work (J. R. Cohen &
R. A. Poldrack, unpublished), there was no evidence
of practice effects in SSRT estimates obtained using
adaptive methods. Previous work has also shown that
the mean proportion of inhibition, when measured using
a fixed set of SSDs, exhibits high test-retest reliability.
Future Directions
Future work on the stop signal task should focus on at
least 4 areas. First, a better characterization of response
inhibition function across a range of neuropsychiatric
disorders is needed, in order to characterize whether there
is any specificity to the inhibitory deficits suggested by the
existing literature. These studies should also include unaffected relatives, particularly given the potential for antipsychotic drugs to interfere with inhibitory functions.
Second, research is needed on ways to develop better
adaptive methods for estimation of SSRT and better procedures to prevent strategic adaptations that result in
slowing of go response times. Third, work is needed
on ways to better delineate different aspects of the
stop signal paradigm and their potential effects on reli130
ability and neural systems validity. Relatively little

work to date has directly compared different forms of
the stop signal task (eg, using visual vs auditory stop signals or using standard vs selective stopping tasks).
Fourth, the field needs a better characterization of the
genetic architecture of response inhibition. There is evidence from candidate gene studies for association with
specific genes, but these await validation by genomewide association studies.
General Conclusion
The discussion above aimed to present a summary of the
data upon which decisions were made at the third
CNTRICS meeting in regards to which tasks to nominate
for translation into clinical trials. In this selection process
and in the descriptions provided above, much emphasis
was placed on the evidence regarding construct validity,
the neural substrates of the tasks, and the availability of
animal models. This emphasis was intentional because
the guiding theme of the CNTRICS initiative has been
that we need to leverage the advances in cognitive neuroscience to develop better and more effective treatments
for cognitive impairment in schizophrenia. One thing
that should be clear from the above descriptions though
is that much work is needed to make these paradigms
derived from the basic science literaturepractical and
feasible for use in clinical trials. For some, this process
might seem daunting and might lead one to fall back
on traditional tasks for which such psychometric and
practicality information is already available. However,
we would argue strongly that the hard work that lies
ahead is well worth it and critically necessary so as to
move the field forward. The use of paradigms with a tight
grounding in modern theories of human and animal cognitive neuroscience may allow us to pursue a more rationally guided pathway to drug development and testing
that has the potential to point the way toward more
effective treatment pathways. Yes, it is true that development work for some of these tasks will not end up being
successful and some may not survive the translation
process in a way that allows them to be put to good
use in clinical trials. However, even if only some tasks
are successfully translated for use in clinical trials, and
these paradigms end up improving our ability to design
and evaluate effective cognitive treatments, then the
whole process will have been worth the time and effort
of the many people involved.
Funding
NIH, the McDonnell Center for Higher Brain Function,
and NARSAD to D.M.B; NIMH, the Robert Wood
Johnson Foundation and NARSAD to C.S.C; C.S.C
has served as a consultant for Pfizer, Eli Lilly, and Roche.
NIMH, the McDonnell Foundation, and NARSAD to
patients with schizophrenia, as well as subjects with

other forms of psychosis, exhibited impaired response
inhibition (as measured by the slope of inhibition functions because a fixed SSD method was used). Vink
et al164 examined inhibitory control using an adaptation
of the stop signal task with a visual stop signal, in patients
with schizophrenia and unaffected siblings. SSRT was
not estimated, but instead the accuracy of stopping
was reported. They found that patients were impaired
at stopping, particularly when stop signals became
more likely (ie, after a larger number of go trials). Siblings
showed a marginally significant effect in the same direction. Bellgrove et al165 examined stop signal performance
in a group of early-onset schizophrenic patients diagnosed as either paranoid or undifferentiated. SSRT
was estimated using an adaptive staircase method, with
separate SSRT estimates for each hand. The undifferentiated patients exhibited a substantial deficit in SSRT for
the left hand; no deficits were observed for the paranoid
group. Finally, Enticott et al166 compared SSRT in
schizophrenic inpatients and matched controls and found
that schizophrenics showed increased SSRT. Thus, although not conclusive, there is evidence from several
studies for deficits in response inhibition in schizophrenic
patients and, perhaps more important, in unaffected siblings as well.
D.M.B; NIH to T.S.B; NIH; James S. McDonnell Foundation for Dr. Poldrack to R.A.P; MRC, Wellcome
Trust, McDonnell Foundation, and E.U. (FP6) to
T.W.R. He consults for Cambridge Cognition, E. Lilly,
GlaxoSmithKline and Allon Therapeutics.
Acknowledgments
References
1. Reichenberg A, Harvey PD. Neuropsychological impairments in schizophrenia: integration of performance-based
and brain imaging findings. Psychol Bull. 2007;133:833858.
2. Kerns JB, Berenbaum H. Cognitive impairments associated
with formal thought disorder in people with schizophrenia.
J Abnorm Psychol. 2002;111:211224.
3. Delawalla Z, Barch DM, Fisher Eastep JL, et al. Factors
mediating cognitive deficits and psychopathology among
siblings of individuals with schizophrenia. Schizophr Bull.
2006;32:525537.
4. Barch DM, Mitropoulou V, Harvey PD, New AS, Silverman
JM, Siever LJ. Context-processing deficits in schizotypal personality disorder. J Abnorm Psychol. 2004;113:556568.
5. McClure MM, Bowie CR, Patterson TL, et al. Correlations
of functional capacity and neuropsychological performance
in older patients with schizophrenia: evidence for specificity
of relationships? Schizophr Res. 2007;89:330338.
6. Kerns JG, Nuechterlein KH, Braver TS, Barch DM. Executive functioning component mechanisms and schizophrenia.
Biol Psychiatry. 2008;64:2633.
7. Carter CS, Barch DM, Buchanan RW, et al. Identifying
cognitive mechanisms targeted for treatment development
in schizophrenia: an overview of the first meeting of the
Cognitive Neuroscience Treatment Research to Improve
Cognition in Schizophrenia Initiative. Biol Psychiatry. 2008;
64:410.
8. Carter CS, Barch DM. Cognitive neuroscience-based
approaches to measuring and improving treatment effects
on cognition in schizophrenia: the CNTRICS initiative.
Schizophr Bull. 2007;33:11311137.
9. Mathewson K, Dywan J, Snyder PJ, Tays W, Segalowitz S.
Electrophysiological correlates of older and younger adults
response to error feedback using a maze learning task. Psychophysiology. in press.
10. Snyder AM, Maruff P, Pietrzak RH, Cromer JR, Snyder PJ.
Effect of treatment with stimulant medication on nonverbal
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
131
We would like to thank Jody Conrad, Carol Cox, and Deb

