Abstract

Studies of a variety of higher cognitive functions consistently activate a region of anterior cingulate cortex (ACC), situated posterior to the genu and superior to the corpus callosum. However, it is not clear whether the same ACC region is activated for all response modalities (e.g. vocal and manual) and/or all processing domains (e.g. verbal and spatial). To explore this question, we used rapid event-related functional magnetic resonance imaging and a spatial Stroop task with conditions tapping both verbal and spatial processing. We also employed novel methods that allowed us to acquire the accuracy and reaction times of both manual and vocal responses. We found one large ACC region that demonstrated significant response conflict effects with both vocal and manual responses, and three ACC regions that demonstrated significant response conflict effects with both spatial and verbal processing. We did not find any ACC regions that demonstrated activity selective to either a specific response modality or processing domain. Thus, our results suggest that the same regions of ACC are responsive to conflict arising with both manual and vocal output and with both spatial and verbal processing.

Introduction

The role of the anterior cingulate cortex (ACC) in higher cognition has received a great deal of attention in recent years. In part, this is because studies of a wide range of higher cognitive functions, including working memory, verbal fluency, selective attention, and long-term memory have consistently found activation of the ACC. More specifically, the region typically activated in these studies is situated posterior to the genu of the ACC, and superior to the corpus callosum. In the nomenclature of Picard and Strick, this ACC region is located in the rostral cingulate zone (rCZ) (Picard and Strick, 1996). In recent work, we have proposed a hypothesis as to why this region of the ACC is activated in a wide range of cognitive tasks. Specifically, we have hypothesized that this ACC region is active under a range of task conditions because it evaluates the demand for cognitive control by monitoring for the occurrence of response conflict in information processing. A number of studies now provide support for this hypothesis (Carter et al., 1998; Botvinick et al., 1999, 2001; Barch et al., 2000; Carter et al., 2000; Braver et al., 2001). However, an additional question is whether this ACC region is functionally homogeneous, or whether it can be further functionally segregated along some domain, such as the nature of the response to be made or the nature of the processing domain. In particular, the goals of this study were to determine whether or not the same ACC region responds to conflict in across response modalities (e.g. both vocal and manual) and/or across processing domains (e.g. verbal and spatial).

As noted above, our hypothesis about ACC function is that it serves to evaluate the demand for cognitive control by monitoring for response conflict or crosstalk in information processing. By crosstalk we mean interference or interactions in the processing of two stimuli or responses that occurs when the pathways for this processing overlap. Thus, our hypothesis predicts that the ACC should be active whenever there is a high degree of competition between incompatible responses. A number of studies now provide support for this hypothesis, demonstrating activation of a similar ACC region in a variety of task domains that all elicit response conflict. These task domains can be loosely grouped into three categories. First, this ACC region is active in task conditions in which a prepotent response tendency has to be overcome, such as in studies of the Stroop task and the Go/No-go task (Pardo et al., 1990; Bench et al., 1993; Paus et al., 1993; Taylor et al., 1994, 1997; Carter et al., 1995, 2000; Kawashima et al., 1996; Casey et al., 1997; George et al., 1997; Bush et al., 1998; Derbyshire et al., 1998; Botvinick et al., 1999; Brown et al., 1999; Konishi et al., 1999; Peterson et al., 1999; Braver et al., 2001). Second, this ACC region is active in task conditions when the response to be made is not fully constrained by the task context, such as in studies of verb generation, verbal fluency, and stem completion (Petersen et al., 1989; Frith et al., 1991, 1993; Friston et al., 1993; Raichle et al., 1994; Buckner et al., 1995; Yetkin et al., 1995; Warburton et al., 1996; Phelps et al., 1997; Thompson-Schill et al., 1997; Crosson et al., 1999). Lastly, ACC activity is also commonly found when individuals produce errors, a situation that typically either elicits, or is the result of, response conflict (Hohnsbein et al., 1989; Gehring et al., 1990; Dahaene et al., 1994; Carter et al., 1998; Kiehl et al., 2000).

Although a growing body of evidence supports the hypothesis that the ACC is responsive to conflict in information processing, a number of questions about the role of the ACC in cognition remain. In particular, one important issue is whether the same ACC region responds to conflict in all response modalities. There are a number of reasons to raise this question. First, in prior work, Paus and colleagues found that response conflict tasks using oculomotor, manual and speech responses activated slightly different ACC regions (Paus et al., 1993). Paus found that oculomotor and speech responses activated regions in the rostral ACC, speech responses activated regions in the intermediate ACC, and manual response activated regions in the caudal ACC. These researchers argued that this pattern was consistent with data on the somatotopic organization of cingulate cortex in monkeys, and interpreted their results as consistent with the hypothesis that the ACC participates in facilitating appropriate motor responses, and inhibiting inappropriate ones. However, since only peak activations were reported, the extent of overlap in ACC regions activated by different response modalities was not clear.

