The need to select information among competing al-
ternatives is ubiquitous. Oftentimes, successful cognition
depends on the ability to focus resources on goal-relevant
information while filtering out or inhibiting irrelevant infor-
mation. How selective attention operates and whether and
how irrelevant information is inhibited or otherwise filtered
out has been a major focus of research since the inception
of experimental psychology. For the past 15 years, cogni-
tive neuroscientists have used neuroimaging to uncover the
brain mechanisms underlying the processes responsible for
handling irrelevant information. Much of this research has
used variants of classic cognitive interference resolution
tasks, each different in its superficial characteristics but
sharing the common requirement to resolve conflict. What
have we learned from this large corpus of data?
Examining the multitude of studies focusing on interfer-
ence resolution tells an extremely varied story. Figure 1A
shows a plot of the peaks of activation of 47 studies that
purport to examine interference resolution (see the stud-
ies listed in Table 1). Ostensibly, there appears to be little
consistency in these data. Several factors may be contrib-
uting to the massive interstudy variance. First, Figure 1A
includes activations from different tasks, subjects, equip-
ment, scanning parameters, and statistical analyses. If we
constrain our focus to just one task, however, the activa-
tions do not appear to be much more consistent. Figure 1B
shows the activations arising just from the Stroop task
(Stroop, 1935), and these do not appear any more orderly.
Indeed, the variability among the reported peaks across
all interference resolution tasks corroborates behavioral
findings that correlations in performance among different
interference resolution tasks are low (Kramer, Humphrey,
Larish, Logan, & Strayer, 1994; Shilling, Chetwynd, &
Rabbitt, 2002). Indeed, even simple changes in task pa-
rameters appear to produce very different results (e.g., de
Zubicaray, Andrew, Zelaya, Williams, & Dumanoir, 2000;
MacLeod, 1991). It seems clear that understanding inter-
ference resolution will take deeper analytic methods that
interrogate possible strategic and mechanistic differences.
Some researchers have attempted to examine the neural
signatures of various interference resolution tasks within
the same subjects to uncover whether any consistency
can be found (Fan, Flombaum, McCandliss, Thomas, &
Posner, 2003; Liu, Banich, Jacobson, & Tanabe, 2004;
Peterson et al., 2002; Wager et al., 2005). These efforts
have revealed that activations in different tasks overlap in
a number of regions but that there are also regions unique
to one task or another. What underlies these commonali-
ties and differences?
At this point, there have been a sufficient number of
studies of interference resolution to begin to answer these
questions. Here, we will attempt to sift through the inter-
Copyright 2007 Psychonomic Society, Inc.
Interference resolution: Insights from a
meta-analysis of neuroimaging tasks
Derek evan nee
University of Michigan, Ann Arbor, Michigan
Tor D. Wager
Columbia University, New York, New York
University of Michigan, Ann Arbor, Michigan
A quantitative meta-analysis was performed on 47 neuroimaging studies involving tasks purported to require
the resolution of interference. The tasks included the Stroop, flanker, go/no-go, stimulus–response compatibil-
ity, Simon, and stop signal tasks. Peak density-based analyses of these combined tasks reveal that the anterior
cingulate cortex, dorsolateral prefrontal cortex, inferior frontal gyrus, posterior parietal cortex, and anterior
insula may be important sites for the detection and/or resolution of interference. Individual task analyses reveal
differential patterns of activation among the tasks. We propose that the drawing of distinctions among the pro-
cessing stages at which interference may be resolved may explain regional activation differences. Our analyses
suggest that resolution processes acting upon stimulus encoding, response selection, and response execution
may recruit different neural regions.
Cognitive, Affective, & Behavioral Neuroscience
2007, 7 (1), 1-17
D. E. Nee, firstname.lastname@example.org
2 Nee, Wager, aNd JoNides
study variance in the interference resolution literature and
pick out the consistencies among studies and tasks. In ad-
dition to trying to uncover the neural basis of interference
resolution, we shall also consider why variations in tasks
and task parameters may lead to separable patterns of neu-
ral activation. Although the meta-analytic methods used
here preclude us from drawing strong conclusions about
interference resolution (because they rely on reported peak
coordinates from previous studies), they allow us to begin
to form hypotheses that further investigations can either
confirm or deny (e.g., Fox, Laird, & Lancaster, 2005).
For our analyses, we included six tasks that have been
prominent in the interference resolution literature: the go/
no-go task, flanker task, Stroop task, stimulus–response
compatibility (SRC) task, Simon task, and stop signal task
(all described below). Studies were included only if they
reported peaks of activation in standardized coordinate
space (Talairach or MNI). Notably absent are tasks used
to examine the resolution of proactive interference (e.g.,
Jonides, Smith, Marshuetz, Koeppe, & Reuter-Lorenz,
1998), since a review of these data has already been pub-
lished (Jonides & Nee, 2006). Furthermore, we do not
include the antisaccade task in this mix, because models
of this task are already at the single-unit level and our
coarse techniques of analysis would be unable to inform
this literature further (Munoz & Everling, 2004). We in-
cluded neuroimaging studies in which either PET or fMRI
was used between 1990 and 2005 and in which normal,
healthy, young adults were examined.2 Although we rec-
ognize that there may be differences between blocked
and event-related designs in terms of neural activations,
there were insufficient studies to examine each separately.
Therefore, we have combined both types of designs in
our analyses. Forty-seven studies met our criteria and are
listed in Table 1. When possible, we restricted our analy-
ses to correct trials only.
Go/no-go. In the go/no-go task, subjects are required to
respond to one stimulus (e.g., the letter “Y”) but to withhold
a response to another stimulus (“X”). Responses are labeled
go trials, whereas trials on which a response is to be withheld
are called no-go trials. It has been argued that as the number
of go trials preceding a no-go trial increases, a greater pre-
potent tendency to respond is formed (de Zubicaray et al.,
2000; Durston, Thomas, Worden, Yang, & Casey, 2002;
Durston, Thomas, Yang, et al., 2002; Rubia et al., 2001).
This prepotent response must be resolved in order to per-
form properly on no-go trials. Our analyses included con-
trasts of no-go versus go responses.
Flanker. The flanker task requires a subject to attend to
a centrally fixated stimulus while ignoring flanking stimuli
(Eriksen & Eriksen, 1974). In a paradigmatic case, the cen-
tral stimulus can be a letter (e.g., “H”), which subjects learn
to associate with a given response (say, a left keypress).
Flanking stimuli can be of three types. First, the flankers
can be identical to the imperative stimulus. In this case, both
the relevant and the irrelevant stimuli are consistent (HHH).
We will refer to this trial type as identical. Flankers can also
be different from the central stimulus (“S,” for instance),
but the participants are instructed to map these stimuli onto
the same (say, left) response as the target stimulus (SHS).
This trial type is called congruent. Finally, stimuli can dif-
fer not only in form from the relevant stimulus, but also in
response pairing (“G” mapped onto a right keypress). This
is what we call an incongruent trial (GHG). Thus, on identi-
cal trials, no conflict is present. On congruent trials, there is
stimulus conflict, but not response conflict, and on incon-
gruent trials, there is stimulus, as well as response, conflict
(Kornblum, Stevens, Whipple, & Requin, 1999; van Veen,
Cohen, Botvinick, Stenger, & Carter, 2001; Zhang, Zhang,
& Kornblum, 1999). Our analyses included contrasts of
both incongruent versus congruent responses and incon-
gruent versus identical responses. There were insufficient
studies to tease these two contrasts apart.
Stimulus–response compatibility. In the SRC
paradigm, a subject is required to switch between two
Figure 1. (A) Peaks from the 47 studies included in the meta-analysis, plotted in a single
brain. (B) Peaks from the studies in which the Stroop task was used.1
iNterfereNce resolutioN 3
stimulus–response mappings. One mapping, referred to
as compatible, is directly suggested by the stimulus. For
example, a typical SRC task might employ arrows as stim-
uli, in which case a compatible mapping might be a left
keypress to an arrow pointing left and a right keypress to
an arrow pointing right. An incompatible mapping would
require a left keypress to a rightward-pointing arrow and
a right keypress to a leftward-pointing arrow. Thus, in the
Studies Included in the Meta-Analysis Catalogued by Which Tasks They Included
Adleman et al.
Banich et al.
Banich et al.
Bench et al.
Bunge et al.
Carter et al.
Casey et al.
Dassonville et al.
Derbyshire et al.
Durston, Thomas, Worden, et al.
Durston, Thomas, Yang, et al.
Fan et al.
Garavan et al.
Garavan et al.
Hazeltine et al.
Iacoboni et al.
Iacoboni et al.
Kiehl et al.
Konishi et al.
Konishi et al.
Leung et al.
Liddle et al.
Liu et al.
Maclin et al.
Menon et al.
Milham et al.
Milham et al.
Milham et al.
Milham & Banich
Pardo et al.
Paus et al.
Perlstein et al.
Peterson et al.
Peterson et al.