Tussing, whose efforts have been invaluable in the
CNTRICS process. In addition, we would like to thank
the other members of the executive control breakout
group, including Andrew Blackwell, Nelson Cowan,
Randy Engle, Judith Jaeger, Rich Keefe, John Kerns,
Ann Kring, Gina Kuperberg, Steve Luck, Henry
Mahncke, Paul Maruff, Michael Minzenberg, Holly
Moore, Venkatesha Murthy, Keith Nuechterlein,
Patricio ODonnell, Brad Postle, Nathan Pradeep,
Charan Ranganath, Martin Sarter, Larry Seidman,
Magnus Sjogren, Ed Smith, Mark Tricklebank, Sophia
Vinogradov, and Andy Yonelinas.
11.
executive function and visuomotor speed in children with attention deficit/hyperactivity disorder (ADHD). Child Neuropsychol. 2008;14:211226.
Williams MT, Morford LL, McCrea AE, Wood SL, Vorhees
CV. Administration of D,L-fenfluramine to rats produces
learning deficits in the Cincinnati water maze but not the
Morris water maze: relationship to adrenal cortical output.
Neurotoxicol Teratol. 2002;24:783796.
Snyder PJ, Jackson CE, Piskulic D, Olver J, Norman TR,
Maruff P. Spatial working memory and problem solving in
schizophrenia: the effect of symptom stabilization with atypical antipsychotic medication. Psychiatry Res. in press.
Pietrzak RH, Maruff P, Mayes LC, Roman SA, Sosa JA,
Snyder PJ. An examination of the construct validity and
factor structure of the Groton Maze Learning Test, a new
measure of spatial working memory, learning efficiency,
and error monitoring. Arch Clin Neuropsychol. 2008;23:
433445.
Frank MJ, Loughry B, OReilly RC. Interactions between
frontal cortex and basal ganglia in working memory: a computational model. Cogn Affect Behav Neurosci. 2001;1:137160.
Pietrzak RH, Cohen H, Snyder PJ. Spatial learning efficiency and error monitoring in normal aging: an investigation using a novel hidden maze learning test. Arch Clin
Neuropsychol. 2007;22:235245.
Pietrzak RH, Maruff P, Mayes LC, Roman S, Sosa J, Snyder
PJ. Construct validity and factor structure of the groton maze
learning test: a new measure of spatial working memory,
learning efficiency, and error monitoring. under review.
Snyder AM, Maruff P, Pietrzak RH, Cromer JR, Snyder PJ.
Effect of treatment with Stimulant Medication on Nonverbal Executive Function and Visuomotor Speed in Children
with Attention Deficit/Hyperactivity Disorder (ADHD).
Child Neuropsychol. 2008;14(3):211226.
OReilly RC, Noelle DC, Braver TS, Cohen JD. Prefrontal
cortex and dynamic categorization tasks: representational
organization and neuromodulatory control. Cereb Cortex.
2002;12:246257.
Roberts AC, Robbins TW, Everitt BJ. The effects of intradimensional and extradimensional shifts on visual discrimination learning in humans and non-human primates. Q J
Exp Psychol B. 1988;40:321341.
Hampshire A, Owen AM. Fractionating attentional control
using event-related fMRI. Cereb Cortex. 2006;16:16791689.
Rogers RD, Andrews TC, Grasby PM, Brooks DJ, Robbins
TW. Contrasting cortical and subcortical activations produced by attentional-set shifting and reversal learning in
humans. J Cogn Neurosci. 2000;12:142162.
Dias R, Robbins TW, Roberts AC. Dissociation in prefrontal cortex of affective and attentional shifts. Nature.
1996;380:6972.
Chudasama Y, Robbins TW. Dissociable contributions of
the orbitofrontal and infralimbic cortex to pavlovian
autoshaping and discrimination reversal learning: further evidence for the functional heterogeneity of the rodent frontal
cortex. J Neurosci. 2003;23:87718780.
Clarke HF, Dalley JW, Crofts HS, Robbins TW, Roberts
AC. Cognitive inflexibility after prefrontal serotonin depletion. Science. 2004;304:878880.
Clarke HF, Walker SC, Crofts HS, Dalley JW, Robbins
TW, Roberts AC. Prefrontal serotonin depletion affects reversal learning but not attentional set shifting. J Neurosci.
2005;25:532538.
D. M. Barch et al.
132
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
predict selective impairments in perceptual attentional setshifting. J Neurosci. 2006;26:78707874.