A second reason to ask whether the location of activity within the rCZ is somatotopically mapped is a more recent review article by Picard and Strick (Picard and Strick, 1996). In this work, Picard and Strick reviewed positron emission tomography (PET) studies demonstrating activation changes in the medial wall of the human cortex. Picard and Strick divided these studies into ones that used either simple or complex motor tasks, defining a complex motor task as one ‘characterized by additional motor or cognitive demands such as the selection of a motor response, and the acquisition of a conditioned association’. Picard and Strick concluded that there was evidence for some somatotopic organization of ACC activations associated with particular output modalities for both simple and complex tasks, as well as differences in the anatomic location of activations associated with simple versus complex motor tasks. In addition, Paus et al. provided another recent review of the ACC literature, concluding that there was evidence for hand–arm responses activating a more caudal cingulate region, and non hand–arm responses activating a more rostral cingulate region (Paus et al., 1998). However, the somatotopic mappings suggested by Picard and Strick versus Paus are slightly different, with Picard and Strick suggesting additional arm regions within the zone Paus suggested contained primarily speech or non arm–hand representations. Further, for the majority of the studies in both reviews, activation peaks for different response modalities were obtained from different studies. Thus, it is difficult to know what contribution differences in task design, imaging, and anatomical localization methods may have made to differences in the location of activations.

A third reason to ask whether the location of activity in the rCZ is somatotopically mapped is the results of a recent study by Turken and Swick (Turken and Swick, 1999). These authors report work with a focal right ACC lesion patient. The paradigms used to assess the patient included both tasks that should elicit response conflict (e.g. the Stroop, a divided attention task) and more simple response preparation tasks. Across all three tasks, both simple and complex, the patient exhibited a selective impairment in responding when manual, but not vocal responses were required. On the surface, the results of this study are consistent with the hypothesis that the ACC may exhibit somatotopic mapping. However, the precise location of the lesion in this patient may actually not correspond to the region of ACC typically activated by response conflict. Specifically, the lesion was primarily located in BA 24, and included the dorsal and ventral banks of the cingulate sulcus, but did not extend into the paracingulate sulcus. In addition, the lesion did not extend into the supplementary motor area (SMA), but did extend down to the corpus callosum. Thus, the majority of the lesion in this patient may actually lie below the rostral cingulate zone identified by Picard and Strick, which was located mainly in BA 32, with some extension into dorsal BA 24. As noted above, the rostral cingulate zone is the area we believe is most strongly associated with the monitoring of conflict. Further, the lesion in this patient clearly extended into the caudal cingulate zone identified by Picard and Strick, which they associated more with the production of simple as compared to complex manual responses.

As described above, several lines of evidence suggest that there may be somatotopic mapping of ACC activations associated with response conflict. However, an update of Picard and Strick's review provides a somewhat different picture regarding somatotopic organization within the rostral cingulate zone. Specifically, in Tables 1–3, we list the locations of ACC activations from the tasks that Picard and Strick categorized as complex, as well additional studies we could find that fell into any of the categories of tasks outlined above (inhibition of prepotent response, underdetermined responding, commission of errors), and which provided standardized stereotactic coordinates in the reference frame of Talairach and Tournoux for their ACC activations. Figure 1 plots these activations on a mid-sagittal slice from the Talairach coordinate system.