Ravnkilde et al.
Rubia et al.
Ruff et al.
Schumacher & D’Esposito
Sylvester et al.
Tamm et al.
Taylor et al.
Taylor et al.
Ullsperger & von Cramon
van Veen et al.
Wager et al.
Watanabe et al.
Zysset et al.
Note—Contrasts that were reported in the study are indicated in the table (I–C, incongruent–congruent or incompatible–compatible;
I–N, incongruent–neutral; I–Id, incongruent–identical). There were a total of 6 flanker studies, contributing 79 peaks; 14 go/no-go studies,
contributing 139 peaks; 9 SRC studies, contributing 62 peaks; 12 Stroop (I–N) studies, contributing 158 peaks; 11 Stroop (I–C) studies,
contributing 190 peaks; 4 Simon studies, contributing 64 peaks; and 1 stop signal study, contributing 6 peaks.
4 Nee, Wager, aNd JoNides
incompatible condition, a prepotent response that is sug-
gested by the stimulus and developed by previous compat-
ible responses must be overcome. Our analyses included
incompatible minus compatible contrasts.
Stroop. In the Stroop task, subjects must identify the
hue in which a word is printed while ignoring the refer-
ent of the word. There are three basic types of trials in a
typical Stroop task: incongruent, congruent, and neutral.
On congruent trials, both the color of the word and the
word’s referent elicit the same response (e.g., the word
“red” printed in red ink). On incongruent trials, the color
and referent of the word elicit different responses (the
word “green” printed in red ink). Neutral trials may be of
several types, but for all neutral trials, the referent of the
stimulus does not provide a competing response to the hue
(e.g., a series of Xs printed in red, or the word “lot” printed
in red). Our analyses included both incongruent minus
congruent and incongruent minus neutral contrasts.
Simon. The Simon task is similar to the Stroop task,
except that the irrelevant stimulus dimension is spatial.
For example, in a paradigmatic Simon task, a relevant
stimulus is presented at various spatial locations. The
stimulus (say, a colored circle) might appear either to the
right or to the left of fixation. The circle is mapped onto
a left or a right response (e.g., red–left, blue–right), and
subjects must respond to the stimulus while ignoring the
potentially distracting spatial placement of the stimulus. It
has been found that reaction times are longer when the lo-
cation of the stimulus is incompatible with the response it
elicits (a red circle presented to the right of fixation) than
when the location is compatible with the response, due to
the resolution of interference caused by the irrelevant spa-
tial dimension of the stimulus. We included incompatible
minus compatible contrasts in our analyses.
Stop signal. The stop signal task requires a subject to
cease executing a readied response. In a typical stop signal
task, a subject is required to respond to a stimulus but to
withhold the response if a tone is heard. Varying the onset
of the tone, relative to the response, can affect the error
rates (responses not withheld) and, thus, the demands on
conflict resolution processes. Our analyses include stop
versus go responses.
We used a data-driven approach to discovering which
regions of the brain were most consistently reported in the
corpus of studies. To this end, we employed a density analy-
sis technique, which has been successfully used in other
meta-analyses (Wager, Jonides, & Reading, 2004; Wager,
Phan, Liberzon, & Taylor, 2003) and is similar to other
voxel-based methods (Fox et al., 2005; Laird et al., 2005).
The density technique is similar to the activation likelihood
estimate (ALE) method used in some other meta-analyses
(Turkeltaub, Eden, Jones, & Zeffiro, 2002), with one dis-
tinction. The density technique examines the spatial con-
sistency among reported peaks and locates brain voxels in
which the density of reported peaks exceeds what would be
expected by chance. The ALE method assesses the prob-
ability that at least one activation peak fell within that voxel
by assessing the union of probability values across individ-
ual peaks. Although the methods give very similar results,
we tested the null hypothesis that the spatial distribution of
peaks is random, whereas the ALE method tests the null hy-
pothesis that in no studies was a particular voxel activated.
The density analysis was conducted as follows. We first
converted all Talairach peaks into MNI space, in order to
have all the data mapped into a common stereotactic space
(www.mrc-cbu.cam.ac.uk/Imaging/). Next, we plotted all
of the peaks reported in each study onto a canonical brain
(avg152T1.img; SPM, Wellcome Department of Imaging
Neuroscience, www.fil.ion.ucl.ac.uk/spm/). We included
only positive activations, since deactivations are incon-
sistently reported and difficult to interpret (Phan, Wager,
Taylor, & Liberzon, 2002; Wager et al., 2003). We then
calculated a peak density estimate for each of the 2 3 2 3
2 mm voxels in the brain; this was defined as the number
of n peaks in the analysis contained within a sphere of 10–
20 mm (depending on analysis, described below) surround-
ing that voxel, divided by the volume of the sphere. Thus,
the units of density reported are peaks per cubic millimeter
of brain tissue. In order to determine a density distribu-
tion for the null hypothesis, we conducted a Monte Carlo
simulation with 5,000 iterations per analysis, assuming no
systematic spatial organization of the voxels. For each iter-
ation, n points corresponding to the n reported peaks were
distributed randomly throughout the gray and white matter
of the brain (excluding ventricles and sinus spaces). White
matter was included because many reported peaks fall
within white matter near white/gray matter boundaries.3
The density estimate map across the brain for the peaks
as actually reported in the literature was then compared
with this null distribution, using a significance threshold of
the 95th percentile of the null distribution (p , .05, brain-
wise, one-tailed). The test statistic is the density of reported
peaks in the local area around the voxel being tested, and
the Monte Carlo simulation provides p values that reflect
how (un)likely it is to obtain the observed density if peaks
were actually randomly (uniformly) distributed throughout
the brain. A low p value would indicate that the null hy-
pothesis uniform distribution of peaks is unlikely to result
in a cluster as dense as the one observed. If the density es-
timate of a given voxel was significantly greater than what
would be expected by the simulated null distribution, we
took this voxel to be active for that particular analysis.4
Active voxels were grouped into contiguous voxels, using
SPM2’s contiguity assessment procedures (spm_cluster.m;
Wellcome Department of Imaging Neuroscience); that is,
if voxels share at least one vertex, they are considered to be
part of the same contiguous region. The resulting clusters
are reported in Table 2. Localization of these clusters was
performed by first converting the clusters back into Talai-
rach space (www.mrc-cbu.cam.ac.uk/Imaging/) and then
consulting a standard brain atlas (Talairach & Tournoux,
We performed a separate density analysis for each in-
terference contrast: go/no-go, flanker, SRC, and Stroop.
Due to the small number of studies in which the Simon
and stop signal tasks were investigated, we were unable to
perform a density analysis on these tasks. In addition, we
performed a density analysis on all of the studies taken
iNterfereNce resolutioN 5
together. For the individual studies, a density sphere with a
20-mm radius was used. We used a larger sphere for these
analyses because few studies and, therefore, few coordi-
nates were available for each of these tasks. For the analy-
sis that combined all the tasks, we used a smaller region
of 10-mm radius, consistent with the size used in previous
such meta-analyses (Wager et al., 2003).
The density analysis performed on the combination of all
the tasks produced significant clusters bilaterally in the dor-
solateral prefrontal cortex (DLPFC), inferior frontal gyrus
(IFG), anterior cingulate cortex (ACC), and posterior parietal
cortex (PPC) (see Figure 4). Table 2 summarizes the results.
Individual Task Analyses
Density analyses performed on each task individually
by and large revealed a proper subset of the analysis of the
combination of all the tasks (see Table 2 and Figure 3).
Go/no-go. For the go/no-go task, the most prominent
cluster was in the right DLPFC, extending inferiorly into
the right IFG and insula. There were also significant clus-
ters in the left DLPFC, ACC, and right PPC, but these
were smaller in extent. There were also small clusters in
the right occipital cortex.
Flanker. The flanker task produced a significant clus-
ter in the right DLPFC. Another smaller cluster was found
in the right insula, but the extent of the inferior cluster was
not nearly the size of the one found in the go/no-go task.
Stimulus–response compatibility. The SRC task pro-
duced reliable clusters most prominently in the bilateral
PPC, but primarily right lateralized. Clusters were also
found in the left supplementary motor area and premotor
cortex, as well as in the ACC.
Stroop. Clusters from the Stroop task were primar-
ily left lateralized. There was a large cluster in the left
DLPFC that extended inferiorly to the insula. In addition,
we found a very large cluster in the medial frontal cortex,
All tasks combined
medial frontal/anterior cingulate
right dorsolateral prefrontal cortex
left premotor cortex
left inferior frontal/insula
right dorsolateral prefrontal cortex
right inferior frontal/insula
right inferior parietal lobule
left inferior parietal lobule
right inferior parietal lobule
left dorsolateral prefrontal cortex
right dorsolateral prefrontal/inferior frontal
left dorsolateral prefrontal cortex
right angular gyrus
anterior cingulate cortex
left middle frontal gyrus
left anterior cingulate cortex
left dorsolateral prefrontal cortex
right inferior occipital gyrus
right middle occipital gyrus
left premotor/supplementary motor area
anterior cingulate cortex
right premotor/supplementary motor area
left premotor cortex
left anterior cingulate cortex
right dorsolateral prefrontal cortex
medial frontal/anterior cingulate cortex
left dorsolateral prefrontal cortex/inferior frontal
right inferior parietal lobule
right dorsolateral prefrontal cortex
right dorsolateral prefrontal cortex
right dorsolateral prefrontal cortex
Note—Coordinates are reported in MNI space. Voxels is the area of the region in voxels. Only clusters of 5 voxels or more
are reported. BA, Brodmann area.