Weed MR, Taffe MA, Polis I, et al. Performance norms for
a rhesus monkey neuropsychological testing battery: acquisition and long-term performance. Cogn Brain Res.
1999;8:185201.
Broberg BV, Dias R, Glenthoj BY, Olsen CK. Evaluation of
a neurodevelopmental model of schizophreniaearly postnatal PCP treatment in attentional set-shifting. Behav Brain
Res. 2008;190:160163.
Elliott R, McKenna PJ, Robbins TW, Sahakian BJ. Neuropsychological evidence for frontostriatal dysfunction in
schizophrenia. Psychol Med. 1995;25:619630.
Elliott R, Sahakian BJ, McKay AP, Herrod JJ, Robbins
TW, Paykel ES. Neuropsychological impairments in unipolar depression: the influence of perceived failure on subsequent performance. Psychol Med. 1996;26:975989.
Pantelis C, Barber FZ, Barnes TR, Nelson HE, Owen AM,
Robbins TW. Comparison of set-shifting ability in patients
with chronic schizophrenia and frontal lobe damage. Schizophr Res. 1999;37:251270.
Turner DC, Clark L, Pomarol-Clotet E, McKenna PJ,
Robbins S, Sahakian B. Modafinil improves cognition
and attentional set shifting in patients with chronic schizophrenia. 2004;29(7):13631373.
Hutton SB, Puri BK, Duncan LJ, Robbins TW, Barnes TR,
Joyce EM. Executive function in first-epsiode schizophrenia.
Psychol Med. 1998;28:463473.
Joyce E, Hutton S, Mutsatsa S, et al. Executive dysfunction
in first-episode schizophrenia and relationship to duration of
untreated psychosis: the West London Study. Br J Psychiatry Suppl. 2002;43:s38s44.
Jazbec S, Pantelis C, Robbins T, Weickert T, Weinberger
DR, Goldberg TE. Intra-dimensional/extra-dimensional
set-shifting performance in schizophrenia: impact of distractors. Schizophr Res. 2007;89:339349.
Robbins TW, James M, Owen AM, et al. A study of
performance on tests from the CANTAB battery sensitive
to frontal lobe dysfunction in a large sample of normal
volunteers: implications for theories of executive functioning and cognitive aging. J Int Neuropsycho Soc. 1998;
4:474490.
Lowe C, Rabbitt P. Test-re-test reliability of the CANTAB
and ISPOCD neuropsychological batteries: theoretical and
practical issues. Neuropsychologia. 1998;36:915923.
Stroop JR. Studies of interference in serial verbal reactions.
J Exp Psychol. 1935;18:643662.
MacLeod CM. Half a century of research on the Stroop
effect: an integrative review. Psychol Bull. 1991;109:163203.
Cohen JD, Barch DM, Carter C, Servan-Schreiber D.
Context-processing deficits in schizophrenia: converging
evidence from three theoretically motivated cognitive tasks.
J Abnorm Psychol. 1999;108:120133.
Barch DM, Carter CS, Cohen JD. Factors influencing
Stroop performance in schizophrenia. Neuropsychology.
2004;18:477484.
Cohen JD, Dunbar K, McClelland JL. On the control of automatic processes: a parallel distributed processing account
of the Stroop effect. Psychol Rev. 1990;97:332361.
Herd SA, Banich MT, OReilly RC. Neural mechanisms
of cognitive control: an integrative model of stroop task
performance and fMRI data. J Cogn Neurosci. 2006;18:
2232.
26. Clarke HF, Walker SC, Dalley JW, Robbins TW, Roberts
AC. Cognitive inflexibility after prefrontal serotonin depletion is behaviorally and neurochemically specific. Cereb Cortex. 2007;17:1827.
27. Chamberlain SR, Menzies L, Hampshire A, et al. Orbitofrontal dysfunction in patients with obsessive-compulsive
disorder and their unaffected relatives. Science. 2008;321:
421422.
28. Cools R, Clark L, Owen AM, Robbins TW. Defining the
neural mechanisms of probabilistic reversal learning using
event-related functional magnetic resonance imaging. J Neurosci. 2002;22:45634567.
29. Birrell JM, Brown VJ. Medial frontal cortex mediates perceptual attentional set shifting in the rat. J Neurosci.
2000;20:43204324.
30. Park SB, Coull JT, McShane RH, et al. Tryptophan depletion
in normal volunteers produces selective impairments in learning and memory. Neuropharmacology. 1994;33:575588.
31. Rogers RD, Blackshaw AJ, Middleton HC, et al. Tryptophan depletion impairs stimulus-reward learning while
methylphenidate disrupts attentional control in healthy
young adults: implications for the monoaminergic basis of
impulsive behaviour. Psychopharmacology (Berl). 1999;146:
482491.
32. Middleton HC, Sharma A, Agouzoul D, Sahakian BJ,
Robbins TW. Idazoxan potentiates rather than antagonizes
some of the cognitive effects of clonidine. Psychopharmacology (Berl). 1999;145:401411.
33. Mehta MA, Manes FF, Magnolfi G, Sahakian BJ, Robbins
TW. Impaired set-shifting and dissociable effects on tests of
spatial working memory following the dopamine D2 receptor antagonist sulpiride in human volunteers. Psychopharmacology (Berl). 2004;176:331342.
34. Lange KW, Paul GM, Robbins TW, Marsden CD. L-Dopa
and frontal cognitive function in Parkinsons disease. Adv
Neurol. 1993;60:475478.
35. Boulougouris V, Glennon JC, Robbins TW. Dissociable
effects of selective 5-HT(2A) and 5-HT(2C) receptor antagonists on serial spatial reversal learning in rats. Neuropsychopharmacology. 2007;33:20072019.
36. Roberts AC, De Salvia MA, Wilkinson LS, et al. 6-Hydroxydopamine lesions of the prefrontal cortex in monkeys enhance performance on an analog of the Wisconsin Card
Sort Test: possible interactions with subcortical dopamine.
J Neurosci. 1994;14:25312544.
37. Crofts HS, Dalley JW, Collins P, et al. Differential effects of
6-OHDA lesions of the frontal cortex and caudate nucleus
on the ability to acquire an attentional set. Cereb Cortex.
2001;11:10151026.
38. Lapiz MD, Bondi CO, Morilak DA. Chronic treatment with
desipramine improves cognitive performance of rats in an attentional set-shifting test. Neuropsychopharmacology.
2007;32:10001010.
39. Tait DS, Brown VJ, Farovik A, Theobald DE, Dalley JW,
Robbins TW. Lesions of the dorsal noradrenergic bundle
impair attentional set-shifting in the rat. Eur J Neurosci.
2007;25:37193724.
40. Tunbridge EM, Bannerman DM, Sharp T, Harrison PJ.
Catechol-o-methyltransferase inhibition improves setshifting performance and elevates stimulated dopamine release in the rat prefrontal cortex. J Neurosci. 2004;24:
53315335.
41. Liston C, Miller MM, Goldwater DS, et al. Stress-induced
alterations in prefrontal cortical dendritic morphology
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
patients during a Stroop task. Neuropsychopharmacology.

