The Multi-Source Interference Task: validation study with fMRI in individual subjects


Dorsal anterior cingulate cortex (dACC) plays critical roles in cognitive processing, but group-averaging techniques have generally been required to obtain significant dACC activation in functional neuroimaging studies. Development of a task that reliably and robustly activates dACC within individuals is needed to improve imaging studies of neuropsychiatric disorders and localization of dACC in normal volunteers. By combining sources of cognitive interference (Stroop, Eriksen and Simon) with factors known to increase dACC activity, the Multi-Source Interference Task (MSIT) maximally taxes dACC, making it possible to reliably activate dACC within individuals using functional magnetic resonance imaging (fMRI). In this study, eight normal adult volunteers performed the MSIT during fMRI. We compared fMRI responses and performance data between interference and control trials. Significant dACC activation (P<1.7×10−4) was observed in all eight individuals and in the group-averaged fMRI data. In addition to dACC activation, group data also showed activation of presumably networked regions including dorsolateral prefrontal, premotor, and parietal cortices. The MSIT's reaction time interference effect (overall mean 312±61 ms) was up to 10 times greater than that of its component predecessors and temporally stable over hundreds of trials. The robustness, reliability and stability of the neuroimaging and performance data should make the MSIT a useful task with which to study normal human cognition and psychiatric pathophysiology.


Abnormalities of different anterior cingulate cortex (ACC) subdivisions have been implicated in the pathophysiology of many neuropsychiatric disorders such as attention deficit/hyperactivity disorder,1,2 depression,3,4,5 schizophrenia,6,7 Alzheimer's disease,8 anxiety disorders,9,10,11 obsessive compulsive disorder12,13 and post-traumatic stress disorder.14,15 However, ACC is not a homogeneous region, and future progress in elucidating the neural substrates of psychiatric disorders will require refined understanding of these cognitive, motor, and emotional cingulate subdivisions. While appreciating the existence of these functional subdivisions can improve psychiatric research efforts, this alone will not be sufficient to push forward to the next levels, such as the development of functional imaging-based diagnostic tests for neuropsychiatric disorders. To accomplish this, we must develop standardized tasks with sufficient power to reliably and robustly activate ACC subdivisions within individual subjects. Cognitive activation paradigms used with functional magnetic resonance imaging (fMRI) to date, however, generally have not produced robust dorsal ACC (dACC) activation in individuals, and instead have had to rely on group-averaging techniques. The present study details the development and validation of the Multi-Source Interference Task (MSIT)—an fMRI task designed to specifically activate and assess the functional integrity of the ACC subdivision in selected neuropsychiatric disorders. The MSIT combines multiple dimensions of cognitive interference (ie Stroop,16 Eriksen,17 and Simon18) with decision-making and other factors known to activate dACC (target detection, novelty detection, error detection, response selection, stimulus/response competition, and task difficulty)19—thereby maximally recruiting dACC neurons and providing a robust fMRI response within this cortical area in individual subjects.

Convergent evidence from anatomical, connectionist, electrophysiological, and functional neuroimaging sources has established that dACC plays critical roles in complex cognitive and motor processing20,21,22,23,24,25 and reward-based decision-making.19 Reviews and meta-analyses have differentiated dACC from other cingulate subdivisions, including perigenual anterior cingulate cortex (pACC), which is involved in emotional processing,3,10,21,26,27,28,29,30 and the caudal cingulate motor area (cCMA), which assists with simple motor movements.22,29,31,32 Furthermore, task-dependent reciprocal responses have been repeatedly observed in dACC and pACC, (that is, complex cognitive tasks produce increased dACC activity and decreased pACC activity, while tasks involving emotion conversely produce decreased dACC activity and increased pACC activity.3,21,23,33,34

The MSIT's design was hypothesis-driven, emerging from the conceptualization of dACC as an admixture of heterogeneous neurons that variously anticipate and detect targets, indicate novelty, influence motor response selection, signal rewards and detect errors.19 This ‘dACC heterogeneity model’ is built upon a large body of monkey single unit recording studies that have established that dACC encompasses such heterogeneous neuron types,35,36,37,38,39 and was supported by the results from a recent human fMRI study using a reward-based decision-making task.19 In a related manner, human functional imaging studies of anticipation,40,41 target detection,42 novelty detection,43,44 response selection,20,22,45,46,47,48,49,50,51 reward,52,53 and error detection44,54,55,56,57 have all independently activated dACC, supporting the heterogeneity conceptualization.

Cognitive interference tasks, in which the processing of one stimulus feature impedes the simultaneous processing of a second stimulus attribute, have also activated dACC in group-averaged functional neuroimaging studies. These would include Stroop and Stroop-like tasks,1,20,46,49,58,59,60,61,62,63,64,65,66 Eriksen Flanker-type tasks,67,68,69 and Simon effect task variants.20 This is not surprising, as such tasks place high demands on target detection, response selection, and performance monitoring circuits. Interestingly, a recent study by Grachev et al70 went beyond the traditional forms of functional imaging and, using the Color Stroop and in vivo proton magnetic resonance spectroscopy, found biochemical differences in dACC that were correlated with the degree of cognitive interference (ie N-acetyl aspartate levels in the right dACC were significantly lower in subjects showing greater cognitive interference as measured by the Stroop reaction time interference effect).

While full descriptions of the cognitive interference tasks that the MSIT was derived from are beyond the scope of this paper, brief descriptions are given here. The essence of the prototypical cognitive interference task, the Color Stroop,16,59 is that subjects take longer to name the color of the ink that color words are written in when the ink color and color word are incongruent (eg the word red written in blue ink, correct answer ‘blue’) than when they do match (blue written in blue ink) or when the word is a noncolor word (house written in blue ink). In the Eriksen Flanker Task17 subjects take longer to identify a centrally located target letter (and make more errors) when the target letter is flanked by incongruent distractor letters (eg DDTDD) than when it is flanked by the same letter (eg TTTTT). The Simon Effect18 denotes the cognitive interference produced by spatial incongruence between the target and response (eg given a two-button response pad, it takes longer for subjects to respond using the left button when the target appears on the right [and vice versa] than when the stimulus and response positions correspond).