As can be seen in Figure 1, there is a large area of rCZ in which activations associated with both manual and vocal responses are intermixed, with no clear visual evidence for somatotopic organization within this region of the rCZ. However, Picard and Strick suggested that there may be two subdivisions of the rCZ, an anterior one (rCZa) and a posterior one (rCZp), each of which may have separate regions for face and arm representations. In addition, Picard and Strick suggested that the caudal cingulate zone (just in front of and behind the Vca line) primarily contains arm representations. Thus, similar to Picard and Strick, we separated the reviewed activations into three subregions: (i) caudal cingulate zone (cCZ: posterior to Y = +6 and +29 mm < Z < +46 mm); (ii) posterior rostral cingulate zone (rCZp: +5 mm < Y < +26 mm, +26 mm < Z < +51 mm); and (iii) anterior rostral cingulate zone (rCZa: +24 mm < Y < +26 mm; +9 mm < Z < +41 mm). Seven activations fell within the cCZ, of which five were associated with manual and two with vocal/eye responses, providing some evidence consistent with the hypothesis that the cCZ contains arm representations. In both the rCZa and the rCZp, Picard and Strick suggested that the arm representations were caudal to the face representations. If this were true, then activations associated with manual, as compared with vocal or eye responses, should have more posterior Y coordinate values and/or more superior Z coordinate values. To evaluate the presence or absence of somatotopy in these rCZ subdivisions quantitatively, we compared the Y and Z coordinates for manual versus vocal/eye responses using t-tests. For the rCZp, these analyses did not indicate any significant differences in either the Y (manual M = 16.3; vocal/eye M = 17.1) or Z (manual M = 38.7; vocal/eye M = 36.1) coordinates (P < 0.10). Similarly, for the rCZa region, these analyses again did not indicate any significant differences in either the Y (manual M = 34; vocal/eye M = 31.1) or Z (manual M = 24.3; vocal/eye M = 21.8) coordinates (P < 0.10). Thus, with the inclusion of more recent studies assessing response conflict, no clear evidence emerges for somatotopy in either the rCZa or the rCZp.

A second issue regarding activity in the rCZ is whether the same ACC region responds to conflict in all processing domains (e.g. verbal and spatial). Other regions of cortex demonstrate evidence for lateralized activation in response to differences in processing domain. For example, inferior frontal cortex (BA 44/6) shows relatively greater right sided activation for non- verbal processing, but relatively greater left-sided activation for verbal processing (D'Esposito et al., 1998; Kelley et al., 1998). Thus, one might hypothesize that the ACC might demonstrate a similar lateralization of function, with the right ACC being more responsive to conflict arising with non-verbal processing, while left ACC might be more responsive to conflict arising from verbal processing. Unfortunately, there is little research to date relevant to addressing this hypothesis. The vast majority of the studies that have examined inhibition of a prepotent response, underdetermined responding, or errors have used verbal processing. In the imaging domain, the exception to this is studies that have examined antisaccades, which can be conceived of as non-verbal processing. However, these antisaccade studies have primarily used eye movements as the response modality. Thus, it is impossible to determine whether the location of ACC activity in such studies reflects the nature of the processing or the modality of the response. The Turken and Swick study of the ACC lesion study did include conditions in their Stroop task that could be conceived of as both verbal (e.g. attend to the word) and non-verbal (e.g. attend to direction of the arrow). However, their patient showed impairment in both the word and arrow conditions when manual responses were required, although the pattern of impairment was slightly different in the two conditions.

As described above, our review of the literature suggests that, while there is reason to question whether the location of activity in the rCZ is somatotopically mapped, the research to date does not provide a clear answer to this question. In addition, there is almost no evidence to date regarding whether the location of ACC activity might vary according to the nature of the processing domain. Thus, the goal of the current study was to test directly these two hypotheses in healthy individuals using event-related functional magnetic resonance imaging (fMRI). Specifically, we used a version of the spatial Stroop task that allowed us to fully cross processing domain (e.g. verbal versus non-verbal) with response modality (e.g. vocal versus manual). In addition, we used fMRI methods that allowed us to acquire both the content and the latency of participants oral responses in vocal response conditions.

Materials and Methods

Participants

Thirteen neurologically normal right-handed subjects participated in this study. Subjects were seven males and six females, with a mean age of 23 (range 19–33 years). Subjects were paid $25 an hour for participation, and gave informed consent in accordance with guidelines set by the Human Studies Committee at Washington University. Because of technical problems, behavioral data from one subject and imaging data from a separate subject were unusable, and thus these subjects were excluded from analysis.

Behavioral Procedures and Cognitive Tasks

A Power Macintosh computer (Apple, Cupertino, CA) and PsyScope software (Cohen et al., 1993) displayed all visual stimuli. A LCD projector (Sharp, model XGE850) projected stimuli onto a screen placed at the head of the bore. Subjects viewed the screen via a mirror fastened to the head coil. Subjects responded either by pushing a fiber-optic light-sensitive key- press connected to a PsyScope Button Box (Carnegie Mellon University, Pittsburgh, PA) that recorded both accuracy and reaction time, or by making an overt vocal response (described in more detail below). Stimuli appeared for 1000 ms, followed by a 2200 ms inter-trial interval.