6 Nee, Wager, aNd JoNides
including the ACC. To a lesser extent, there was also a
cluster in the left PPC. There were also clusters in the right
DLPFC and PPC, but these clusters were much smaller in
extent than the ones found in the left hemisphere. Finally,
there was also a small cluster in the thalamus.
Despite the seemingly random scatter of activation
pictured in Figure 1, our density analysis yielded reliable
clusters of activation in many areas that have often been
implicated in interference resolution (see Figures 2 and 4).
This network of regions may, therefore, be involved in in-
terference resolution in general. However, a look at our
individual task analyses reveals that each task reliably ac-
tivates a subset of these regions. Understanding why each
task loads differentially on a distinct subset of regions may
be the key to understanding how the brain resolves conflict.
Each task included in this study relies on different
methods for inducing cognitive conflict. It is likely that
these different forms of conflict act upon different neural
mechanisms. For instance, mechanisms that filter out dis-
tracting visual information may be useful in the flanker,
Stroop, and Simon tasks, in which conflict is produced by
competing irrelevant stimuli, but these same mechanisms
would not be relevant for the go/no-go task, in which there
are no visual distractors. Therefore, examining the dif-
ferences in the kinds of conflict each task produces and
differences in the neural activations that accompany each
kind of conflict resolution may shed light on the neural
mechanisms underlying interference resolution.
Go/No-Go and Stop Signal
It is clear that the go/no-go task induces conflict in
mechanisms responsible for selecting and executing an
appropriate response. As some authors have argued, re-
sponse selection and response execution may be distin-
guishable stages of processing (Rubia et al., 2001; Rubia,
Smith, Brammer, & Taylor, 2003). Therefore, when
subjects attempt to overcome the prepotent tendency to
respond in the go/no-go task, they may accomplish this
either by biasing decision processes toward selecting the
appropriate response or by restraining an inappropriate re-
sponse from being executed and later selecting the appro-
priate response. In the former case, interference resolution
acts upon response selection, and in the latter, it acts upon
response execution. At which stage conflict is resolved
is likely influenced by the experimental parameters. For
instance, as the proportion of go to no-go trials increases,
a greater prepotency to respond is formed, which may
heavily bias response selection processes in favor of re-
sponding, thereby making a subject more reliant upon
mechanisms of restraint that act upon response execution
(de Zubicaray et al., 2000; Garavan, Ross, & Stein, 1999).
It is likely also that speeded responding would produce a
similar effect. Although changes in task parameters would
be interesting to explore, we have an insufficient number
of studies in which the go/no-go task has been explored to
warrant meta-analytic techniques. Therefore, for specifics
on how the neural mechanisms underlying interference
resolution change as task parameters differ, we rely on
By far the most reliable activation we found in the go/
no-go task was in the right frontal cortex, including the
DLPFC and inferior frontal regions. Somewhat specu-
latively, we can tease apart what parts of this activation
may be due to response selection and what may be due
to response execution. One approach is to examine what
neural changes occur as the go/no-go task becomes more
or less difficult. Presumably, by the logic we have pre-
sented, increased difficulty caused by an increased pre-
potency to respond requires a greater contribution of
All Tasks Combined
Figure 2. Results of a peak density analysis performed on all of the 47 studies included.
Regions are reported in Table 2.
iNterfereNce resolutioN 7
resolution mechanisms acting upon response execution.
Several studies in which this has been examined have re-
ported that activation in the right IFG increases with in-
creased task difficulty (Durston, Thomas, Worden, et al.,
2002; Durston, Thomas, Yang, et al., 2002; Garavan et al.,
1999). Another study in which the number of no-go tri-
als was parametrically varied showed that as the number
of no-go trials increased, reaction times increased, and
errors decreased, suggesting a shift toward more con-
trolled responding (de Zubicaray et al., 2000). This shift
in response style was accompanied by an increase in the
right DLPFC. Taken together, it appears that in the go/no-
go task, right IFG activation underlies resolution during
response execution, whereas right DLPFC activation ac-
companies more controlled resolution, perhaps during the
selection of a response.
Although our reasoning is somewhat speculative, it cor-
roborates well the results in the literature concerning the
stop signal task. In the stop signal task, the subject must
restrain a response when a stop signal occurs, thereby re-
lying solely upon mechanisms that resolve conflict during
the execution of a response. Indeed, neuroimaging studies
in which the stop signal task has been examined have im-
plicated the right IFG for this kind of interference resolu-
tion (Rubia et al., 2001; Rubia et al., 2003). Even stronger
evidence for this case is made by lesion evidence. It has
been shown that as the size of a lesion in the right IFG
increases, performance in the stop signal task gets poorer,
thereby implicating the right IFG as a region that is vital
to the resolution of conflict during response execution
(Aron, Fletcher, Bullmore, Sahakian, & Robbins, 2003;
Aron, Robbins, & Poldrack, 2004). Although we had an
insufficient number of stop signal studies to examine this
task separately, the combination of neuroimaging and le-
sion evidence appears to provide strong support for the
notion that the right IFG is heavily involved in resolving
conflict due to response execution.
Our examination of the flanker task revealed signifi-
cant clusters in the right DLPFC and right insula. Notably,
these areas overlapped with the frontal areas activated by
the go/no-go task, suggesting that these regions may un-
derlie common mechanisms (see Figure 5; Wager et al.,
2005). What might these mechanisms be? As was de-
scribed above, the flanker task can involve stimulus con-
flict, when the distractor stimuli and target stimuli do not
match, and response conflict, when the distractor stimuli
are mapped onto a response different from that for the
target stimuli. Since the go/no-go task does not include
stimulus conflict, the overlapping activations most likely
result from response conflict. However, in our discussion
Individual Task Density Analyses
Stroop Stroop (I–C) Stroop (I–N)
Figure 3. Results of peak density analyses performed separately on the go/no-go, flanker,
stimulus–response compatibility (SRC), and Stroop tasks. Also included are separate density
analyses performed on studies investigating the incongruent versus congruent Stroop con-
trast and the incongruent versus neutral Stroop contrast.
8 Nee, Wager, aNd JoNides
of the go/no-go task, we delineated two forms of response
conflict: response selection conflict and response execu-
tion conflict. Furthermore, we implicated right DLPFC
activation with resolution of response selection conflict
and right inferior frontal activation with the resolution of
conflict during response execution. Do these implications
match up with the flanker data?
The overlap in the right DLPFC appears to be concor-
dant with the idea that the right DLPFC is involved in in-
terference resolution during response selection. Incongru-
ent flankers bias response selection processes against the
appropriate response, thereby requiring resolution pro-
cesses to overcome this bias. Therefore, in both the flanker
and the go/no-go tasks, there is a need to select against a
bias toward an inappropriate response. However, the low
error rates typically found with the flanker task suggest
that there is little need to restrain a response during re-
sponse execution. Therefore, the inferior frontal overlap
appears to be somewhat puzzling.
Whereas the go/no-go task produced a cluster that in-
corporated both the right IFG and the insula, the inferior
frontal cluster in the flanker task was found exclusively
in the insula. We did not distinguish the right IFG from
the insula in our earlier discussion, mainly because the
role of the insula in cognitive tasks remains unclear. One
study that compared the go/no-go, flanker, and SRC
tasks in the same subjects found common anterior insula
activation among the tasks, activation which correlated
with behavioral performance (Wager et al., 2005). These
authors argued that since all three tasks have resolution
processes acting upon response selection in common, the
insula is involved somehow in response selection pro-
cesses. If this is the case, the common insula activation
we found in our meta-analysis is orderly, in that it may
reflect common mechanisms of response selection in the
go/no-go and flanker tasks. However, other authors have
argued that the anterior insula is involved in the restrain-
ing of inappropriate responses (Garavan et al., 1999). If
z = 0 z = 2z = 4 z = 6 z = 8z = 10
z = 12z = 14z = 16 z = 22z = 24 z = 26
z = 28z = 30z = 32 z = 34z = 36 z = 38
z = 40 z = 42z = 44 z = 46z = 48 z = 50
z = 52z = 54z = 56
Figure 4. Slice renderings of the peak density analysis performed on all of the 47 studies included.
iNterfereNce resolutioN 9
this is the case, the insula cluster found here is somewhat
A closer look into our flanker studies reveals that in one
study, a speeded flanker task was used that produced nearly
chance accuracy on incongruent trials (Ullsperger & von
Cramon, 2001). The difficulty of this task may have shifted
resolution processes to response execution, consistent with
our logic for the go/no-go task. Indeed, 11 of the 14 inferior
frontal peaks found in our flanker analysis were contributed
by this study alone. Therefore, if the inferior frontal region
shared by the go/no-go and flanker tasks really does reflect
resolution processes acting upon response execution, our
flanker result appears to be compatible with this account.