An ideal functional neuroimaging-based diagnostic test to be used in assessing the functional integrity of dACC or any other cortical structure(s) in neuropsychiatric disorders should possess the following characteristics: (1) It must produce reliable and robust activation of the cortical region(s)-of-interest (ROI) within healthy individuals. (2) It should be hypothesis-driven, (that is, pre-existing evidence should support a mechanism explaining why the task would be expected to recruit the RO). (3) It should include collection of concomitant imaging and performance data (reaction times and accuracy). (4) Testing procedures must be standardized. (5) The task instructions should be easy to learn and retain so that the task can be performed by subjects with impaired cognition (eg schizophrenia) and by subjects across a wide age spectrum (to enable developmental studies in children and studies of elderly subjects). (6) It should not require an excessive time commitment as children and elderly subjects tend to tire more easily than young adults. (7) It should not be language-specific (in order to allow cross-cultural studies). (8) Performance data should vary within a relatively narrow range in healthy volunteers. (9) Imaging and performance data should be related. (10) Imaging and performance data should show temporal stability (ie it should display sufficient test–retest reliability to permit longitudinal and treatment studies). (11) Imaging and performance data should be sensitive to changes with successful treatment. (12) Results should be disorder specific.

The MSIT was designed with attention to these characteristics. Our primary objective in this validation study was to show that the interference minus control comparison would produce reliable and robust dACC activation within individual subjects. Based on the available evidence cited above, we also predicted that in contrast to dACC activation, performance of the MSIT would yield decreased pACC activity and no significant difference in cCMA activity. Finally, we predicted that reaction times would be longer during interference trials than during control trials.

Materials and methods


Informed consent was obtained following the established guidelines of the Massachusetts General Hospital Subcommittee on Human Subjects. All eight subjects (four males, four females, mean age 30.4 years [s.d. 5.6 years]) were strongly right-handed as per the Edinburgh Handedness Inventory.71 All had normal or corrected-to-normal vision. No subject had a history of neurological, major medical, or psychiatric disorder, nor did any have a history of serious head injury. None were taking medication.

Task description

For the MSIT (see Figure 1), subjects were given a button-press and instructed that the keypad buttons represented one, two, and three from left to right. They were told to use the index, middle and ring fingers of the right hand to respond. They were instructed that sets of three numbers (1and/or 2 and/or 3) and/or letters (x) would appear in the center of the screen every 1.75 s, and that one number would always be different from the other two (matching distractor) numbers or letters. Subjects were asked to report, via button-press, the identity of the number that was different from the other two items.

Figure 1

MSIT trial examples. Examples of single trials of the two types of stimuli are shown. Subjects are asked to report, via button-press, the identity of the number that differs from the other two items. During ‘control’ trials, the distractors were always the letter ‘x’, target numbers were always large and placed congruently with their position on the button box. During ‘interference’ trials, the distractors were other numbers (either 1, 2, or 3), target numbers could be large or small and they were never placed congruently with their position on the button box. In both top examples, the correct answer would be to press button ‘1’, and in both bottom examples, the correct answer would be to press button ‘2’.

Subjects were informed that scans would begin and end with fixation of a white dot for 30 s, and that between these times there would be two trial types that would appear in alternating 42 s long blocks. In trials with numbers and letters (control trials), the target number would always match its position on the button-press (eg the number ‘1’ would appear in the first [leftmost] position). In contrast, during the trials involving all numbers (interference trials), the target would never match its position. It was then emphasized that they were to ‘report what the target number was regardless of its position’. Further, subjects were informed that in the numbers and letters (control) condition, the target number would always be large, while in the numbers only (interference) condition the target number would sometimes be large and sometimes be small (in actuality, unbeknownst to subjects, the number of large target trials and small target interference trials were equal). For all trials, subjects were instructed to answer as quickly as possible but to make sure that they gave the right answer.

After instructions were reviewed, and just prior to entering the scanner, subjects completed a 1 min long computerized practice version of the task (20 control trials followed by 20 interference trials). Subjects completed four scans each of the MSIT, with each scan lasting 6 min and 36 s. RTs for erroneous or null responses were discarded.

Subjects completed 24 trials during each (control/interference) block, 96 trials of each type during a single scan, and 384 total trials of each type during the four-scan session (except for one subject who, due to technical problems, only had three scans). The order of presentation, regarding the control and interference blocks, was fixed within runs (ie it was always FCICICICIF). Four pseudorandom sequences were used—these were counterbalanced across subjects. Within each block each stimulus appeared an equivalent number of times (ie each of the three control stimuli 1xx, x2x, xx3 appeared eight times in each block, while each of the 24 different interference stimuli combinations appeared once in each block). Eye movements were not monitored.

fMRI procedures

fMRI was performed in a Siemens 3 T Allegra high-speed echo-planar imaging device (Munich, Germany) using a quadrature head coil (see Bush et al19 and Sperling et al72). Subjects lay on a padded scanner couch in a dimly illuminated room, and wore foam earplugs. Foam padding stabilized the head. Stimuli were generated via MacStim 2.2.1 (West Melbourne, Australia) on a Macintosh 250 MHz Powerbook™ (Cupertino, CA,USA) and projected onto a screen secured to the head coil. Subjects viewed images on a tilted mirror placed in front of their eyes. Stimuli subtended approximately 1° of the visual angle vertically.

High-resolution structural images (1.0×1.0×1.3 mm,3 magnetization-prepared rapid acquisition with gradient echoes (MP-RAGE), 128 slices, 2562 matrix, echo time (TE)=3.3 ms; repetition time (TR)=30 ms; flip=40°) were collected for 3-D reconstruction either during the functional scanning session or in a separate session. Functional sessions began with an initial sagittal localizer scan, followed by a set of high-resolution (22 coronal slices, perpendicular to the anterior commissure–posterior commissure line and extending posterior from a point approximately y=60 mm, 1.5 mm2 in-plane resolution (ipr) × 5 mm thick, skip 1 mm) inversion time T1-weighted echo-planar images (inversion time (TI)/TE/TR=1200/29/6000 ms; number of excitations (NEX)= 4) and T2 conventional anatomical scans (2562 matrix, TI/TE/TR=1200/104/11 ms, NEX=2). The co-registered functional series (TE/TR=30/1500 ms, 264 images/slice, flip=90°, FOV=20 cm2, 642 matrix, ipr=3.125 mm2) lasted 6 min and 36 s. Subjects completed four functional scans for a total of 1056 total images per subject. Data sets were motion-corrected using AFNI (Milwaukee, WI, USA) and normalized to represent percent signal change from the mean signal during the fixation condition.