We used a factorial design, fully crossing two spatial Stroop tasks (attend to location, attend to word) with two response types (manual, vocal) and four trial types (fixation, congruent, neutral, incongruent). In all task conditions, subjects were presented with a word either to the right or left of a central fixation point. In the attend to location task (referred to as ‘location’), subjects were told to respond to the location of the word, and ignore its content. In the attend to word task (referred to as ‘word’), subjects were told to respond to the content of the word and ignore its location. For fixation trials for both tasks, subjects simply saw a centrally presented fixation cross. For congruent trials in both tasks, subjects saw either the word ‘right’ presented to the right of fixation, or the word ‘left’ presented to the left of fixation. For incongruent trials in both tasks, subjects saw either the word ‘right’ presented to the left of fixation or the word ‘left’ presented to the right of fixation. In neutral trials for the location task, subjects saw the words ‘home’ and ‘great’ presented either to the left or right of fixation. In neutral trials for the word task, subjects saw the words ‘right’ and ‘left’ presented in the center of the computer screen. Participants were asked to make a manual response on half of the trials, pressing either the right or left button of the fiber-optic button box. Participants were asked to make a vocal response on the remaining trials, saying either ‘right’ or ‘left’ aloud. Trials were blocked by task and response modality, such that each participant performed two runs of each of the following (with order counterbalanced across participants): (i) location, manual response; (ii) location, vocal response; (iii) word, manual response; and (iv) word, vocal response. During each run, the four trial types (fixation, congruent, neutral, incongruent) were presented equally often in a continuous series of 112 intermixed trials. To allow rapid event-related analyses (Dale and Buckner, 1997), trial types were pseudorandomly intermixed with first-order counterbalancing (each trial type followed ever other trial type equally often). Four different such pseudorandomly intermixed orders were created, and used twice for each subject (once for a word run and once for a location run). List order was counterbalanced across subjects. To create a stable task baseline, each functional run began with 16 s of visual fixation and ended with an additional 35.2 s of visual fixation.

Acquisition of Vocal Responses

Participants' overt vocal responses were acquired through the use of an elastic tube and a condenser microphone. The elastic tube was positioned over the participant's mouth and taped to the head coil. The elastic tube ran the length of the participant's body to the door of the scanner room. The condenser microphone was taped to the inside of the elastic tube at the end near the scanner door and run under the scanner door. The signal from the microphone was then split, going both into a standard tape- recorder (to record the content of the participant's response) and into the PsyScope button box attached to the Macintosh computer running PsyScope software. The connection into the button box allowed a voice- activated response key to record the reaction time of the participant's vocal response. Normally, the noise generated by fast changing echoplanar gradients preclude accurate acquisition of overt vocal responses. To avoid this, a quiet interval of 800 ms was interleaved with each frame acquisition (TR = 2400 ms + 800 ms Quiet = 3200 ms trial) (Akbudak et al., 1999). Although such a quiet period was not needed for manual responses, the identical procedure was used for both manual and vocal response runs to enhance comparability across different response modalities. The accuracy of participant's performance in the vocal response condition was coded by listening to the tape-recordings made during the scanning session.

Scanning Procedures

Images were acquired on a Siemens 1.5 Tesla Vision System (Erlangen, Germany) with a standard circularly polarized head coil. A pillow and tape were used to minimize head movement. Headphones dampened scanner noise and enabled communication with participants. Structural images were acquired using a high resolution (1.2511 mm) sagittal 3-D MP-RAGE (Mugler and Brookeman, 1990) T1-weighted sequence (TR = 9.7 ms, TE = 4 ms, flip = 12, TI = 300 ms). Functional images were acquired using an asymmetric spin-echo echo-planar sequence (TR = 2400 ms, Quiet Period = 800 ms, TE = 50 ms, flip = 90). During each functional scanning run 128 sets of 16 contiguous, 8 mm thick axial images were acquired parallel to the anterior–posterior commissure plane (3.75 × 3.75 mm in-plane resolution), allowing complete brain coverage at a high signal-to-noise ratio (Conturo et al., 1996). Each run lasted ~7 min, with a 2 min rest period between runs.