2008.
Gratton G, Coles MG, Donchin E. Optimizing the use of information: strategic control of activation of responses. J Exp
Psycho Gen. 1992;121:480506.
Rabbitt PMA. Errors and error correction in choiceresponse tasks. J Exp Psychol. 2008;33(11):26192625.
Laming D. Choice reaction performance following an error.
Acta Psychol. 1979;43:199224.
Botvinick MM, Braver TS, Carter CS, Barch DM, Cohen
JC. Evaluating the demand for control: Anterior cingulate
cortex and conflict monitoring. Report nr Technical Report
98.1. Pittsburgh, PA: Center for the Neural Basis of Cognition; 1998.
Fan J, McCandliss BD, Fossella J, Flombaum JI, Posner
MI. The activation of attentional networks. Neuroimage.
2005;26:471479.
Fan J, Flombaum JI, McCandliss BD, Thomas KM, Posner
MI. Cognitive and brain consequences of conflict. Neuroimage. 2003;18:4257.
Fossella J, Sommer T, Fan J, et al. Assessing the
molecular genetics of attention networks. BMC Neurosci.
2002;3:14.
Fan J, Fossella J, Sommer T, Wu Y, Posner MI. Mapping
the genetic variation of executive attention onto brain activity. Proc Natl Acad Sci U S A. 2003;100:74067411.
Wang K, Fan J, Dong Y, Wang CQ, Lee TM, Posner MI.
Selective impairment of attentional networks of orienting
and executive control in schizophrenia. Schizophr Res. 2005;
78:235241.
Fan J, McCandliss BD, Sommer T, Raz A, Posner MI. Testing the efficiency and independence of attentional networks.
J Cogn Neurosci. 2002;14:340347.
Ridderinkhof KR. Micro- and macro-adjustments of task
set: activation and suppression in conflict tasks. Psychol
Res. 2002;66:312323.
Kerns JG. Anterior cingulate and prefrontal cortex activity
in an FMRI study of trial-to-trial adjustments on the Simon
task. Neuroimage. 2006;33:399405.
Sturmer B, Leuthold H, Soetens E, Schroter H, Sommer W.
Control over location-based response activation in the
Simon task: behavioral and electrophysiological evidence.
J Exp Psychol Hum Percept Perform. 2002;28:13451363.
Kerns JG, Cohen JD, MacDonald AW, III, Cho RY,
Stenger VA, Carter CS. Anterior cingulate conflict monitoring and adjustments in control. Science. 2004;303:
10231026.
Carter CS, Macdonald AM, Botvinick M, et al. Parsing
executive processes: strategic versus evaluative functions of
the anterior cingulate cortex. Proc Natl Acad Sci. 2000;97:
19441948.
Egner T, Hirsch J. The neural correlates and functional integration of cognitive control in a Stroop task. Neuroimage.
2005;24:539547.
Egner T, Delano M, Hirsch J. Separate conflict-specific cognitive control mechanisms in the human brain. Neuroimage.
2007;35:940948.
Botvinick MM, Braver TS, Barch DM, Carter CS, Cohen
JC. Conflict monitoring and cognitive control. Psychol
Rev. 2001;108:624652.
Brown JW, Braver TS. Learned predictions of error likelihood in the anterior cingulate cortex. Science.
2005;307:11181121.
133
59. Dehaene S, Kerszberg M, Changeux JP. A neuronal model