A voxel-wise t-test was employed to test if greater activation occurred in dACC during interference trials vs control trials. The dACC ROI was defined based on a modification of a meta-analysis of 64 imaging studies that reported ACC activation during cognitively demanding tasks.19,21 The modification (added to better assess homology, based on the monkey electrophysiology studies and anatomical work in humans) was that only ACC cortex superior to the corpus callosum between y=0 and +30 mm and within the cingulate sulcus (or between the cingulate and paracingulate sulci inclusive in cases displaying double parallel cingulate sulci) was considered.19 It refers to the same cortical region that has been called the anterior cingulate cognitive division (ACcd),21 rostral cingulate zone,22 or midcingulate cortex.8 This was done for each subject on his/her anatomical scan (see Figure 2). For this a priori defined dACC region encompassing 250 voxels, statistical significance (of P<0.05 Bonferroni-corrected for multiple comparisons) was defined as P<1.7×10−4. Resultant statistical maps were displayed in pseudocolor, scaled according to significance, projected onto the high-resolution anatomical scan slices in native and Talairach space.73

Figure 2

Scan and reaction time data: group-averaged and individuals. Top: Group-averaged fMRI and RT results are shown in the larger top panels. Scan data are displayed on group-averaged slices in radiological convention (coronal slice at y=17 mm [left] and axial slice at z=24 mm [right]). On the coronal slice, the aqua box represents the group-averaged ROI. The gray box indicates the approximate borders of the enlarged areas shown for individuals (below). The axial slice highlights activity in other presumably networked regions such as dorsolateral prefrontal cortex (DLPFC) and posterior parietal cortex (PPC)—other regions that are commonly activated by Stroop-like tasks (for a full listing of activated regions, please refer to Table 2). At right, mean RTs are shown for the 16 interference–neutral word block pairs collected while subjects were being scanned. Black bars represent interference trials, gray bars control trials, and the thick black line represents the difference between the two. Error bars indicate standard errors of the mean. Bottom: In the eight panels below, fMRI and RT data are shown for each of the individuals. Aqua lines indicate the dACC ROI for that slice for each individual. Significance (color) scales are identical for all scan data in this figure.

We also conducted group-averaged analyses in Talairach atlas space using voxel-wise t-tests to test if greater activation occurred in dACC during interference trials vs control trials. This analysis was conducted for all interference minus control blocks both overall (all scans from all subjects) and on a scan-by-scan basis for the first three scans from all subjects (since one subject only completed three scans). For these group evaluations, the dACC ROI was defined as in past group studies, that is, anterior cingulate cortex anterior to y=0 mm, posterior to y=+30 mm, and within 15 mm of the midline.1 For completeness, while our main interest lies in evaluating dACC response, loci of activation in other regions meeting the same threshold are reported (although these post hoc identified regions should be independently and prospectively confirmed).

To quantify the degree to which the locations of individual dACC activations matched that of the group-averaged data, we calculated the mean distance (and SD) of each individual maximum from the ipsilateral group average (for the left hemisphere we measured the distances from individual maxima to a point midway between the two left-sided dACC maxima, which were found in close proximity to one another).


Behavioral results

Figure 2 displays the robust and reliable RT data and illustrates how the performance data (and scan data) of the individuals closely matches that of the group as a whole. Planned comparisons revealed that, as a group, subjects showed a highly significant overall mean increase in RT of 312 ms (±61 ms, s.d.) during interference condition as compared to the control condition (t=13.6, df=6, P<0.001). In fact, this RT increase during the interference condition was highly significant for each scan individually: Scan 1 (t=11.0, df=6, P<0.001), Scan 2 (t=13.3, df=6, P<0.001), Scan 3 (t=10.8, df=6, P<0.001), and Scan 4 (t=18.6, df=6, P<0.001). Thus, the RT interference effect remains significant throughout the entire experiment (Figure 2, Table 1). Subjects showed a significant increase in errors during interference trials as compared to control trials ( Table 1, paired two-tailed Student's t-test, P<0.01).

Table 1 Performance data. Group-averaged reaction time and accuracy data during different trial types are shown

The effects of target characteristics on RT and accuracy within interference trials were examined ( Table 1). Target size did not have an appreciable effect on either RT or accuracy. A repeated-measures ANOVA revealed that RTs varied across target position (F=5.6, df =2,14, P<0.02). Follow-up t-tests revealed that RTs were significantly greater when the target was in position 1 vs position 2 (t=3.3, df=7, P<0.02) and marginally greater when the target was in position 1 vs position 3 (t=2.1, df=7, P<0.07). RTs for targets in positions 2 and 3 did not differ from each other (t=0.6, df=7, P<0.55). Error rates did not differ across target positions (F=0.6, df=2,14, P<0.58).

fMRI results

As illustrated in Figures 2 and 3, dACC was significantly activated (P<1.7×10−4) in eight out of eight single subjects (and in the group average) when comparing all interference blocks with all control blocks (ie 16 block-pairs per subject). These figures show the tight spatial correlation between the individual and group data (the mean distance of individual dACC activations from the group average was only 8.5 mm [SD 6.1 mm]). Additionally, the group data also showed temporal stability in that significant dACC activation (P<10−4) was observed when comparing interference blocks with control blocks within each of the first three scans.

Figure 3

MSIT: comparison to meta-analysis. This figure plots the group (aqua stars) and individual (small aqua squares) fMRI activation data observed during the MSIT atop the results from a previous meta-analysis of cognitive and emotional neuroimaging studies.21 The dACC is activated by cognitively challenging tasks (red dots, yellow, and orange triangles) such as Stroop and Stroop-like tasks, divided attention tasks, and complex response selection tasks. The pACC is activated by tasks that relate to affective or emotional content, or symptom provocation (blue dots and blue diamond). As predicted, the MSIT activated dACC, and the individual data clustered tightly around the group-averaged maxima. Also, the MSIT activations overlapped those observed in this same group of subjects when performing a reward-based decision-making task (the purple triangle indicates group maxima, and the purple rectangle indicates the limited spatial extent of the activations within individuals)—see Bush et al.19 Abbreviation: CC, corpus callosum.

As expected, prospective analysis revealed significant (P<1.7×10−4) bilateral signal decreases in the pACC (2 foci: x/y/z of 7/45/9 and −2/45/3) during the interference blocks (relative to the control blocks). Also as predicted, we did not observe significant activation in cingulate cortex caudal to dACC in the interference minus control contrast.

While our focus was on dACC, group-averaged fMRI data also showed that the MSIT coactivated other brain regions involved in cognition, target detection, response selection, motor planning, and motor output. These areas included dorsolateral prefrontal cortex, premotor cortex, and parietal cortex (see Table 2). These group effects are likely veridical, as these same areas were significantly activated in at least 75% of the individuals.