Movement Estimation and Correction

Functional images were corrected for movement using a six-parameter rigid-body rotation and translation correction (Friston et al., 1994; Snyder, 1996). Two sets of estimated movement parameters (Pitch, Roll, Yaw, X, Y, Z) were obtained from this algorithm. The first set was the difference of the current image from the immediately preceding image, which will be referred to as incremental movement. The second set was the difference of the current image from the reference image (the first image acquired), which will be referred to as absolute movement. For Pitch, Roll and Yaw, the parameters are expressed in degrees. For X, Y and Z the parameters are expressed in millimeters. The absolute values of these estimates were used to examine the degree of increased movement associated with producing overt vocal responses. We analyzed the movement data using two-factor ANOVAs, with image as the random factor (all images acquired were analyzed), task (location, word) and response modality (manual, vocal) as within-subject factors and the six movement parameters (Pitch, Roll, Yaw, X, Y, Z) as dependent variables. For absolute movement, there were no significant main effects of either task or response modality. This result is consistent with our prior research demonstrating no increase in absolute movement during overt vocal responding (Barch et al., 1999, 2000). For incremental movement, the ANOVAs demonstrated no significant main effects of task, but significantly main effects of response modality for all six parameters (all Ps < 0.05). For all parameters, incremental movement was greater during vocal than manual responding. However, the magnitude of incremental movement during vocal responding was still relatively small, again consistent with the results of our prior research on overt vocal responding (Barch et al., 1999, 2000).

Susceptibility Artifacts

To assess potential reductions in signal-to-noise (SNR) associated with overt vocal responses, we quantified SNR by calculating the mean SNR (mean/variance) for each participant, for each slice location, separately for vocal and manual response runs. Paired-sample t-tests indicated significantly reduced SNR during vocal compared to manual responses for 12 out of the 16 slices (P < 0.05). To determine the magnitude of the SNR reduction, we calculated the average percent decrease in SNR {[(Manual SNR – Vocal SNR)/Manual SNR)] × 100} across participants for each slice. The average decrease in SNR ranged from 0 to 22% across slices (M = 11%), and was relatively small (e.g. < 10%) in the more superior slices. However, the reduction in SNR did increase in the more inferior slices, which is not surprising given that these slices are closer to the throat and mouth region.

Image Analysis Procedures

Functional imaging data were analyzed according to the following procedures. Following movement correction, all functional images were scaled to achieve a whole-brain mode value (used in place of mean because of its reduced sensitivity to variation in brain margin definition) of 1000 for each scanning run (to reduce the effect of scanner drift or instability). Functional images were then resampled into 3 mm isotropic voxels, transformed into standardized atlas space (Talairach and Tournoux, 1988), and smoothed with a Gaussian filter (6 mm FWHM). A General Linear Model (Friston et al., 1994; Worsley and Friston, 1995; Josephs et al., 1997; Zarahn et al., 1997; Miezin et al., 2000) was used to analyze the pre-processed data on a voxel-by-voxel basis. Estimates of the magnitudes of each effect were derived from the model. Specifically, the seven time-points following the stimulus were cross-correlated with a series of five lagged hemodynamic response functions, each separated by 1 s, in order to account for possible variation in the onset of the hemodynamic response function (Boynton et al., 1996; Dale and Buckner, 1997; Buckner et al., 1998). These magnitude estimates were then analyzed using appropriately designed ANOVAs and t-tests (as described in more detail below), treating subject as a random effect. Statistical parametric maps of the voxel-wise t- and F-values were thresholded for significance using a cluster-size algorithm (Forman et al., 1995). This algorithm takes account of the spatial extent of activation to correct for multiple comparisons. The specific thresholds used for each analysis are described below. In graphic displays, all effects are described in terms of percent signal change. Percent signal change was defined as signal magnitude divided by the mean signal intensity across all functional runs of the intercept term of the linear model.