of a global workspace in effort cognitive tasks. Proc Natl
Acad Sci U S A. 1998;95:1452914534.
60. Kane MJ, Engle RW. Working-memory capacity and the control of attention: the contributions of goal neglect, response
competition, and task set to Stroop interference. J Exp Psychol
Gen. 2003;132:4770.
61. MacDonald AW, III, Cohen JD, Stenger VA, Carter CS.
Dissociating the role of the dorsolateral prefrontal and anterior cingulate cortex in cognitive control. Science. 2000;
288:18351838.
62. Donohue SE, Wendelken C, Bunge SA. Neural correlates of
preparation for action select as a function of specific task
demands. J Cogn Neurosci. 2008;20:694706.
63. Demakis GJ. Frontal lobe damage and tests of executive
processing: a meta-analysis of the category test, Stroop test
and trail-making test. J Clin Exp Neuropsychol. 2004;26:
441450.
64. Barch DM, Carter CS. Amphetamine improves cognitive
function in medicated individuals with schizophrenia and
in healthy volunteers. Schizophr Res. 2005;77:4358.
65. Henik A, Salo R. Schizophrenia and the stroop effect. Behav
Cogn Neurosci Rev. 2004;3:4259.
66. Haddon JE, George DN, Killcross AS. Contextual control
of biconditional task performance: evidence for cue and response competition in rats. Q J Exp Psychol (Colchester).
2007;61:.
67. Haddon JE, Killcross S. Contextual control of choice performance: behavioral, neurobiological, and neurochemical
influences. Ann N Y Acad Sci. 2007;1104:250269.
68. Marquis JP, Killcross S, Haddon JE. Inactivation of the prelimbic, but not infralimbic, prefrontal cortex impairs the
contextual control of response conflict in rats. Eur J Neurosci. 2007;25:559566.
69. Stoet G, Snyder LH. Correlates of stimulus-response congruence in the posterior parietal cortex. J Cogn Neurosci.
2007;19:194203.
70. Stoet G, Snyder LH. Effects of the NMDA antagonist ketamine on task-switching performance: evidence for specific
impairments of executive control. Neuropsychopharmacology. 2006;31:16751681.
71. Kerns JG, Cohen JD, MacDonald AW, III, et al. Decreased
conflict- and error-related activity in the anterior cingulate
cortex in subjects with schizophrenia. Am J Psychiatry.
2005;162:18331839.
72. Weiss EM, Siedentopf C, Golaszewski S, et al. Brain activation patterns during a selective attention testa functional
MRI study in healthy volunteers and unmedicated patients
during an acute episode of schizophrenia. Psychiatry Res.
2007;154:3140.
73. Yucel M, Pantelis C, Stuart GW, et al. Anterior cingulate
activation during Stroop task performance: a PET to MRI
coregistration study of individual patients with schizophrenia. Am J Psychiatry. 2002;159:251254.
74. Fields RB, Van Kammen DP, Peters JL, et al. Clonidine
improves memory function in schizophrenia independently
from change in psychosis. Schizoph Res. 1988;1:417423.
75. Snitz BE, Macdonald AW, III, Carter CS. Cognitive deficits
in unaffected first-degree relatives of schizophrenia patients:
a meta-analytic review of putative endophenotypes. Schizophr Bull. 2006;32:179194.
76. Becker TM, Kerns JG, Macdonald AW, III, Carter CS. Prefrontal dysfunction in first-degree relatives of schizophrenia
D. M. Barch et al.
134
116. Tieges Z, Richard Ridderinkhof K, Snel J, Kok A. Caffeine