Table 2 Regions activated by the MSIT


There were three principal findings:

(1) The MSIT produced reliable and robust fMRI activation of dACC within individual subjects. Activation observed in dACC within single subjects was tightly clustered around the dACC foci identified by group averaging of data.

(2) Predictably, in contrast to dACC activation during the interference minus control comparison, fMRI signal decreased in pACC, and did not increase significantly in either the caudal cingulate motor zone or posterior cingulate cortex. Together, the findings support the existence of functional subdivisions of cingulate cortex.

(3) The MSIT successfully produced a state of cognitive interference, as indicated by prolonged reaction times and increased errors during interference blocks as compared to control blocks. As with the fMRI data, the individuals’ performance data closely paralleled the group data.

Issues for studies of clinical populations

The MSIT possesses many of the qualities deemed desirable in a functional neuroimaging-based diagnostic test: (1) It reliably and robustly activates dACC within individuals. (2) Mechanistically, it is hypothesis-driven. (3) It permits collection of concomitant imaging and performance data. (4) Testing procedures are standardized. (5) The task instructions are easy to learn and retain so that the task can be performed by subjects with impaired cognition and by subjects across a wide age spectrum. (6) It can be completed in a short period of time. (7) It is not language-specific, which can facilitate cross-cultural studies (ie at least across those cultures that use Arabic numerals). (8) Performance data varied within a relatively narrow range. (9) Imaging and performance data were related (ie increased RT was associated with higher fMRI signal). (10) Imaging and performance data showed temporal stability. Thus, it appears that the MSIT can be a useful task in studies of neuropsychiatric patients and normal volunteers.

Many of the characteristics that make the MSIT a good candidate for a clinical task would appear to represent substantive advancements over previous tasks. Although many tasks (such as the Color and Counting Stroops, go/no-go tasks, and divided attention tasks) activate dACC reliably,21 these studies have historically relied on group averaging rather than analyses in single subjects. While reward-based decision-making19 and verb generation tasks74 have produced reliable fMRI activation in single subjects and provided valuable information with respect to dACC mechanisms, both present drawbacks in terms of clinical use. The reward-based decision-making task must be run in event-related format, and due to design issues such as infrequent target presentation, requires extensive averaging and nearly 2 h of dedicated scanning time. In contrast, the MSIT took less than an hour and potentially could be completed in about 20 min. The verb generation task has variable and nonstandardized output (making it difficult to quantify and compare performance) and, as Raichle et al75 showed, is extremely sensitive to minimal practice. Practice-related decrements in dACC activation, also observed within the first few minutes of the Counting Stroop62 were desirable in terms of understanding learning-related activity in dACC—but this type of temporal instability is an undesirable trait for a clinical test. In contrast, the MSIT yields temporally stable performance and imaging data, making it useful in treatment and other types of longitudinal studies.

It is important to note, however, that a number of imaging parameters were chosen to specifically maximize individual activation, and these alone (rather than factors specific to the MSIT) could conceivably have led to the enhanced ability to detect activation within individuals. For example, the current study used a high field strength 3 T Siemens magnet, and averaged four runs per individual, whereas most previous tasks have been run on 1.5 T magnets and averaged only two runs. Prior to developing the MSIT, we did conduct pilot studies of Color and Counting Stroops, go/no-go tasks, and other response override tasks (using up to four runs on a different high field General Electric Signa 3 T magnet [Milwaukee, WI, USA]) and these did not produce consistent single subject dACC activation, but this does not conclusively demonstrate that the MSIT is a ‘superior’ task, nor should one conclude that the other tasks are incapable of producing single subject activation with equivalent scanning parameters. Future comparative studies of the different tasks using the identical subjects, scan parameters and analysis methods might produce interesting results.

The RT interference effect (mean 312 ms) was extremely robust. It was much larger than has typically been seen in imaging studies of the Color Stroop (mean=142 ms, range=97–235 ms),58,60,64,65,76,77 the Counting Stroop (mean=38 ms, range=29–46 ms),1,62 Eriksen Flanker-type tasks (mean=102 ms, range 19–267 ms),67,68,69 or Simon effect task variants (39 ms in Hazeltine et al,78 and approximately 10–30 ms in Barch et al20). The fact that the MSIT's mean RT interference was at least double those of these other tasks indicates that the strategy of combining tasks to maximize cognitive interference was successful.

The ability to produce activation within individuals is particularly valuable to patient-based studies. Better localization will improve power to detect differences in future studies. This can help studies of normal function of dACC as well as patient/group studies. It can also help future patient studies by permitting the elimination of a potential confound (ie apparent ‘relative hypofunction’ of a brain region in a patient group (compared to the control group) could be spuriously caused not by actual hypofunction within the individuals with the disorder, but rather by greater spatial heterogeneity of the location of a normally functioning brain region in the patient group). Of course, while we stress the value of the individual study, group-averaged MSIT data can also be used with the advantages of greater power, fewer subjects, and higher confidence.

In addition to combining Stroop, Eriksen, and Simon interference effects, the MSIT was designed to recruit as many dACC neurons as possible (anticipation, target detection, novelty detection, error detection, response selection) in order to maximally activate dACC. Interference blocks (in block format) were expected to induce greater anticipation and designed to maximally tax response selection. Also, target detection and novelty detection requirements were greater during interference blocks, since these included 24 different stimulus combinations, as compared to three control task stimuli. Errors were significantly higher during the interference condition. It should be noted that while no specific single cell representations have been shown to exist within dACC for stimulus and/or response competition detection56 and task difficulty,79 these are factors that the ‘heterogeneity model of dACC’ predicts would, if differentially present between parts of a task, lead to increased dACC activation.19

Admittedly, the interference and control portions of the task were not perfectly matched in all characteristics save one, as is the goal in traditional psychology experiments. However, it must be emphasized that this was by design, since the goal here was not to study some aspect of dACC function in isolation, but rather to develop an adequately controlled task that would robustly activate dACC. The interference and control blocks did have virtually identical visual and motor requirements—beyond that, task design decisions were made to maximize dACC activity. Future work should be done using different permutations of the MSIT in order to study the effects of changing different task characteristics, such as target size, position, novelty, etc. For example, one could examine the effects of instructing subjects to report the position of the target stimulus rather than its identity.

Dorsal ACC activation in individuals—implications for studies of normal cognition

The MSIT advances our understanding of the localization of dACC in normal humans. Figure 3 shows that it activated the same area that has been activated repeatedly by other cognitively demanding tasks,21 as well as by the reward-based decision-making task that was performed by the same subjects in a different session.19 Both the MSIT and the reward-based decision-making task yielded bilateral activation in the majority of these individuals, even though the button-press responses were performed solely with the right hand in both studies, tacitly arguing in favor of a more generalized cognitive integration function rather than a lateralized, solely motor function for dACC.