The SRC task is similar to the go/no-go task in that sub-
jects must overcome a prepotent tendency to respond inap-
propriately in order to perform the task correctly. However,
z = –14z = –12 z = –10z = –8 z = 10 z = 12
z = 14z = 16z = 18 z = 20z = 22z = 24
z = 26z = 28 z = 30z = 32 z = 34z = 36
z = 38z = 40 z = 42z = 44z = 46 z = 48
z = 50z = 52 z = 54z = 56z = 58 z = 60
z = 62
Go/no-go & flanker
Figure 5. Slice renderings showing the results of the peak density analyses performed on the go/no-go (red), flanker
(green), and stimulus–response compatibility (SRC; blue) tasks. Activation overlap between the go/no-go and flanker
tasks is depicted in yellow.
10 Nee, Wager, aNd JoNides
the prepotency in the two tasks is somewhat different. In the
go/no-go task, the prepotent tendency to respond is due to
the immediately preceding context. In other words, a subject
has responded to several go trials and is, therefore, likely to
respond. In the SRC task, however, the prepotency to re-
spond inappropriately is due not to the immediately preced-
ing context but, rather, to the subject’s familiarity with the
stimulus in general. For example, it is more natural, on the
basis of previously learned responses, to respond to a left
arrow with a left response. However, on incompatible trials,
subjects must overcome this learned tendency to respond ap-
propriately. Put another way, one major difference between
the go/no-go and the SRC tasks is a difference in time scale:
In one, a response has been learned due to an immediately
preceding context; in the other, a response has been learned
over the course of a lifetime. As our data illustrate, these
differences result in very different neural patterns.
In contrast to the go/no-go and flanker tasks, which
produced predominately frontal activation, the bulk of
the activation in the SRC task was in the parietal cortex.
The largest cluster was in the left posterior parietal cortex,
which included the intraparietal sulcus. Additional clus-
ters were found in the anterior cingulate and premotor/
supplementary motor area.
We can contrast the go/no-go and flanker tasks with the
SRC task to help understand these differences. Let us begin
with what is common between them. Although a study com-
paring the go/no-go, flanker, and SRC tasks in the same sub-
jects found several neural regions in common to all (Wager
et al., 2005), the only overlap we found was that between the
go/no-go and the SRC tasks in the left premotor cortex. This
region has been implicated in response selection (Iacoboni,
Woods, & Mazziotta, 1998), and repetitive transcranial mag-
netic stimulation (rTMS) performed on this region impairs
performance on incompatible trials (Praamstra, Kleine, &
Schnitzler, 1999). Therefore, the SRC task may have some
component of interference resolution during response selec-
tion in common with the go/no-go task.
What is interesting about the neural pattern of results
found for the SRC task is that the areas implicated are
exactly those regions implicated in a meta-analysis of
switching attention (Wager et al., 2004). Indeed, a study in
which the SRC task was directly compared with a switch-
ing task showed close parallels between the neural signa-
tures of the two tasks (Sylvester et al., 2003). One possi-
bility is that interference resolution in the SRC task is very
similar to switching. On incompatible trials, the prepotent
response may automatically be elicited, and subjects may
need to switch their response set to activate the appropri-
ate response. An alternative but similar proposal is that
activation for both switching and the SRC task indicates
the need to select among competing stimulus–response
associations. However, further testing will be needed to
verify whether the type of resolution involved in the SRC
task is truly more akin to switching than is the resolution
involved in the go/no-go and flanker tasks.
Like the flanker task, the Stroop task involves filtering
out distracting irrelevant information that can compete
with the appropriate response. However, unlike the flanker
task, in which the distractors are adjacent to the imperative
stimulus, in the Stroop task, the target and the distractor
are different attributes of the same object. In addition, due
to the automatic nature of reading the distracting material,
an incorrect response is highly prepotent on incongruent
trials. Therefore, in the Stroop task, there appears to be
a greater demand for selective attention to filter out the
In the event that selective attention fails to filter out
irrelevant information completely, it is likely that the ir-
relevant information will bias toward the inappropriate
response. How interference resolution proceeds in this
case depends on the specifics of the paradigm. It has been
argued that the verbal response Stroop task is very dif-
ferent from a manual response Stroop task, due to the
verbal task’s having an automatic mapping of stimulus to
response, whereas the manual case has an arbitrary map-
ping (MacLeod, 1991). Due to the movement involved in
verbal responses, neuroimaging has relied, by and large,
on manual responses. A manual response version of the
Stroop task most likely relies upon response selection
because the subject is required to select among the arbi-
trary mappings provided by the experimenter. In this case,
as with the go/no-go, flanker, and SRC tasks, resolution
mechanisms must act upon response selection to favor the
correct response in the face of strong competition.
Overall, the Stroop task is similar to the other tasks
studied here in its reliance on interference resolution act-
ing upon response selection. However, it differs from the
other tasks in its greater need for selective attention. In
addition, of the tasks studied here, the Stroop task is the
most verbal in nature. Beginning with what is common,
the Stroop task overlaps with the go/no-go and flanker
tasks in the right DLPFC and with the SRC task in the left
premotor/supplementary motor area. Consonant with the
idea that the Stroop task shares response selection compo-
nents with these tasks, all of these regions have been im-
plicated in the resolution of interference during response
selection (Bunge, Hazeltine, Scanlon, Rosen, & Gabrieli,
2002; Durston, Thomas, Worden, et al., 2002; Durston,
Thomas, Yang, et al., 2002; Iacoboni et al., 1998; Praam-
stra et al., 1999). In addition, the Stroop task overlaps both
the go/no-go and the SRC tasks in the ACC. This region
has been the subject of much debate, mostly centered
around its function as a monitor involved in the resolu-
tion of response conflict (e.g., Botvinick, Braver, Barch,
Carter, & Cohen, 2001). Once again, this is compatible
with the notion that the Stroop task shares response selec-
tion components with the other tasks.
Although we failed to find significant overlap, there
was a close correspondence in the left posterior parietal
cortex (BA 7) between the Stroop task (]22, ]64, 46) and
the SRC task (]16, ]62, 48). Both tasks share the need
to overcome an overly learned prepotent response: in the
SRC task, the tendency to respond left to a left-pointing
arrow, and in the Stroop task, the tendency to read writ-
ten words. Earlier, we speculated that resolution in the
SRC task may be similar to switching from the prepotent
response set to the appropriate response set or selecting
iNterfereNce resolutioN 11
among stimulus–response associations. It is worth noting
that regions that are highly related to switching accord-
ing to a meta-analysis of switching tasks (the premotor
cortex, the intraparietal sulcus, and the anterior cingulate)
are present in the Stroop analysis, as they are in the SRC
analysis (Wager et al., 2004). Therefore, the Stroop and
SRC tasks may share the same sort of switch or stimulus–
response association-related interference resolution.
Other authors have also speculated that the Stroop task
shares mechanisms with switching tasks (Brass, Derrfuss,
Forstmann, & von Cramon, 2005; Derrfuss, Brass, Neu-
mann, & von Cramon, 2005). These authors proposed that
both tasks share the need for the updating of task repre-
sentations. On the basis of a meta-analysis of Stroop and
switching tasks, these authors proposed that a region in
the left frontal cortex, termed the inferior frontal junc-
tion, may mediate this function (Derrfuss et al., 2005). In
accordance with this claim, we found a large left frontal
cluster in our Stroop analysis that overlapped with this
proposed region. We might expect to find similar activa-
tion in the SRC task, which would ostensibly require the
same task-representation-updating function. However, we
did not find reliable clusters in this region in our SRC
analyses, although it is possible that this result was due to
insufficient power. Perhaps more puzzling is the finding
that a separate meta-analysis of 31 switching studies also
failed to show reliable clusters in the left inferior frontal
junction (Wager et al., 2004), although these authors did
find a large left dorsolateral prefrontal region that was
more anterior to the inferior frontal junction at a reduced
threshold. Further examination will be required to provide
a consensus regarding the role of the inferior frontal junc-
tion and its relation to the Stroop and switching tasks.