strengthens action monitoring: evidence from the errorrelated negativity. Brain Res Cogn Brain Res. 2004;21:8793.
117. Zirnheld PJ, Carroll CA, Kieffaber PD, ODonnell BF,
Shekhar A, Hetrick WP. Haloperidol impairs learning and
error-related negativity in humans. J Cogn Neurosci.
2004;16:10981112.
118. de Bruijn ER, Sabbe BG, Hulstijn W, Ruigt GS, Verkes RJ.
Effects of antipsychotic and antidepressant drugs on action
monitoring in healthy volunteers. Brain Res. 2006;1105:122129.
119. Amiez C, Joseph JP, Procyk E. Anterior cingulate errorrelated activity is modulated by predicted reward. Eur J
Neurosci. 2005;21:34473452.
120. Schall JD, Stuphorn V, Brown JW. Monitoring and control
of action by the frontal lobes. Neuron. 2002;36:309322.
121. Emeric EE, Brown JW, Leslie M, Pouget P, Stuphorn V,
Schall JD. Performance monitoring local field potentials in
the medial frontal cortex of primates: anterior cingulate cortex. J Neurophysiol. 2008;99:759772.
122. Olson CR, Gettner SN. Neuronal activity related to rule and
conflict in macaque supplementary eye field. Physiol Behav.
2002;77:663670.
123. Nakamura K, Roesch MR, Olson CR. Neuronal activity in
macaque SEF and ACC during performance of tasks involving conflict. J Neurophysiol. 2005;93:884908.
124. Mansouri FA, Buckley MJ, Tanaka K. Mnemonic function
of the dorsolateral prefrontal cortex in conflict-induced behavioral adjustment. Science. 2007;318:987990.
125. Narayanan NS, Laubach M. Neuronal correlates of posterror slowing in the rat dorsomedial prefrontal cortex.
J Neurophysiol. 2008;100:520525.
126. Alain C, McNeely HE, Yu H, Christensen BK, West R.
Neurophysiological evidence of error monitoring deficits in
patients with schizophrenia. Cereb Cortex. 2002;12:840846.
127. Kopp B, Rist F. An event-related brain potential substrate
of disturbed response monitoring in paranoid schizophrenic
patients. J Abnorm Psychol. 1999;108:337346.
128. Mathalon DH, Dedor M, Faustman WO, Gray M, Askari
N, Ford JM. Response-monitoring dysfunction in schizophrenia: an event-related brain potential study. J Abnorm
Psychol. 2002;111:2241.
129. Morris SE, Heerey EA, Gold JM, Holroyd CB. Learningrelated changes in brain activity following errors and performance feedback in schizophrenia. Schizophr Res. 2008;99:
274285.
130. Morris SE, Yee CM, Nuechterlein KH. Electrophysiological
analysis of error monitoring in schizophrenia. J Abnorm
Psychol. 2006;115:239250.
131. Laurens KR, Ngan ETC, Bates AT, Kiehl KA, Liddle PF.
Rostral anterior cingulate cortex dysfunction during error
processing in schizophrenia. Brain. 2003;126:610622.
132. Polli FE, Barton JJ, Thakkar KN, et al. Reduced errorrelated activation in two anterior cingulate circuits is related
to impaired performance in schizophrenia. Brain. 2008;131:
971986.
133. Polli FE, Barton JJ, Vangel M, Goff DC, Iguchi L,
Manoach DS. Schizophrenia patients show intact immediate
error-related performance adjustments on an antisaccade
task. Schizophr Res. 2006;82:191201.
134. Logan G, Cowan N. On the ability to inhibit thought and
action: a theory of an act of control. Psychol Rev. 1984;91:
295327.
96. Brown JW, Reynolds JR, Braver TS. A computational

model of fractionated conflict-control mechanisms in taskswitching. Cognit Psychol. 2007;55:3785.
97. Brown JW, Braver TS. A computational model of risk, conflict, and individual difference effects in the anterior cingulate cortex. Brain Res. 2008;1202:99108.
98. Mayr U, Awh E, Laurey P. Conflict adaptation effects in the
absence of executive control. Nat Neurosci. 2003;6:450452.
99. Ullsperger M, Bylsma LM, Botvinick MM. The conflict adaptation effect: its not just priming. Cogn Affect Behav Neurosci. 2005;5:467472.
100. Notebaert W, Gevers W, Verbruggen F, Liefooghe B. Topdown and bottom-up sequential modulations of congruency
effects. Psychon Bull Rev. 2006;13:112117.
101. Botvinick MM, Nystrom L, Fissel K, Carter CS, Cohen JD.
Conflict monitoring versus selection-for-action in anterior
cingulate cortex. Nature. 1999;402:179181.
102. Holroyd CB, Coles MG. The neural basis of human error
processing: reinforcement learning, dopamine, and the
error-related negativity. Psychol Rev. 2002;109:679709.
103. Gehring WJ, Goss B, Coles MGH, Meyer DE, Donchin E.
A neural system for error detection and compensation. Psychol Sci. 1993;4:385390.
104. Debener S, Ullsperger M, Siegel M, Fiehler K, von Cramon
DY, Engel AK. Trial-by-trial coupling of concurrent electroencephalogram and functional magnetic resonance imaging identifies the dynamics of performance monitoring.
J Neurosci. 2005;25:1173011737.
105. Egner T, Hirsch J. Cognitive control mechanisms resolve
conflict through cortical amplification of task-relevant information. Nat Neurosci. 2005;8:17841790.
106. Marco-Pallares J, Camara E, Munte TF, RodriguezFornells A. Neural mechanisms underlying adaptive actions
after slips. J Cogn Neurosci. 2008;20:15951610.
107. Fellows LK, Farah MJ. Is anterior cingulate cortex necessary for cognitive control? Brain. 2005;128:788796.
108. di Pellegrino G, Ciaramelli E, Ladavas E. The regulation of
cognitive control following rostral anterior cingulate cortex
lesion in humans. J Cogn Neurosci. 2007;19:275286.
109. Gehring WJ, Knight RT. Prefrontal-cingulate interactions in
action monitoring. Nat Neurosci. 2000;3:516520.
110. Burle B, Roger C, Allain S, Vidal F, Hasbroucq T. Error
negativity does not reflect conflict: a reappraisal of conflict
monitoring and anterior cingulate cortex activity. J Cogn
Neurosci. 2008;20:16371655.
111. Botvinick MM. Conflict monitoring and decision making:
reconciling two perspectives on anterior cingulate function.
Cogn Affect Behav Neurosci. 2007;7:356366.
112. Carter CS, van Veen V. Anterior cingulate cortex and conflict detection: an update of theory and data. Cogn Affect
Behav Neurosci. 2007;7:367379.
113. Riba J, Rodriguez-Fornells A, Munte TF, Barbanoj MJ. A
neurophysiological study of the detrimental effects of alprazolam on human action monitoring. Brain Res Cogn Brain
Res. 2005;25:554565.
114. de Bruijn ER, Hulstijn W, Verkes RJ, Ruigt GS, Sabbe BG.
Drug-induced stimulation and suppression of action monitoring in healthy volunteers. Psychopharmacology (Berl).
2004;177:151160.
115. Riba J, Rodriguez-Fornells A, Morte A, Munte TF,
Barbanoj MJ. Noradrenergic stimulation enhances human
action monitoring. J Neurosci. 2005;25:43704374.
152. Nambu A, Tokuno H, Takada M. Functional significance of