Different functions have been ascribed to dACC, including attention-for-action/target selection,80,81 motivational valence assignment,63,82,83 motor response selection,47,50,51,79 error detection/performance monitoring,57,84 competition monitoring,56,67 anticipation40,63 working memory,85 novelty detection,43,44,63 reward assessment,52,53 and reward-based decision-making.19 Monkey single-unit recording studies have established that dACC is comprised of an admixture of heterogeneous cell types that anticipate and detect targets, encode novelty, influence motor preparation and execution, evaluate rewards, and signal errors,35,36,37,38,39 and fMRI results in humans indicate that a similar architecture exists in humans.19 While much work needs to be done to validate such a heterogeneity-based ‘local intracortical network model of dACC’, the heterogeneity conceptualization has the capacity to be refined into a unifying, neurobiologically plausible model that can explain the diverse results from neuroimaging and electrophysiological studies.19,21 It should be noted that the construction of the MSIT task was based on, but not dependent on, this heterogeneity concept. Thus, while the data are consistent with the heterogeneity concept, they do not rule out other proposed dACC mechanisms.

The results support the existence of separable dorsal (dACC) ‘cognitive’ and perigenual (pACC) ‘emotional’ subdivisions of ACC. The MSIT activated dACC, but not pACC. Similar dACC activity has been obtained in response to cognitively challenging tasks, and can be contrasted with activity seen in pACC (rostral/subgenual areas 24/25/32/33) during processing of emotional stimuli.21 Notably, in companion studies run in the same group of subjects during the same scanning session, the cognitive Counting Stroop activated dACC,62 while the emotional Counting Stroop activated pACC.10 The current data also are consistent with meta-analyses of ACC activations that showed reciprocal deactivations (ie decreases in the ‘cognitive’ dACC during emotional tasks and decreases in the ‘emotional’ pACC during ‘cognitive’ tasks,3,21,23 as the MSIT produced significantly less fMRI signal in pACC during the interference blocks than during the control blocks. Obviously, the terms ‘cognitive’ and ‘emotional’ are gross over-simplifications, and the actual relevance of ‘deactivations’ in imaging studies must be elucidated, but the MSIT data are consistent with previous findings supporting functional subdivisions of ACC.

No significant activation was observed in the caudal cingulate motor zone, which many conclude is involved in the execution of simple motor acts,22,29 or posterior cingulate cortex, which is involved in visuospatial processing.29,86 Together with the dACC and pACC data, these findings also support the conclusion that cingulate cortex contains functional subdivisions that differentially subserve cognitive, motor, and emotional processing.

Other brain regions

Although the current study focused on dACC, it should be emphasized that this in no way implies that dACC is the only region responsible for processing cognitive interference tasks. The MSIT also recruited many other brain regions including dorsolateral prefrontal cortex, premotor areas, and parietal cortex. This was expected, as many researchers have concluded that these structures subserve cognitive processing in parallel-distributed fashion.82,87,88 DLPFC has often been reported to be coactivated with dACC during cognitive tasks,24,89,90,91 premotor cortex is responsible for planning and execution of nonautomatic tasks,92,93 and parietal cortex has been shown to be activated during target detection tasks94,95 and Stroop tasks.1,58,62,64,96 Furthermore, interesting recent work by different groups utilizing combined transcranial magnetic stimulation (TMS) and neuroimaging has shown that TMS of lateral prefrontal cortex can modulate ACC activity97,98 indicating the strong inter-regional connectivity and suggesting another possible direction for research (ie technique refinements may produce dACC activation via remote TMS stimulation). While the precise roles these structures play in such tasks remain to be determined, the convergent data argue that they interact as a network, and the challenge ahead is to determine the role of the individual components of these distributed brain circuits.


The MSIT produced reliable and robust activation of dACC in individuals and this closely matched the group-averaged data. Coupled with the reliable and robust performance data, the MSIT displayed many of the characteristics desired in a neuroimaging-based diagnostic test. As such, it can be expected to serve as a useful fMRI probe in searching for the neural substrates of various neuropsychiatric disorders such as attention deficit disorder, schizophrenia, obsessive–compulsive disorder, and depression. Future studies can also use task manipulations to help us to better understand mechanisms of attention, response selection, and cognition.


  1. 1

    Bush G, Frazier JA, Rauch SL, Seidman LJ, Whalen PJ, Jenike MA et al. Anterior cingulate cortex dysfunction in attention-deficit/hyperactivity disorder revealed by fMRI and the Counting Stroop. Biol Psychiatry 1999; 45: 1542–1552.

  2. 2

    Swanson J, Castellanos FX, Murias M, LaHoste G, Kennedy J . Cognitive neuroscience of attention deficit hyperactivity disorder and hyperkinetic disorder. Curr Opin Neurobiol 1998; 8: 263–271.

  3. 3

    Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin RK, Jerabek PA et al. Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 1999; 156: 675–682.

  4. 4

    Drevets WC . Neuroimaging studies of mood disorders. Biol Psychiatry 2000; 48: 813–829.

  5. 5

    George MS, Ketter TA, Parekh PI, Rosinsky N, Ring HA, Pazzaglia PJ et al. Blunted left cingulate activation in mood disorder subjects during a response interference task (the Stroop). J Neuropsychiatry Clin Neurosci 1997; 9: 55–63.

  6. 6

    Carter CS, MacDonald 3rd AW, Ross LL, Stenger VA . Anterior cingulate cortex activity and impaired self-monitoring of performance in patients with schizophrenia: an event-related fMRI study. Am J Psychiatry 2001; 158: 1423–1428.

  7. 7

    Kiehl KA, Liddle PF . An event-related functional magnetic resonance imaging study of an auditory oddball task in schizophrenia. Schizophr Res 2001; 48: 159–171.

  8. 8

    Vogt BA, Vogt LJ, Nimchinsky EA, Hof PR . Primate cingulate cortex chemoarchitecture and its disruption in Alzheimer's disease. In: Bloom FE, Bjorklund A, Hokfelt T (Eds.) Handbook of Chemical Neuroanatomy, The Primate Nervous System. Amsterdam: Elsevier, 1997, pp 455–528.

  9. 9

    Simpson Jr JR, Drevets WC, Snyder AZ, Gusnard DA, Raichle ME . Emotion-induced changes in human medial prefrontal cortex: II. During anticipatory anxiety. Proc Natl Acad Sci USA 2001; 98: 688–693.