In contrast to the other tasks studied here, the Stroop
task is highly left lateralized, most prominently in the left
DLPFC and inferior frontal regions. Part of this lateraliza-
tion may be due to the strongly verbal nature of the Stroop
task. Indeed, some authors have implicated left inferior
frontal regions in the resolution of verbal conflict (Jonides
& Nee, 2006; Jonides et al., 1998; Leung, Skudlarski,
Gatenby, Peterson, & Gore, 2000). Some of the lateraliza-
tion may also be due to the greater need for selective atten-
tion processes involved in filtering out strongly competitive
irrelevant information. Consonant with this idea, one study
examined differences during a preparatory period preced-
ing either the Stroop task or the reverse Stroop task where
subjects make the easier response of responding to the
word rather than the color (MacDonald, Cohen, Stenger,
& Carter, 2000). These authors found greater left DLPFC
activation for the Stroop task than for the reverse Stroop
task. If we believe that the selective attention demands are
greater for the more difficult task, it follows that the left
DLPFC may be engaged in preparation for high demands
on selective attention. However, this version of the Stroop
task involved switching between Stroop and reverse Stroop
and produced an abnormally large reverse-Stroop effect, so
any conclusions must be drawn with caution.
Perhaps better evidence regarding the involvement of
the left DLPFC in selective attention comes from several
studies that examined differences when competing stimuli
are response eligible versus response ineligible (Liu, Ban-
ich, Jacobson, & Tanabe, 2006; Milham & Banich, 2005;
Milham, Banich, & Barad, 2003; Milham et al., 2001).
In these studies, the subjects learned a mapping of some
colors to response keys. These were response-eligible col-
ors. Contrasting with these colors were other colors that
did not have mapped responses. Since these items were
not available for response, they were response ineligible.
Importantly, when used as distracting words, response-
eligible words caused both stimulus and response conflict
on incongruent trials, whereas response-ineligible words
caused only stimulus conflict. Therefore, examining neu-
ral responses to response-ineligible trials, in comparison
with neutral trials, isolates processes involved in resolv-
ing stimulus conflict. Indeed, several studies in which this
paradigm has been examined have shown the left DLPFC
as being related to resolving stimulus conflict (Liu et al.,
2006; Milham & Banich, 2005; Milham et al., 2003; Mil-
ham et al., 2001). This lends support to the idea that the
left DLPFC is involved in selective attention.
Unlike in our other tasks, we included both incongruent
versus congruent and incongruent versus neutral peaks
in our analyses, in that there were many studies in which
each contrast was examined. Since we had sufficient data,
we also explored whether the incongruent versus congru-
ent contrast differed from the incongruent versus neutral
contrast, as some authors have reported (Bench et al.,
1993; Carter, Mintun, & Cohen, 1995; Taylor, Kornblum,
Lauber, Minoshima, & Koeppe, 1997). Indeed, there
were significant differences (see Figures 3 and 6). The
incongruent minus neutral contrast exhibited far greater
left DLPFC and left posterior parietal activation, whereas
the incongruent versus congruent contrast revealed larger
ACC activation. What this must mean is that congruent
trials produce greater activation in the left DLPFC and
posterior parietal cortex than do neutral trials and less ac-
tivation in the ACC than do neutral trials.
Although we are uncertain what exactly to make of
these differences, we can provide some speculation. Un-
like neutral trials, congruent trials provide a competing
response-eligible stimulus (Milham et al., 2002). If the
strategy of the subject is to try hard to ignore the irrel-
evant word, the fact that the word is part of the color set
may trigger mechanisms involved in selecting the correct
stimulus dimension (color). By this account, we would
expect increases in the left DLPFC during congruent trials
to filter out potentially distracting information. However,
since the responses indicated by both the word and the
color are the same, there is no conflict at the response se-
lection stage. Therefore, the reduction in actual response
conflict may decrease demand on the ACC. These specu-
lations are supported by a study that examined regions
specifically recruited by conflict (incongruent . congru-
ent and neutral trials) and those by competition (incongru-
ent and congruent . neutral) (Milham & Banich, 2005).
In this study, there was greater left dorsolateral prefrontal
activity associated with competition (although still sig-
nificant activation, to a lesser extent, in the left DLPFC
for conflict), consonant with the idea that the left DLPFC
is involved in both incongruent and congruent trials where
12 Nee, Wager, aNd JoNides
there is a competing response-eligible word. By contrast,
there was greater ACC activity associated with conflict (al-
though a smaller, dissociable region of the ACC produced
competition-related activation). These results corroborate
well our finding of greater left DLPFC activation for the
incongruent minus neutral contrast and greater ACC acti-
vation for the incongruent minus congruent contrast.
Putting It Together
We began by noting a network of regions involved in in-
terference resolution and then interrogated the individual
tasks, to attempt to understand the functions that the indi-
vidual pieces within this network are performing. What is
arrived at by piecing together the individual facts is a pro-
posal of separate interference resolution mechanisms act-
ing upon different stages of processing. Specifically, from
the go/no-go and stop signal data, it appears as though right
inferior frontal regions are heavily involved in restrain-
ing an inappropriate response during response execution.
Commonalities in the go/no-go, flanker, SRC, and Stroop
tasks implicate the right DLPFC and ACC in interference
resolution during response selection. For cases such as the
Stroop and SRC tasks, the intraparietal sulcus and premo-
tor cortex may also be involved during response selection,
perhaps as a means of switching from inappropriate to
appropriate response sets or selecting among competing
stimulus–response associations. Finally, the Stroop data
point to the left DLPFC for resolution of stimulus conflict,
perhaps via selective attention mechanisms, and to the left
inferior frontal regions for resolution of verbal conflict.
z = 12 z = 14z = 16 z = 18z = 20
z = 22z = 24 z = 26z = 28z = 30
z = 32 z = 34 z = 36z = 38 z = 40
z = 42 z = 44z = 46z = 48 z = 50
z = 52z = 54z = 56 z = 58
Figure 6. Slice renderings of the two different Stroop contrasts, including their overlap. I, incongruent;
C, congruent; N, neutral.
iNterfereNce resolutioN 13
We are quick to note that these are merely hypotheses
borne out of the meta-analysis, rather than conclusions.
Each of these hypotheses needs further testing.
To bolster these hypotheses, we performed a logistic
regression in order to investigate whether resolution of
interference at different stages of processing would pre-
dict activation in a given region. We coded each study
by whether the task included resolution during response
execution, response selection, or stimulus encoding and
examined whether these predictors explained activation
in the right IFG, left IFG, left DLPFC, right DLPFC, and
ACC. In addition, since we hypothesized that resolution of
verbal information may involve the left IFG, we performed
a separate logistic regression on the left IFG, using ver-
bal conflict as a predictor (Jonides & Nee, 2006; Jonides
et al., 1998; Leung et al., 2000; Nelson, Reuter-Lorenz,
Sylvester, Jonides, & Smith, 2003). The results supported
our hypotheses. Stimulus conflict significantly predicted
activation in the left DLPFC (Wald 5 5.58, p 5 .018) and
the left IFG (Wald 5 6.9, p 5 .008). Verbal conflict also
predicted left IFG activation (Wald 5 6.5, p 5 .01). Fi-
nally, conflict during response execution marginally pre-
dicted activation in the right IFG (Wald 5 3.27, p 5 .07).
We note that conflict during response selection did not
significantly predict activation in any region, but this is
most likely due to the fact that all tasks other than the one
stop signal task included in our meta-analysis elicit con-
flict during response selection and, therefore, this predic-
tor had insufficient variance to explain activation.
These results are consistent with the idea that differ-
ent neural regions are responsible for the resolution of in-
terference at different stages of processing. However, we
recognize that there may be other ways to organize inter-
ference resolution processes as well. Some authors have
carefully distinguished several different forms of conflict,
each of which may require its own dissociable resolution
mechanisms (Kornblum, Hasbroucq, & Osman, 1990;
Kornblum et al., 1999; Zhang et al., 1999). Unfortunately,
most neuroimaging studies of interference resolution con-
found several of these forms of conflict, thereby making
it difficult to distinguish among them. Further investiga-
tion is needed to determine whether interference resolu-
tion mechanisms can be more finely dissociated than we
Mechanisms of Interference Resolution
We have implicated several regions as important in the
resolution of interference, but we have not speculated how
this conflict is resolved. Resolution may proceed via the
facilitation of appropriate information, inhibition of inap-
propriate information, a combination of the two, or some
other strategy, such as switching response sets (Hasher,
Zacks, & May, 1999; MacLeod, Dodd, Sheard, Wilson,
& Bibi, 2003). We believe that the extant data cannot yet
penetrate this question, so we remain agnostic as to how
interference is resolved.
Relation to Other Work
Several other meta-analyses have been performed to
look for consistencies among neuroimaging data (Cabeza
& Nyberg, 2000; Duncan & Owen, 2000; Johnson et al.,
2005; Wager et al., 2004; Wager & Smith, 2003). For ex-
ample, Duncan and Owen demonstrated that regions of
the frontal cortex, including the ACC and the dorsolat-
eral and ventrolateral prefrontal cortices, were recruited
by diverse cognitive demands not exclusive to conflict.
Their analyses of 19 studies produced little if any discern-
ible dissociation among the various tasks studied when all
peaks were plotted on the same canonical brain. Similar to
their analysis, combining all of our studies produced the
same network of regions. However, when each task was
interrogated individually, we found dissociations within
this network. Why did we find dissociations when Duncan
and Owen did not?