the cortico-subthalamo-pallidal hyperdirect pathway. Neurosci Res. 2002;43:111117.
153. Tannock R, Schachar RJ, Carr RP, Chajczyk D, Logan
GD. Effects of methylphenidate on inhibitory control in
hyperactive children. J Abnorm Child Psychol. 1989;17
:473491.
154. Aron AR, Dowson JH, Sahakian BJ, Robbins TW. Methylphenidate improves response inhibition in adults with attention-deficit/hyperactivity disorder. Biol Psychiatry. 2003;54:
14651468.
155. Fillmore MT, Rush CR, Hays L. Acute effects of cocaine in
two models of inhibitory control: implications of non-linear
dose effects. Addiction. 2006;101:13231332.
156. Chamberlain SR, Muller U, Blackwell AD, Clark L,
Robbins TW, Sahakian BJ. Neurochemical modulation of
response inhibition and probabilistic learning in humans.
Science. 2006;311:861863.
157. Overtoom CC, Verbaten MN, Kemner C, et al. Effects of
methylphenidate, desipramine, and L-dopa on attention
and inhibition in children with Attention Deficit Hyperactivity Disorder. Behav Brain Res. 2003;145:715.
158. Congdon E, Lesch KP, Canli T. Analysis of DRD4 and
DAT polymorphisms and behavioral inhibition in healthy
adults: implications for impulsivity. Am J Med Genet B Neuropsychiatr Genet. 2008;147B:2732.
159. Eagle DM, Robbins TW. Inhibitory control in rats performing a stop-signal reaction-time task: effects of lesions of the
medial striatum and d-amphetamine. Behav Neurosci.
2003;117:13021317.
160. Eagle DM, Robbins TW. Lesions of the medial prefrontal
cortex or nucleus accumbens core do not impair inhibitory
control in rats performing a stop-signal reaction time task.
Behav Brain Res. 2003;146:131144.
161. Eagle DM, Baunez C, Hutcheson DM, Lehmann O, Shah
AP, Robbins TW. Stop-signal reaction-time task performance: role of prefrontal cortex and subthalamic nucleus.
Cereb Cortex. 2008;18:178188.
162. Eagle DM, Tufft MR, Goodchild HL, Robbins TW. Differential effects of modafinil and methylphenidate on stopsignal reaction time task performance in the rat, and
interactions with the dopamine receptor antagonist cisflupenthixol. Psychopharmacology (Berl). 2007;192:193206.
163. Badcock JC, Michie PT, Johnson L, Combrinck J. Acts of
control in schizophrenia: dissociating the components of inhibition. Psychol Med. 2002;32:287297.
164. Vink M, Ramsey NF, Raemaekers M, Kahn RS. Striatal
dysfunction in schizophrenia and unaffected relatives. Biol
Psychiatry. 2006;60:3239.
165. Bellgrove MA, Chambers CD, Vance A, Hall N, Karamitsios M, Bradshaw JL. Lateralized deficit of response inhibition in early-onset schizophrenia. Psychol Med. 2006;36:
495505.
166. Enticott PG, Ogloff JR, Bradshaw JL. Response inhibition
and impulsivity in schizophrenia. Psychiatry Res. 2008;
15:157(13):2514.
135
135. Boucher L, Palmeri TJ, Logan GD, Schall JD. Inhibitory