  10. 10

    Whalen PJ, Bush G, McNally RJ, Wilhelm S, McInerney SC, Jenike MA et al. The emotional counting Stroop paradigm: a functional magnetic resonance imaging probe of the anterior cingulate affective division. Biol Psychiatry 1998; 44: 1219–1228.

  11. 11

    Benkelfat C, Bradwejn J, Meyer E, Ellenbogen M, Milot S, Gjedde A et al. Functional neuroanatomy of CCK4-induced anxiety in normal healthy volunteers. Am J Psychiatry 1995; 152: 1180–1184.

  12. 12

    Gehring WJ, Himle J, Nisenson LG . Action-monitoring dysfunction in obsessive-compulsive disorder. Psychol Sci 2000; 11: 1–6.

  13. 13

    Rauch SL, Jenike MA, Alpert NM, Baer L, Breiter HC, Savage CR, et al. Regional cerebral blood flow measured during symptom provocation in obsessive-compulsive disorder using oxygen 15-labeled carbon dioxide and positron emission tomography. Arch Gen Psychiatry 1994; 51: 62–70.

  14. 14

    Shin LM, Whalen PJ, Pitman RK, Bush G, Macklin ML, Lasko NB et al. An fMRI study of anterior cingulate function in posttraumatic stress disorder. Biol Psychiatry 2001; 50: 932–942.

  15. 15

    Shin LM, McNally RJ, Kosslyn SM, Thompson WL, Rauch SL, Alpert NM et al. Regional cerebral blood flow during script-driven imagery in childhood sexual abuse-related PTSD: A PET investigation. Am J Psychiatry 1999; 156: 575–584.

  16. 16

    Stroop JR . Studies of interference in serial verbal reactions. J Exp Psychol 1935; 18: 643–662.

  17. 17

    Eriksen BA, Eriksen CW . Effects of noise letters upon the identification of a target letter in a nonsearch task. Perception Psychophysics 1974; 16: 143–149.

  18. 18

    Simon JR, Berbaum K . Effect of conflicting cues on information processing: the ‘Stroop effect’ vs the ‘Simon Effect’. Acta Psychol 1990; 73: 159–170.

  19. 19

    Bush G, Vogt BA, Holmes J, Dale AM, Greve D, Jenike MA et al. Dorsal anterior cingulate cortex: a role in reward-based decision making. Proc Natl Acad Sci USA 2002; 99: 523–528.

  20. 20

    Barch DM, Braver TS, Akbudak E, Conturo T, Ollinger J, Snyder A . Anterior cingulate cortex and response conflict: effects of response modality and processing domain. Cereb Cortex 2001; 11: 837–848.

  21. 21

    Bush G, Luu P, Posner MI . Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci 2000; 4: 215–222.

  22. 22

    Picard N, Strick PL . Motor areas of the medial wall: a review of their location and functional activation. Cereb Cortex 1996; 6: 342–353.

  23. 23

    Drevets WC, Raichle ME . Reciprocal suppression of regional cerebral blood flow during emotional versus higher cognitive processes: implications for interactions between emotion and cognition. Cognition Emotion 1998; 12: 353–385.

  24. 24

    Duncan J, Owen AM . Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends Neurosci 2000; 23: 475–483.

  25. 25

    Paus T . Primate anterior cingulate cortex:where motor control, drive and cognition interface. Nat Rev Neurosci 2001; 2: 417–424.

  26. 26

    Devinsky O, Morrell MJ, Vogt BA . Contributions of anterior cingulate cortex to behaviour. Brain 1995; 118 (Part 1): 279–306.

  27. 27

    Mega MS, Cummings JL, Salloway S, Malloy P . The limbic system: an anatomic, phylogenetic, and clinical perspective. J Neuropsychiatry Clin Neurosci 1997; 9: 315–330.

  28. 28

    Allman JM, Hakeem A, Erwin JM, Nimchinsky E, Hof P . The anterior cingulate cortex. The evolution of an interface between emotion and cognition. Ann N Y Acad Sci 2001; 935: 107–117.

  29. 29

    Vogt BA, Finch DM, Olson CR . Functional heterogeneity in cingulate cortex: the anterior executive and posterior evaluative regions. Cereb Cortex 1992; 2: 435–443.

  30. 30

    Posner MI, Rothbart MK . Attention self-regulation and consciousness. Philos Trans R Soc London B Biol Sci 1998; 353: 1915–1927.

  31. 31

    Picard N, Strick PL . Imaging the premotor areas. Curr Opin Neurobiol 2001; 11: 663–672.

  32. 32

    Shima K, Aya K, Mushiake H, Inase M, Aizawa H, Tanji J . Two movement-related foci in the primate cingulate cortex observed in signal-triggered and self-paced forelimb movements. J Neurophysiol 1991; 65: 188–202.

  33. 33

    Gusnard DA, Akbudak E, Shulman GL, Raichle ME . Medial prefrontal cortex and self-referential mental activity: relation to a default mode of brain function. Proc Natl Acad Sci USA 2001; 98: 4259–4264.

  34. 34

    Simpson Jr, JR, Snyder AZ, Gusnard DA, Raichle ME . Emotion-induced changes in human medial prefrontal cortex: I. During cognitive task performance. Proc Natl Acad Sci USA 2001; 98: 683–687.

  35. 35

    Koyama T, Kato K, Tanaka YZ, Mikami A . Anterior cingulate activity during pain-avoidance and reward tasks in monkeys. Neurosci Res 2001; 39: 421–430.

  36. 36

    Niki H, Watanabe M . Prefrontal and cingulate unit activity during timing behavior in the monkey. Brain Res 1979; 171: 213–224.

  37. 37

    Nishijo H, Yamamoto Y, Ono T, Uwano T, Yamashita J, Yamashima T . Single neuron responses in the monkey anterior cingulate cortex during visual discrimination. Neurosci Lett 1997; 227: 79–82.

  38. 38

    Shima K, Tanji J . Role for cingulate motor area cells in voluntary movement selection based on reward. Science 1998; 282: 1335–1338.

  39. 39

    Procyk E, Tanaka YL, Joseph JP . Anterior cingulate activity during routine and non-routine sequential behaviors in macaques. Nat Neurosci 2000; 3: 502–508.

  40. 40

    Murtha S, Chertkow H, Beauregard M, Dixon R, Evans A . Anticipation causes increased blood flow to the anterior cingulate cortex. Hum Brain Mapping 1996; 4: 103–112.