Figure 7 shows a plot of all of the peaks included in this
study, color coded by the particular task contributing the
peak. From this figure, it is difficult to discern dissociable
patterns. Examining the data in this way demonstrates the
clear need for clustering techniques. It is possible that with
the inclusion of more studies and a clustering technique,
dissociable patterns may emerge from the tasks studied by
Duncan and Owen (2000).
Other meta-analyses in which particular tasks have been
looked at have shown regions overlapping with the regions
we find here. As was mentioned earlier, a meta-analysis
of switching tasks produced clusters in the parietal and
premotor cortex similar to the areas we found for the SRC
and Stroop tasks, indicating that resolution for these tasks
may have a commonality with switching (Wager et al.,
2004). Furthermore, a meta-analysis of working memory
tasks implicated several frontal and parietal regions found
here (Wager & Smith, 2003). This corroborates findings
linking working memory with susceptibility to interfer-
ence (de Fockert, Rees, Frith, & Lavie, 2001; Engle, Kane,
& Tuholski, 1999; Hester, Murphy, & Garavan, 2004;
Kane, Bleckley, Conway, & Engle, 2001; Kane & Engle,
2003; Kim, Kim, & Chun, 2005). Working memory tasks
often require not only the active maintenance of informa-
tion, but also the filtering out of distraction and selecting
among representations for both maintenance and response
processes. Consonant with the idea that the left DLPFC is
important for selective attention, increasing demand on
the left DLPFC by imposing a working memory load in-
creases interference from irrelevant perceptual material
(de Fockert et al., 2001). In addition, increasing demand
on the right DLPFC by increasing working memory load
decreases go/no-go performance, perhaps due to shared
components of response selection (Hester et al., 2004).
These results suggest a close tie between working mem-
ory and interference resolution.
Finally, refreshing, or bringing to focus an item in mind,
recruits the bilateral frontal cortex, ACC, and PPC (John-
son et al., 2005). Refreshing verbal material preferentially
activates left inferior frontal regions, whereas no other
refresh-related region demonstrates a verbal preference
(Johnson et al., 2005). The left inferior frontal gyrus may
be important in selecting the appropriate verbal mate-
rial to refresh (Jonides & Nee, 2006; Thompson-Schill,
D’Esposito, Aguirre, & Farah, 1997). This verbal selection
role is consonant with our data demonstrating that the left
14 Nee, Wager, aNd JoNides
IFG is needed for interference resolution of verbal conflict.
It is likely that other commonalities between regions found
here and in refreshing also reflect selection of various sorts
of representations. Further work is needed to examine the
relation between interference resolution and refreshing.
Examining the combination of many tasks that involve
interference resolution revealed that a network including
the bilateral DLPFC, inferior frontal regions, the PPC,
and the ACC may underlie the resolution of conflict. We
hypothesize that separating functions by the stage of pro-
cessing at which conflict is resolved may provide a use-
ful framework for understanding interference resolution.
Although future research will be needed to test these hy-
potheses and add further to our understanding of how each
region performs interference resolution, our data suggest
that the right IFG is important during response execution,
the right DLPFC and ACC during response selection, and
the left DLPFC during stimulus encoding. In addition,
switching-related regions in the intraparietal sulcus and
premotor cortex may contribute to some forms of interfer-
This material is based on work supported by the National Science
Foundation under Grant 0520992 and by a National Science Foundation
Graduate Research Fellowship. Correspondence concerning this article
should be addressed to D. E. Nee, Department of Psychology, Univer-
Figure 7. Peaks from all 47 studies plotted in a canonical brain, color coded by task. GNG,
go/no-go; SRC, stimulus–response compatibility.
iNterfereNce resolutioN 15
sity of Michigan, 530 Church St., Ann Arbor, MI 48109-1043 (e-mail:
Adleman, N. E., Menon, V., Blasey, C. M., White, C. D., Warsof-
sky, I. S., Glover, G. H., & Reiss, A. L. (2002). A developmental
fMRI study of the Stroop color–word task. NeuroImage, 16, 61-75.
Aron, A. R., Fletcher, P. C., Bullmore, E. T., Sahakian, B. J., &
Robbins, T. W. (2003). Stop-signal inhibition disrupted by damage
to right inferior frontal gyrus in humans. Nature Neuroscience, 6,
Aron, A. R., Robbins, T. W., & Poldrack, R. A. (2004). Inhibition
and the right inferior frontal cortex. Trends in Cognitive Sciences, 8,
Banich, M. T., Milham, M. P., Atchley, R., Cohen, N. J., Webb, A.,
Wszalek, T., et al. (2000). fMRI studies of Stroop tasks reveal
unique roles of anterior and posterior brain systems in attentional se-
lection. Journal of Cognitive Neuroscience, 12, 988-1000.
Banich, M. T., Milham, M. P., Jacobson, B. L., Webb, A., Wszalek, T.,
Cohen, N. J., & Kramer, A. F. (2001). Attentional selection and
the processing of task-irrelevant information: Insights from fMRI
examinations of the Stroop task. Progress in Brain Research, 134,
Bench, C. J., Frith, C. D., Grasby, P. M., Friston, K. J., Paulesu, E.,
Frackowiak, R. S. J., & Dolan, R. J. (1993). Investigations of the
functional anatomy of attention using the Stroop test. Neuropsycho-
logia, 31, 907-922.
Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., &
Cohen, J. D. (2001). Conflict monitoring and cognitive control. Psy-
chological Review, 108, 624-652.
Brass, M., Derrfuss, J., Forstmann, B., & von Cramon, D. Y. (2005).
The role of the inferior frontal junction area in cognitive control.
Trends in Cognitive Sciences, 9, 314-316.
Bunge, S. A., Hazeltine, E., Scanlon, M. D., Rosen, A. C., & Ga-
brieli, J. D. E. (2002). Dissociable contributions of prefrontal and
parietal cortices to response selection. NeuroImage, 17, 1562-1571.
Cabeza, R., & Nyberg, L. (2000). Imaging cognition II: An empirical
review of 275 PET and fMRI studies. Journal of Cognitive Neurosci-
ence, 12, 1-47.
Carter, C. S., Mintun, M., & Cohen, J. D. (1995). Interference and
facilitation effects during selective attention: An H215O PET study of
Stroop task performance. NeuroImage, 2, 264-272.
Casey, B. J., Thomas, K. M., Davidson, M. C., Kunz, K., & Franzen,
P. L. (2002). Dissociating striatal and hippocampal function devel-
opmentally with a stimulus–response compatibility task. Journal of
Neuroscience, 22, 8647-8652.
Dassonville, P., Lewis, S. M., Zhu, X. H., Ugurbil, K., Kim, S. G.,
& Ashe, J. (2001). The effect of stimulus–response compatibility on
cortical motor activation. NeuroImage, 13, 1-14.
de Fockert, J. W., Rees, G., Frith, C. D., & Lavie, N. (2001). The
role of working memory in visual selective attention. Science, 291,
Derbyshire, S. W. G., Vogt, B. A., & Jones, A. K. P. (1998). Pain and
Stroop interference tasks activate separate processing modules in an-
terior cingulate cortex. Experimental Brain Research, 118, 52-60.
Derrfuss, J., Brass, M., Neumann, J., & von Cramon, D. Y. (2005).
Involvement of the inferior frontal junction in cognitive control: Meta-
analyses of switching and Stroop studies. Human Brain Mapping,
de Zubicaray, G. I., Andrew, C., Zelaya, F. O., Williams, S. C. R., &
Dumanoir, C. (2000). Motor response suppression and the prepotent
tendency to respond: A parametric fMRI study. Neuropsychologia,
Duncan, J., & Owen, A. M. (2000). Common regions of the human
frontal lobe recruited by diverse cognitive demands. Trends in Neuro-
sciences, 23, 475-483.
Durston, S., Thomas, K. M., Worden, M. S., Yang, Y., & Casey, B. J.
(2002). The effect of preceding context on inhibition: An event-related
fMRI study. NeuroImage, 16, 449-453.
Durston, S., Thomas, K. M., Yang, Y., Uluˇ g, A. M., Zimmerman,
R. D., & Casey, B. J. (2002). A neural basis for the development of
inhibitory control. Developmental Science, 5, F9-F16.
Engle, R. W., Kane, M. J., & Tuholski, S. W. (1999). Individual dif-
ferences in working memory capacity and what they tell us about
controlled attention, general fluid intelligence, and functions of the
prefrontal cortex. In A. Miyake & P. Shah (Eds.), Models of working
memory: Mechanisms of active maintenance and executive control
(pp. 102-134). Cambridge: Cambridge University Press.
Eriksen, B. A., & Eriksen, C. W. (1974). Effects of noise letters upon
the identification of a target letter in a nonsearch task. Perception &
Psychophysics, 16, 143-149.