control in mind and brain: an interactive race model of
countermanding saccades. Psychol Rev. 2007;114:376397.
136. Band GP, van der Molen MW, Logan GD. Horse-race
model simulations of the stop-signal procedure. Acta Psychol (Amst). 2003;112:105142.
137. Logan GD, Schachar RJ, Tannock R. Impulsivity and inhibitory control. Psychol Sci. 1997;8:6064.
138. Lijffijt M, Kenemans JL, Verbaten MN, van Engeland H. A
meta-analytic review of stopping performance in attentiondeficit/hyperactivity disorder: deficient inhibitory motor
control? J Abnorm Psychol. 2005;114:216222.
139. Chamberlain SR, Fineberg NA, Blackwell AD, Robbins
TW, Sahakian BJ. Motor inhibition and cognitive flexibility
in obsessive-compulsive disorder and trichotillomania. Am J
Psychiatry. 2006;163:12821284.
140. Monterosso JR, Aron AR, Cordova X, Xu J, London ED.
Deficits in response inhibition associated with chronic methamphetamine abuse. Drug Alcohol Depend. 2005;79:273277.
141. Fillmore MT, Rush CR. Impaired inhibitory control of behavior in chronic cocaine users. Drug Alcohol Depend. 2002;
66:265273.
142. Colzato LS, van den Wildenberg WP, Hommel B. Impaired
inhibitory control in recreational cocaine users. PLoS ONE.
2007;2:e1143.
143. Aron AR, Fletcher PC, Bullmore ET, Sahakian BJ, Robbins
TW. Stop-signal inhibition disrupted by damage to right inferior frontal gyrus in humans. Nat Neurosci. 2003;6:115116.
144. Floden D, Stuss DT. Inhibitory control is slowed in patients
with right superior medial frontal damage. J Cogn Neurosci.
2006;18:18431849.
145. Nachev P, Wydell H, ONeill K, Husain M, Kennard C. The
role of the pre-supplementary motor area in the control of
action. Neuroimage. 2007;36(suppl. 2):T155T163.
146. Aron AR, Poldrack RA. Cortical and subcortical contributions to Stop signal response inhibition: role of the subthalamic nucleus. J Neurosci. 2006;26:24242433.
147. Aron AR, Behrens TE, Smith S, Frank MJ, Poldrack RA.
Triangulating a cognitive control network using diffusionweighted magnetic resonance imaging (MRI) and functional
MRI. J Neurosci. 2007;27:37433752.
148. Chevrier AD, Noseworthy MD, Schachar R. Dissociation of
response inhibition and performance monitoring in the stop
signal task using event-related fMRI. Hum Brain Mapp.
2007;28:13471358.
149. Rubia K, Smith AB, Brammer MJ, Taylor E. Right inferior
prefrontal cortex mediates response inhibition while mesial
prefrontal cortex is responsible for error detection. Neuroimage. 2003;20:351358.
150. van den Wildenberg WP, van Boxtel GJ, van der Molen
MW, Bosch DA, Speelman JD, Brunia CH. Stimulation of
the subthalamic region facilitates the selection and inhibition
of motor responses in Parkinsons disease. J Cogn Neurosci.
2006;18:626636.
151. Hershey T, Revilla FJ, Wernle AR, et al. Cortical and subcortical blood flow effects of subthalamic nucleus stimulation in PD. Neurology. 2003;61:816821.

CNTRICS Final Task Selection: Executive Control

Uploaded by

Copyright:

Available Formats

CNTRICS Final Task Selection: Executive Control

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CNTRICS Final Task Selection: Executive Control

Uploaded by

Copyright:

Available Formats

Schizophrenia Bulletin vol. 35 no. 1 pp.

CNTRICS Final Task Selection: Executive Control

Deanna M. Barch1,2,4,8, Todd S. Braver2, Cameron

Department of Psychology; 3Department of Psychiatry and;

Co-authors had equivalent input and are listed in alphabetical

To whom correspondence should be addressed; Department of

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

phrenia show deficits on tasks designed to measure such

Rule Generation and Selection

that the 1-2 AX-CPT was an interesting and promising

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

the availability of animal models. The overview article at

CNTRICS Final Task Selection

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

CNTRICS Final Task Selection

(posterior parietal, temporal) areas. The neuroanatomical

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

In humans, there is evidence for double dissociation in

The Switching Stroop

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

at the extradimensional shift stage.51 There are limited

CNTRICS Final Task Selection

responses (eg, the word red written in blue). In addition,

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

Pharmacological and Behavioral Manipulation of Task

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

cue-related anterior cingulate (ACC) activity did not vary

CNTRICS Final Task Selection

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

studies examining test-retest reliability, practice effects,

references a differential deficit rather than a generalized

Tasks Recommended for Immediate Translation for

following an incongruent (iI) vs incongruent following

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

CNTRICS Final Task Selection

Stop signal task

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

Conflict and Error Adaptation in the Stroop Task

primate work. Much of this literature suggests that the

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

CNTRICS Final Task Selection

magnitude of the ERN and error-related behaviors is intact

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

Yohimbine does lead to fewer errors following errors,

STOP TRIAL (with a relatively short Stop Signal Delay)

STOP TRIAL (with a relatively long Stop Signal Delay)

portant to highlight that this task is highly sensitive to the

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

CNTRICS Final Task Selection

Pharmacological or Behavioral Manipulation

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

STN is activated for successful stop vs go trials and that its

ability and neural systems validity. Relatively little

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

patients with schizophrenia, as well as subjects with

CNTRICS Final Task Selection

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on May 11, 2012

We would like to thank Jody Conrad, Carol Cox, and Deb