  41. 41

    Critchley HD, Mathias CJ, Dolan RJ . Neural activity in the human brain relating to uncertainty and arousal during anticipation. Neuron 2001; 29: 537–545.

  42. 42

    Shulman GL, Ollinger JM, Linenweber M, Petersen SE, Corbetta M . Multiple neural correlates of detection in the human brain. Proc Natl Acad Sci USA 2001; 98: 313–318.

  43. 43

    Clark VP, Fannon S, Lai S, Benson R, Bauer L . Responses to rare visual target and distractor stimuli using event-related fMRI. J Neurophysiol 2000; 83: 3133–3139.

  44. 44

    Braver TS, Barch DM, Gray JR, Molfese DL, Snyder A . Anterior cingulate cortex and response conflict: effects of frequency, inhibition and errors. Cereb Cortex 2001; 11: 825–836.

  45. 45

    Mesulam MM, Nobre AC, Kim YH, Parrish TB, Gitelman DR . Heterogeneity of cingulate contributions to spatial attention. Neuroimage 2001; 13(6 Part 1): 1065–1072.

  46. 46

    Peterson BS, Skudlarski P, Gatenby JC, Zhang H, Anderson AW, Gore JC . An fMRI study of Stroop word–color interference: evidence for cingulate subregions subserving multiple distributed attentional systems. Biol Psychiatry 1999; 45: 1237–1258.

  47. 47

    Badgaiyan RD, Posner MI . Mapping the cingulate cortex in response selection and monitoring. Neuroimage 1998; 7: 255–260.

  48. 48

    Elliott R, Dolan RJ . Activation of different anterior cingulate foci in association with hypothesis testing and response selection. Neuroimage 1998; 8: 17–29.

  49. 49

    Pardo JV, Pardo PJ, Janer KW, Raichle ME . The anterior cingulate cortex mediates processing selection in the Stroop attentional conflict paradigm. Proc Natl Acad Sci USA 1990; 87: 256–259.

  50. 50

    Paus T, Petrides M, Evans AC, Meyer E . Role of the human anterior cingulate cortex in the control of oculomotor, manual, and speech responses: a positron emission tomography study. J Neurophysiol 1993; 70: 453–469.

  51. 51

    Turken AU, Swick D . Response selection in the human anterior cingulate cortex. Nat Neurosci 1999; 2: 920–924.

  52. 52

    Knutson B, Westdorp A, Kaiser E, Hommer D . FMRI visualization of brain activity during a monetary incentive delay task. Neuroimage 2000; 12: 20–27.

  53. 53

    Kunig G, Leenders KL, Martin-Solch C, Missimer J, Magyar S, Schultz W . Reduced reward processing in the brains of Parkinsonian patients. Neuroreport 2000; 11: 3681–3687.

  54. 54

    Gehring WJ, Fencsik DE . Functions of the medial frontal cortex in the processing of conflict and errors. J Neurosci 2001; 21: 9430–9437.

  55. 55

    Kiehl KA, Liddle PF, Hopfinger JB . Error processing and the rostral anterior cingulate: an event-related fMRI study. Psychophysiology 2000; 37: 216–223.

  56. 56

    Carter CS, Braver TS, Barch DM, Botvinick MM, Noll D, Cohen JD . Anterior cingulate cortex, error, detection and the online monitoring of performance. Science 1998; 280: 747–749.

  57. 57

    Luu P, Flaisch T, Tucker DM . Medial frontal cortex in action monitoring. J Neurosci 2000; 20: 464–469.

  58. 58

    Banich MT, Milham MP, Atchley R, Cohen NJ, Webb A, Wszalek T et al. fMRI studies of Stroop tasks reveal unique roles of anterior and posterior brain systems in attentional selection. J Cogn Neurosci 2000; 12: 988–1000.

  59. 59

    MacLeod CM, MacDonald PA . Interdimensional interference in the Stroop effect: uncovering the cognitive and neural anatomy of attention. Trends Cogn Sci 2000; 4: 383–391.

  60. 60

    MacDonald 3rd AW, Cohen JD, Stenger VA, Carter CS . Dissociating the role of the dorsolateral prefrontal and anterior cingulate cortex in cognitive control. Science 2000; 288: 1835–1838.

  61. 61

    Derbyshire SW, Vogt BA, Jones AK . Pain and Stroop interference tasks activate separate processing modules in anterior cingulate cortex. Exp Brain Res 1998; 118: 52–60.

  62. 62

    Bush G, Whalen PJ, Rosen BR, Jenike MA, McInerney SC, Rauch SL . The counting Stroop: an interference task specialized for functional neuroimaging—validation study with functional MRI. Hum Brain Mapping 1998; 6: 270–282.

  63. 63

    Taylor SF, Kornblum S, Lauber EJ, Minoshima S, Koeppe RA . Isolation of specific interference processing in the Stroop task: PET activation studies. Neuroimage 1997; 6: 81–92.

  64. 64

    Carter CS, Mintun M, Cohen JD . Interference and facilitation effects during selective attention: an H215O PET study of Stroop task performance. Neuroimage 1995; 2: 264–272.

  65. 65

    Ruff CC, Woodward TS, Laurens KR, Liddle PF . The role of the anterior cingulate cortex in conflict processing: evidence from reverse stroop interference. Neuroimage 2001; 14: 1150–1158.

  66. 66

    Leung HC, Skudlarski P, Gatenby JC, Peterson BS, Gore JC . An event-related functional MRI study of the stroop color word interference task. Cereb Cortex 2000; 10: 552–560.

  67. 67

    Botvinick M, Nystrom LE, Fissell K, Carter CS, Cohen JD . Conflict monitoring versus selection-for-action in anterior cingulate cortex. Nature 1999; 402: 179–181.

  68. 68

    Casey BJ, Thomas KM, Welsh TF, Badgaiyan RD, Eccard CH, Jennings JR et al. Dissociation of response conflict, attentional selection, and expectancy with functional magnetic resonance imaging. Proc Natl Acad Sci USA 2000; 97: 8728–8733.

  69. 69

    van Veen V, Cohen JD, Botvinick MM, Stenger VA, Carter CS . Anterior cingulate cortex, conflict monitoring, and levels of processing. Neuroimage 2001; 14: 1302–1308.

  70. 70

    Grachev ID, Kumar R, Ramachandran TS, Szeverenyi NM . Cognitive interference is associated with neuronal marker N-acetyl aspartate in the anterior cingulate cortex: an in vivo (1)H-MRS study of the Stroop Color–Word task. Mol Psychiatry 2001; 6: 496, 529–539.