Fan, J., Flombaum, J. I., McCandliss, B. D., Thomas, K. M., & Pos-
ner, M. I. (2003). Cognitive and brain consequences of conflict.
NeuroImage, 18, 42-57.
Fox, P. T., Laird, A. R., & Lancaster, J. L. (2005). Coordinate-based
voxel-wise meta-analysis: Dividends of spatial normalization. Report
of a virtual workshop. Human Brain Mapping, 25, 1-5.
Garavan, H., Ross, T. J., Murphy, K., Roche, R. A. P., & Stein, E. A.
(2002). Dissociable executive functions in the dynamic control of be-
havior: Inhibition, error detection, and correction. NeuroImage, 17,
Garavan, H., Ross, T. J., & Stein, E. A. (1999). Right hemispheric dom-
inance of inhibitory control: An event-related functional MRI study.
Proceedings of the National Academy of Sciences, 96, 8301-8306.
Hasher, L., Zacks, R. T., & May, C. P. (1999). Inhibitory control, circa-
dian arousal, and age. In D. Gopher & A. Koriat (Eds.), Attention and
performance XVII: Cognitive regulation of performance. Interaction of
theory and application (pp. 653-675). Cambridge, MA: MIT Press.
Hazeltine, E., Poldrack, R., & Gabrieli, J. D. E. (2000). Neural
activation during response competition. Journal of Cognitive Neuro-
science, 12, 118-129.
Hester, R., Murphy, K., & Garavan, H. (2004). Beyond common
resources: The cortical basis for resolving task interference. Neuro-
Image, 23, 202-212.
Iacoboni, M., Woods, R. P., & Mazziotta, J. C. (1996). Brain–behavior
relationships: Evidence from practice effects in spatial stimulus–
response compatibility. Journal of Neurophysiology, 76, 321-331.
Iacoboni, M., Woods, R. P., & Mazziotta, J. C. (1998). Bimodal (au-
ditory and visual) left frontoparietal circuitry for sensorimotor inte-
gration and sensorimotor learning. Brain, 121, 2135-2143.
Johnson, M. K., Raye, C. L., Mitchell, K. J., Greene, E. J., Cun-
ningham, W. A., & Sanislow, C. A. (2005). Using fMRI to inves-
tigate a component process of reflection: Prefrontal correlates of
refreshing a just-activated representation. Cognitive, Affective, & Be-
havioral Neuroscience, 5, 339-361.
Jonides, J., & Nee, D. E. (2006). Brain mechanisms of proactive inter-
ference in working memory. Neuroscience, 139, 181-193.
Jonides, J., Smith, E. E., Marshuetz, C., Koeppe, R. A., & Reuter-
Lorenz, P. A. (1998). Inhibition in verbal working memory revealed
by brain activation. Proceedings of the National Academy of Sciences,
Kane, M. J., Bleckley, M. K., Conway, A. R., & Engle, R. W. (2001).
A controlled-attention view of working-memory capacity. Journal of
Experimental Psychology: General, 130, 169-183.
Kane, M. J., & Engle, R. W. (2003). Working-memory capacity and the
control of attention: The contributions of goal neglect, response com-
petition, and task set to Stroop interference. Journal of Experimental
Psychology: General, 132, 47-70.
Kiehl, K. A., Liddle, P. F., & Hopfinger, J. B. (2000). Error process-
ing and the rostral anterior cingulate: An event-related fMRI study.
Psychophysiology, 37, 216-223.
Kim, S.-Y., Kim, M.-S., & Chun, M. M. (2005). Concurrent working
memory load can reduce distraction. Proceedings of the National
Academy of Sciences, 102, 16524-16529.
Konishi, S., Nakajima, K., Uchida, I., Kikyo, H., Kameyama, M.,
& Miyashita, Y. (1999). Common inhibitory mechanism in human
inferior prefrontal cortex revealed by event-related functional MRI.
Brain, 122, 981-991.
Konishi, S., Nakajima, K., Uchida, I., Sekihara, K., & Miyashita, Y.
(1998). No-go dominant brain activity in human inferior prefrontal
cortex revealed by functional magnetic resonance imaging. European
Journal of Neuroscience, 10, 1209-1213.
Kornblum, S., Hasbroucq, T., & Osman, A. (1990). Dimensional
overlap: Cognitive basis for stimulus–response compatibility—a
model and taxonomy. Psychological Review, 97, 253-270.
16 Nee, Wager, aNd JoNides
Kornblum, S., Stevens, G. T., Whipple, A., & Requin, J. (1999). The
effects of irrelevant stimuli: 1. The time course of stimulus–stimulus
and stimulus–response consistency effects with Stroop-like stimuli,
Simon-like tasks, and their factorial combinations. Journal of Experi-
mental Psychology: Human Perception & Performance, 25, 688-714.
Kramer, A. F., Humphrey, D. G., Larish, J. F., Logan, G. D., &
Strayer, D. L. (1994). Aging and inhibition: Beyond a unitary
view of inhibitory processing in attention. Psychology & Aging, 9,
Laird, A. R., McMillan, K. M., Lancaster, J. L., Kochunov, P.,
Turkeltaub, P. E., Pardo, J. V., & Fox, P. T. (2005). A compari-
son of label-based review and ALE meta-analysis in the Stroop task.
Human Brain Mapping, 25, 6-21.
Leung, H.-C., Skudlarski, P., Gatenby, J. C., Peterson, B. S., &
Gore, J. C. (2000). An event-related functional MRI study of the
Stroop color word interference task. Cerebral Cortex, 10, 552-560.
Liddle, P. F., Kiehl, K. A., & Smith, A. M. (2001). Event-related fMRI
study of response inhibition. Human Brain Mapping, 12, 100-109.
Liu, X., Banich, M. T., Jacobson, B. L., & Tanabe, J. L. (2004).
Common and distinct neural substrates of attentional control in an
integrated Simon and spatial Stroop task as assessed by event-related
fMRI. NeuroImage, 22, 1097-1106.
Liu, X., Banich, M. T., Jacobson, B. L., & Tanabe, J. L. (2006). Func-
tional dissociation of attentional selection within PFC: Response and
non-response related aspects of attentional selection as ascertained by
fMRI. Cerebral Cortex, 16, 827-834.
MacDonald, A. W., III, Cohen, J. D., Stenger, V. A., & Carter, C. S.
(2000). Dissociating the role of the dorsolateral prefrontal and anterior
cingulate cortex in cognitive control. Science, 288, 1835-1838.
MacLeod, C. M. (1991). Half a century of research on the Stroop effect:
An integrative review. Psychological Bulletin, 109, 163-203.
MacLeod, C. M., Dodd, M. D., Sheard, E. D., Wilson, D. E., &
Bibi, U. (2003). In opposition to inhibition. In B. H. Ross (Ed.), The
psychology of learning and motivation: Advances in research and
theory (Vol. 43, pp. 163-214). San Diego: Academic Press.
Maclin, E. L., Gratton, G., & Fabiani, M. (2001). Visual spatial lo-
calization conflict: An fMRI study. NeuroReport, 12, 3633-3636.
Menon, V., Adleman, N. E., White, C. D., Glover, G. H., & Reiss,
A. L. (2001). Error-related brain activation during a go/nogo response
inhibition task. Human Brain Mapping, 12, 131-143.
Milham, M. P., & Banich, M. T. (2005). Anterior cingulate cortex: An
fMRI analysis of conflict specificity and functional differentiation.
Human Brain Mapping, 25, 328-335.
Milham, M. P., Banich, M. T., & Barad, V. (2003). Competition for
priority in processing increases prefrontal cortex’s involvement in top-
down control: An event-related fMRI study of the Stroop task. Cogni-
tive Brain Research, 17, 212-222.
Milham, M. P., Banich, M. T., Webb, A., Barad, V., Cohen, N. J.,
Wszalek, T., & Kramer, A. F. (2001). The relative involvement of
anterior cingulate and prefrontal cortex in attentional control depends
on nature of conflict. Cognitive Brain Research, 12, 467-473.
Milham, M. P., Erickson, K. I., Banich, M. T., Kramer, A. F.,
Webb, A., Wszalek, T., & Cohen, N. J. (2002). Attentional control
in the aging brain: Insights from an fMRI study of the Stroop task.
Brain & Cognition, 49, 277-296.
Munoz, D. P., & Everling, S. (2004). Look away: The anti-saccade
task and the voluntary control of eye movement. Nature Reviews Neu-
roscience, 5, 218-228.
Nelson, J. K., Reuter-Lorenz, P. A., Sylvester, C. Y., Jonides, J.,
& Smith, E. E. (2003). Dissociable neural mechanisms underlying
response-based and familiarity-based conflict in working memory.
Proceedings of the National Academy of Sciences, 100, 11171-11175.
Pardo, J. V., Pardo, P. J., Janer, K. W., & Raichle, M. E. (1990). The
anterior cingulate cortex mediates processing selection in the Stroop
attentional conflict paradigm. Proceedings of the National Academy
of Sciences, 87, 256-259.