Results

Behavioral Data Analyses

The behavioral data acquired during the scanning session, both errors and reaction times for correct responses, were analyzed using three-way ANOVAs with task (location, word), response modality (manual, vocal) and trial type (congruent, neutral and incongruent) as within-subject factors. The error rates for this task were very low (Fig. 2), although the ANOVA indicated a main effect of trial type [F(2,20) = 14.46, P < 0.01], and a two-way interaction between task and response modality [F(1,10) = 10.06, P < 0.01], which was further moderated by a three-way interaction between task, response modality, and trial type [F(2,20) = 9.02, P < 0.01]. The main effect of trial type reflected the fact that there were more errors in the incongruent condition than either the neutral or congruent conditions, which did not differ. The three-way interaction resulted from there being a bigger increase in errors from the neutral to incongruent condition for manual as compared to vocal responses in the word task, but a bigger increase in errors in the incongruent condition for vocal as compared to manual responses in the location task. For the RT (reaction time) ANOVA, all main effects and inter- actions were significant, including the three-way interaction between task, response modality and trial type [F(2,20) = 23.56, P < 0.01]. Planned contrasts indicated that the main effect of trial type was significant for both vocal [F(2,20) = 16.08, P < 0.01] and manual responses [F(2,20) = 10.75, P < 0.01] in the location task and both vocal [F(2,20) = 6.29, P < 0.01] and manual [F(2,20) = 74.23, P < 0.01] responses in the word task. However, as can be seen in Figure 2, the three-way interaction between task, response modality and trial type resulted from there being a greater slowing of RT between the neutral and incongruent condition for vocal as compared to manual responses in the location task, but a greater slowing of RT between the neutral and incongruent trial types for manual than vocal responses in the word task.

The above analyses of the behavioral data utilized RT measures for vocal responses acquired during the course of fMRI scanning. To determine whether these RT data were valid, we ran an additional 12 subjects in the identical tasks outside of the scanner. The results for these 12 subjects were essentially identical to those with the data acquired in the scanner, including three-way interactions between task, response modality and trial type for both RT [F(2,22) = 9.10, P < 0.001] and errors [F(2,22) = 7.39, P < 0.01]. To quantitatively assess the similarities of the RT data acquired in and outside of the fMRI scanner, we conducted an ANOVA on the RTs, with session (in scanner, out of scanner) as a between-subject factor. There was a main effect of session [F(1,21) = 22.08, P < 0.01] with overall faster RTs outside of the scanner. In addition, there was a session  response modality interaction [F(1,21) = 24.91, P < 0.01]. This interaction reflected the fact that although overall RTs were slower for both manual and vocal response in the scanner as compared to out of the scanner, the slowdown was greater for vocal than manual responses. There were no other interactions between session and any of the other factors.

Common ACC Regions Responsive to Conflict for Vocal and Manual Responses

We began the analysis of the fMRI data by looking for ACC regions that displayed significant effects of response competition with both vocal and manual responses. Our hypotheses were specifically focused on the ACC, thus we focused on voxels falling within either the rCZ or cCZ as defined by Picard and Strick (1996). However, we also examined voxels falling within pre-SMA and SMA as comparison regions, again as defined by Picard and Strick. More specifically, we defined a voxel as being within the rCZ if its Tailarach coordinates placed it within BA 32 superior to Z = +10 mm and anterior to Y = +5 mm, or within BA 24 superior to Z = +10 mm and anterior to Y = +15 mm. We defined a voxel as falling within the cCZ if its Tailarach coordinates placed it within BA 24 with a Y coordinate between +5 and –15 mm and a Z coordinate between +30 and +45 m. We defined a voxel as falling within pre-SMA if its Tailarach coordinates placed it within BA 6 anterior to Y = 0 mm, and as falling within SMA if its Tailarach coordinates placed it within BA 6 posterior to Y = 0 mm.

To examine our hypotheses, we performed ‘conjunction’ analyses similar to those described by Friston and colleagues (Price and Friston, 1997; Friston et al., 1999). To do so, we conducted paired t-tests on all voxels (separately for vocal and manual responses) and examined only those voxels within ACC, SMA, and pre-SMA that demonstrated a significantly greater magnitude of response in the incongruent condition as compared to the neutral condition independently for both vocal and manual responses. Because of the conjunctive nature of this analyses, we set the P-value threshold for the analysis of each individual response modality at P < 0.05 and four voxels. However, requiring a conjunction of significance in both the vocal and manual response condition actually leads to significance threshold of P < 0.0025 (0.05×0.05). As shown in Table 4, this analysis identified a region of ACC that demonstrated significant response competition effects for both manual and vocal responses, located in the rCZ. Figure 3 displays the time courses for this ACC region. For both vocal and manual responses, ACC activity was greater during incongruent trials than during neutral trials.