  71. 71

    Oldfield RC . The assessment and analysis of handedness: the Edinburgh Inventory. Neuropsycholgia 1971; 9: 97–113.

  72. 72

    Sperling R, Greve D, Dale A, Killiany R, Holmes J, Rosas HD et al. Functional MRI detection of pharmacologically induced memory impairment. Proc Natl Acad Sci USA 2002; 99: 455–460.

  73. 73

    Talairach J, Tournoux P . Co-planar Stereotaxic Atlas of the Human Brain. Thieme Medical Publishers: Stuttgart, 1988.

  74. 74

    Crosson B, Sadek JR, Bobholz JA, Gokcay D, Mohr CM, Leonard CM et al. Activity in the paracingulate and cingulate sulci during word generation: an fMRI study of functional anatomy. Cereb Cortex 1999; 9: 307–316.

  75. 75

    Raichle ME, Fiez JA, Videen TO, MacLeod AM, Pardo JV, Fox PT et al. Practice-related changes in human brain functional anatomy during nonmotor learning. Cereb Cortex 1994; 4: 8–26.

  76. 76

    Bench CJ, Frith CD, Grasby PM, Friston KJ, Paulesu E, Frackowiak RS et al. Investigations of the functional anatomy of attention using the Stroop test. Neuropsychologia 1993; 31: 907–922.

  77. 77

    Zysset S, Muller K, Lohmann G, von Cramon DY . Color–word matching stroop task: separating interference and response conflict. Neuroimage 2001; 13: 29–36.

  78. 78

    Hazeltine E, Poldrack R, Gabrieli JD . Neural activation during response competition. J Cogn Neurosci 2000; 12(Suppl 2): 118–129.

  79. 79

    Paus T, Koski L, Caramanos Z, Westbury C . Regional differences in the effects of task difficulty and motor output on blood flow response in the human anterior cingulate cortex: a review of 107 PET activation studies. Neuroreport 1998; 9: R37–R47.

  80. 80

    Petersen SE, Fox PT, Posner MI, Mintun M, Raichle ME . Positron emission tomographic studies of the cortical anatomy of single-word processing. Nature 1988; 331: 585–589.

  81. 81

    Frith CD, Friston K, Liddle PF, Frackowiak RS . Willed action and the prefrontal cortex in man: a study with PET. Proc R Soc London B Biol Sci 1991; 244: 241–246.

  82. 82

    Mesulam MM . Large-scale neurocognitive networks and distributed processing for attention, language and memory. Ann Neurol 1990; 28: 597–613.

  83. 83

    Gehring WJ, Willoughby AR . The medial frontal cortex and the rapid processing of monetary gains and losses. Science 2002; 295: 2279–2282.

  84. 84

    Gehring WJ, Knight RT . Prefrontal-cingulate interactions in action monitoring. Nat Neurosci 2000; 3: 516–520.

  85. 85

    Petit L, Courtney SM, Ungerleider LG, Haxby JV . Sustained activity in the medial wall during working memory delays. J Neurosci 1998; 18: 9429–9437.

  86. 86

    Vogt BA, Nimchinsky EA, Vogt LJ, Hof PR . Human cingulate cortex: surface features, flat maps, and cytoarchitecture. J Comp Neurol 1995; 359: 490–506.

  87. 87

    Goldman-Rakic PS . Topography of cognition: parallel distributed networks in primate association cortex. Annu Rev Neurosci 1988; 11: 137–156.

  88. 88

    Posner MI, Petersen SE . The attention system of the human brain. Annu Rev Neurosci 1990; 13: 25–42.

  89. 89

    Banich MT, Milham MP, Atchley RA, Cohen NJ, Webb A, Wszalek T et al. Prefrontal regions play a predominant role in imposing an attentional ‘set’: evidence from fMRI. Brain Res Cogn Brain Res 2000; 10: 1–9.

  90. 90

    Badgaiyan RD . Executive control, willed actions, and nonconscious processing. Hum Brain Mapping 2000; 9: 38–41.

  91. 91

    Koski L, Paus T . Functional connectivity of the anterior cingulate cortex within the human frontal lobe: a brain-mapping meta-analysis. Exp Brain Res 2000; 133: 55–65.

  92. 92

    Schubotz RI, von Cramon DY . Functional organization of the lateral premotor cortex: fMRI reveals different regions activated by anticipation of object properties, location and speed. Brain Res Cogn Brain Res 2001; 11: 97–112.

  93. 93

    Toni I, Schluter ND, Josephs O, Friston K, Passingham RE . Signal, set- and movement-related activity in the human brain: an event-related fMRI study. Cereb Cortex 1999; 9: 35–49.

  94. 94

    Corbetta M, Kincade JM, Ollinger JM, McAvoy MP, Shulman GL . Voluntary orienting is dissociated from target detection in human posterior parietal cortex. Nat Neurosci 2000; 3: 292–297.

  95. 95

    Rushworth MF, Paus T, Sipila PK . Attention systems and the organization of the human parietal cortex. J Neurosci 2001; 21: 5262–5271.

  96. 96

    George MS, Ketter TA, Parekh PI, Rosinsky N, Ring H, Casey BJ et al. Regional brain activity when selecting a response despite interference: an H215O PET study of the Stroop and an emotional Stroop. Human Brain Mapping 1994; 1: 194–209.

  97. 97

    Paus T, Castro-Alamancos MA, Petrides M . Cortico-cortical connectivity of the human mid-dorsolateral frontal cortex and its modulation by repetitive transcranial magnetic stimulation. Eur J Neurosci 2001; 14: 1405–1411.

  98. 98

    Nahas Z, Teneback CC, Kozel A, Speer AM, DeBrux C, Molloy M et al. Brain effects of TMS delivered over prefrontal cortex in depressed adults: role of stimulation frequency and coil-cortex distance. J Neuropsychiatry Clin Neurosci 2001; 13: 459–470.

Download references


The authors wish to thank Mary T Foley, Larry White, Larry Wald, Meghan Searl, Kerry Carley-Bush, Jordan Alexander, Taryn Brandon, and our subjects for their patience and assistance. Support for this work was provided by NIMH (Scientist Development Award 01611), NARSAD and the Forrest C Lattner Foundation.

Author information

Correspondence to G Bush.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Bush, G., Shin, L., Holmes, J. et al. The Multi-Source Interference Task: validation study with fMRI in individual subjects. Mol Psychiatry 8, 60–70 (2003).

Download citation


  • cingulate
  • Stroop
  • Eriksen
  • Simon
  • cognition
  • interference
  • oddball
  • attention
  • clinical
  • diagnostic

Further reading