Paus, T., Petrides, M., Evans, A. C., & Meyer, E. (1993). Role of the
human anterior cingulate cortex in the control of oculomotor, manual,
and speech responses: A positron emission tomography study. Journal
of Neurophysiology, 70, 453-469.
Perlstein, W. M., Dixit, N. K., Carter, C. S., Noll, D. C., & Cohen,
J. D. (2003). Prefrontal cortex dysfunction mediates deficits in work-
ing memory and prepotent responding in schizophrenia. Biological
Psychiatry, 53, 25-38.
Peterson, B. S., Kane, M. J., Alexander, G. M., Lacadie, C.,
Skudlarski, P., Leung, H. C., et al. (2002). An event-related func-
tional MRI study comparing interference effects in the Simon and
Stroop tasks. Cognitive Brain Research, 13, 427-440.
Peterson, B. S., Skudlarski, P., Gatenby, J. C., Zhang, H. P., An-
derson, A. W., & Gore, J. C. (1999). An fMRI study of Stroop
word–color interference: Evidence for cingulate subregions subserv-
ing multiple distributed attentional systems. Biological Psychiatry,
Phan, K. L., Wager, T., Taylor, S. F., & Liberzon, I. (2002). Func-
tional neuroanatomy of emotion: A meta-analysis of emotion activa-
tion studies in PET and fMRI. NeuroImage, 16, 331-348.
Praamstra, P., Kleine, B. U., & Schnitzler, A. (1999). Magnetic
stimulation of the dorsal premotor cortex modulates the Simon effect.
NeuroReport, 10, 3671-3674.
Ravnkilde, B., Videbech, P., Rosenberg, R., Gjedde, A., & Gade, A.
(2002). Putative tests of frontal lobe function: A PET-study of brain
activation during Stroop’s test and verbal fluency. Journal of Clinical
& Experimental Neuropsychology, 24, 534-547.
Rubia, K., Russell, T., Overmeyer, S., Brammer, M. J., Bullmore,
E. T., Sharma, T., et al. (2001). Mapping motor inhibition: Con-
junctive brain activations across different versions of go/no-go and
stop tasks. NeuroImage, 13, 250-261.
Rubia, K., Smith, A. B., Brammer, M. J., & Taylor, E. (2003). Right
inferior prefrontal cortex mediates response inhibition while mesial
prefrontal cortex is responsible for error detection. NeuroImage, 20,
Ruff, C. C., Woodward, T. S., Laurens, K. R., & Liddle, P. F. (2001).
The role of the anterior cingulate cortex in conflict processing: Evi-
dence from reverse Stroop interference. NeuroImage, 14, 1150-1158.
Schumacher, E. H., & D’Esposito, M. (2002). Neural implementation
of response selection in humans as revealed by localized effects of
stimulus–response compatibility on brain activation. Human Brain
Mapping, 17, 193-201.
Shilling, V. M., Chetwynd, A., & Rabbitt, P. M. (2002). Individual
inconsistency across measures of inhibition: An investigation of the
construct validity of inhibition in older adults. Neuropsychologia, 40,
Stroop, J. R. (1935). Studies of interference in serial verbal reactions.
Journal of Experimental Psychology, 18, 643-662.
Sylvester, C.-Y. C., Wager, T. D., Lacey, S. C., Hernandez, L.,
Nichols, T. E., Smith, E. E., & Jonides, J. (2003). Switching atten-
tion and resolving interference: fMRI measures of executive func-
tions. Neuropsychologia, 41, 357-370.
Talairach, J., & Tournoux, P. (1988). Co-planar stereotaxic atlas of
the human brain: 3-dimensional proportional system. An approach to
cerebral imaging (M. Rayport, Trans.). Stuttgart: Thieme.
Tamm, L., Menon, V., & Reiss, A. L. (2002). Maturation of brain func-
tion associated with response inhibition. Journal of the American
Academy of Child & Adolescent Psychiatry, 41, 1231-1238.
Taylor, S. F., Kornblum, S., Lauber, E. J., Minoshima, S., & Koeppe,
R. A. (1997). Isolation of specific interference processing in the Stroop
task: PET activation studies. NeuroImage, 6, 81-92.
Taylor, S. F., Kornblum, S., Minoshima, S., Oliver, L. M., & Koeppe,
R. A. (1994). Changes in medial cortical blood flow with a stimulus–
response compatibility task. Neuropsychologia, 32, 249-255.
Thompson-Schill, S. L., D’Esposito, M., Aguirre, G. K., & Farah,
M. J. (1997). Role of left inferior prefrontal cortex in retrieval of se-
mantic knowledge: A reevaluation. Proceedings of the National Acad-
emy of Sciences, 94, 14792-14797.
Turkeltaub, P. E., Eden, G. F., Jones, K. M., & Zeffiro, T. A. (2002).
Meta-analysis of the functional neuroanatomy of single-word reading:
Method and validation. NeuroImage, 16, 765-780.
Ullsperger, M., & von Cramon, D. Y. (2001). Subprocesses of per-
formance monitoring: A dissociation of error processing and response
competition revealed by event-related fMRI and ERPs. NeuroImage,
van Veen, V., Cohen, J. D., Botvinick, M. M., Stenger, V. A., &
Carter, C. S. (2001). Anterior cingulate cortex, conflict monitoring,
and levels of processing. NeuroImage, 14, 1302-1308.
iNterfereNce resolutioN 17 Download full-text
Wager, T. D., Jonides, J., & Reading, S. (2004). Neuroimaging studies
of shifting attention: A meta-analysis. NeuroImage, 22, 1679-1693.
Wager, T. D., Phan, K. L., Liberzon, I., & Taylor, S. F. (2003). Va-
lence, gender, and lateralization of functional brain anatomy in emo-
tion: A meta-analysis of findings from neuroimaging. NeuroImage,
Wager, T. D., & Smith, E. E. (2003). Neuroimaging studies of working
memory: A meta-analysis. Cognitive, Affective, & Behavioral Neuro-
science, 3, 255-274.
Wager, T. D., Sylvester, C. Y., Lacey, S. C., Nee, D. E., Franklin, M.,
& Jonides, J. (2005). Common and unique components of response
inhibition revealed by fMRI. NeuroImage, 27, 323-340.
Watanabe, J., Sugiura, M., Sato, K., Sato, Y., Maeda, Y., Matsue, Y.,
et al. (2002). The human prefrontal and parietal association cortices
are involved in no-go performances: An event-related fMRI study.
NeuroImage, 17, 1207-1216.
Zhang, H. H., Zhang, J., & Kornblum, S. (1999). A parallel distrib-
uted processing model of stimulus–stimulus and stimulus–response
compatibility. Cognitive Psychology, 38, 386-432.
Zysset, S., Muller, K., Lohmann, G., & von Cramon, D. Y. (2001).
Color–word matching Stroop task: Separating interference and re-
sponse conflict. NeuroImage, 13, 29-36.
1. It is clear from this figure that certain peaks seem to lie outside
of the canonical brain (avg152T1.img; SPM, Wellcome Department of
Imaging Neuroscience, www.fil.ion.ucl.ac.uk/spm/). In order to plot all
of the reported peaks into a single brain, coordinates that were reported
in Talairach space were converted to MNI space (www.mrc-cbu.cam.
ac.uk/Imaging/). It is possible either that there are some imperfections
with the transformation tool or that some authors incorrectly reported
that their coordinates were in Talairach space when they were actually
in MNI space, causing the transformation to move these peaks outside
of the canonical brain.
2. Several studies included also looked at patient, younger, or older
populations. Data included in our analyses consisted only of those data
extracted from normal, healthy young adults.
3. We acknowledge that although some reported peaks fall within
white matter or gray matter/white matter boundaries, peaks are more
likely to fall within gray matter. Therefore, it may not be appropriate
to distribute simulated peaks uniformly across gray and white matter.
However, the assumption of uniform distribution across gray and white
matter greatly simplifies the analysis. Some studies do report peaks rela-
tively deep in white matter, whether due to spatial imprecision, neurovas-
cular translation in the BOLD effect, or some other factors. The inclusion
of white matter makes the tests here slightly less conservative than they
would be if we included only some white matter (near gray matter struc-
tures, for example) or only gray matter, but the difference is relatively
small. Indeed, analyses that excluded white matter produced very similar
results (as did simulations that increased the number of Monte Carlo
simulations to 10,000). Therefore, we deem that this method offers a
4. We realize that using peaks ignores the volume and significance
level of activation. In addition, our resolution is limited by the density
radius, and the nonconformity of peaks may derive from variations in
smoothing function images. As a result, we merely propose hypotheses
from our data, rather than drawing conclusions. However, we point out
that this and similar techniques have provided useful results in several
published studies (e.g., Phan et al., 2002; Turkeltaub et al., 2002; Wager
et al., 2004; Wager et al., 2003).
(Manuscript received July 6, 2006;
revision accepted for publication December 21, 2006.)