One criticism that might be leveled against the above analysis is the possibility that we found a common ACC region responsive to conflict for both manual and vocal responses because covert articulation may have been engaged in at least one of the manual response conditions (e.g. attend to word, manual response). One might make a similar argument for participants having engaged in ‘covert’ arm movements in at least one of the vocal response conditions (e.g. attend to location). To address this possibility, we conducted a second analysis examining only the manual response condition we thought least likely to engage any covert articulation (i.e. attend to location, manual response) and the vocal response condition we thought least likely to engage any covert arm movements (i.e. attend to word, vocal response). We used conjunction analyses identical to those described above to identify ACC and SMA voxels that demonstrated significant response conflict effects in both the attend to location–manual response condition and the attend to word–vocal response conditions. In general, this analysis produced results similar to those described above. Specifically, as shown in Table 4, we found two regions of ACC and a region of SMA that demonstrated significant response conflict effects for both the location–manual and word–vocal tasks. One of the ACC regions was in the posterior rCZ, and one was in the anterior rCZ.

Different ACC Regions Responsive to Conflict for Vocal Versus Manual Responses

To identify ACC regions demonstrating response competition effects selectively for only one of the two response modalities (e.g. manual or vocal), we conducted two-factor ANOVAs on each voxel, with trial type (neutral, incongruent) and response modality (manual, vocal) as within subject factors. We then examined those voxels demonstrating significant interactions between trial type and response modality. We used a more liberal threshold for this analysis (P < 0.01 and four voxels), to protect against null results due to an overly conservative threshold. This analysis did not identify any regions in ACC that demonstrated conflict effects in one response modality and not the other. However, this lack of an effect in ACC did not reflect an inability to identify any region showing such an interaction. Although these areas were not the focus of this study, we see regions in the cerebellum, inferior frontal cortex, parietal cortex, and SMA that demonstrated significant response conflict effects (e.g. greater event-related activity to incongruent than neutral trials) for manual, but not vocal responses. We also saw a region of middle frontal gyrus that demonstrated significant response conflict effects for vocal but not manual responses.

As noted above, a criticism that might be leveled against the above interaction analysis is the possibility that we failed to find ACC regions responsive in one modality and not the other because covert articulation may have been engaged in at least one of the manual response conditions (e.g. attend to word, manual response) or because ‘covert’ arm movements may have occurred in at least one of the vocal response conditions (e.g. attend to location). Thus, to again address this possibility, we conducted an additional analysis comparing the manual response condition we thought would be least likely to engage any covert articulation (i.e. attend to location, manual response) and the vocal response condition we thought would be least likely to engage any covert arm movements (i.e. attend to word, vocal response). Specifically, we conducted voxel-wise two-factor ANOVAs on each voxel, with trial type (neutral, incongruent) and condition (location–manual vs word–vocal) as within- subject factors. Even with this more stringent analysis, we did not find any ACC regions showing response conflict effects in one of the conditions and not the other. Again, however, although these regions were not the focus of this study, we did see areas in primary motor cortex, frontal insula, and middle frontal gyrus (Table 5) that demonstrated significant response conflict effects (i.e. greater event-related activity to incongruent than neutral trials) for manual but not vocal responses. In addition, we found regions of visual cortex, temporal cortex and inferior and middle frontal cortex that demonstrated significant response conflict effects for vocal but not manual responses.

Our results are somewhat in conflict with previous findings by Paus and colleagues. To determine whether such differences might be related to analysis strategies, we also analyzed the data in a manner similar to Paus et al. Specifically, we conducted separate subtractions (incongruent – neutral) for manual and vocal responses, and identified peaks of activation separately for each response modality. A voxel was identified as a peak if it occurred with a cluster of at least four significantly activated voxels and was located at least 12 mm from any other peak of activation. For this analysis, we used the same significance threshold as in our first ‘conjunction’ analyses (P < 0.05). Without the additional constraint of the conjunction, this is a liberal threshold. However, we choose to err on the side of identifying potential false positives with good power to detect differences in the peaks of ACC activity as a function of response modality, as to provide the fairest comparison to the study by Paus and colleagues. These analyses identified three peaks of ACC activity for response conflict effects with manual responses: (i) BA 24 (X = –10, Y = 3, Z = 36); (ii) BA 24 (X = 1, Y = 9, Z = 36); and (iii) BA 32 (X = 19, Y = 36, Z = 24). For vocal responses, we identified two peaks of activation: (i) BA 32 (X = 7, Y = 3, Z = 42) and (ii) BA 32 (X = 4, Y = 39, Z = 27). These peaks for manual versus vocal responses do not show a clear dissociation in terms of a rostral/ caudal dimension. Thus, even using a similar analyses approach, our findings still differ from Paus and colleagues.

Common ACC Regions Responsive to Conflict in Both the Location or Word Task

We next investigated ACC regions that displayed significant effects of response competition for both the word and location tasks. To do so, we again performed ‘conjunction’ analyses. We again conducted paired t-tests on all voxels (separately for the location and word tasks) and examined only those voxels in ACC, SMA or pre-SMA that demonstrated a significantly greater magnitude of response in the incongruent condition as compared to the neutral condition for both the word and location tasks. As shown in Table 4, this analysis identified three regions of ACC, a region of SMA and a region of pre-SMA that demonstrated significant response competition effects for both the word and location tasks. Not surprisingly, two of these ACC regions were very similar in location to the one identified in the conjunction analysis for vocal and manual responses. The other ACC region was more rostral, falling in the anterior portion of the rCZ. As can be seen in Figure 4, the time courses of the cortical response for these ACC regions indicated that for both the word and location tasks, ACC activity was greater during incongruent trials than during neutral trials.

Different ACC Regions Responsive to Conflict in the Location Versus Word Task

To identify ACC regions demonstrating response competition effects for only one of the two tasks (e.g. word or location), we conducted two-factor ANOVAs on each voxel within ACC and SMA, with trial type (neutral, incongruent) and task (attend to word, attend to location) as within-subject factors. We then examined those voxels demonstrating significant interactions between trial type and task (P < 0.01). We did not find any regions in ACC that demonstrated task-selective conflict effects. Again, however, this lack of an effect in ACC did not reflect an inability to identify any region showing such an interaction. Although not the focus of this study, we did see a number of regions that demonstrated significant response conflict effects for the word task, but not the location task, including regions in visual cortex, cerebellum and parietal cortex (Table 5).

Somatotopy in Primary Motor and Somatosensory Cortex

The current study did not find any evidence for somatotopic mapping of conflict associated activation in ACC, although we did find some evidence for regions in SMA responding selectively to conflict with manual responses. However, to be in a better position to interpret the absence of clear somatotopy in ACC activations, we assessed our ability to identify somatotopic organization of activation in primary motor and somatosensory cortex, as well as supplementary motor cortex. Both human and animal data indicate that somatotopy should be present in primary motor cortex and SMA. Thus, demonstration of such results in the current study would suggest that our methods and design provided the power to detect somatotopic mapping where it is known to exist. Thus, we used paired t-tests to identify regions more responsive to manual than verbal responses, collapsing across task and trial type (P < 0.01 and four voxels). We identified regions more responsive to vocal than manual responses in the same manner, collapsing across task and trial type. As shown in Figure 5, we were able to see clear somatotopic organization of activations in primary motor cortex. Specifically, activations associated with the right-handed button press were clearly left lateralized and superior to the activations associated with vocal responding, which were more bilateral. In addition, within SMA we identified a region associated with manual responding that was located posterior to an SMA region associated with vocal responding, a pattern that conforms well with prior studies on somatotopic mapping in SMA (Picard and Strick, 1996).

Discussion

The major results of this study were that we identified regions of the ACC, within the rCZ, that demonstrated significant response conflict effects across response modalities and across processing domains. More specifically, we found a region of the posterior rCZ that demonstrated significant response conflict effects (greater activity during incongruent than neutral trials) for both manual and vocal responses. In addition, we found three regions of the ACC, one in the anterior rCZ and two in the posterior rCZ, that demonstrated significant response conflict effects for both verbal and non-verbal processing. Further, we did not find any ACC regions that demonstrated either response modality- specific or processing domain-specific response conflict effects. In contrast, we were able to identify clear differences in the regions of activations associated with manual versus vocal responses in primary motor and somatosensory cortex, regions known to display such somatotopic mapping.

These results of the current study in regards to the rCZ conflict with the conclusions drawn by Picard and Strick in their review of the literature, which suggested that there were different regions of both the anterior (rCZa) and posterior (rCZp) rostral cingulate zones associated with manual versus vocal responses, and that such differences are of a magnitude capable of being detected in group averaged analyses. As described above, we did not find any evidence for such somatotopy within either rCZa or rCZp. As such, our current findings are consistent with the review of the literature we presented in the Introduction, which suggested that ACC activity as a function of response conflict for both vocal and manual responses occurred throughout the rCZ, without clear somatotopic mapping. We should note, however, that the results of our study do not preclude the possibility that a finer grained somatotopy might be present in rCZ activations, such as might be detected in single subjects analyses or with a higher resolution magnet. However, our results are not consistent with the hypothesis that rCZ activations in response to conflict are somatotopically organized on a scale such as that suggested by Picard and Strick, who reviewed studies using group averaged data.

The ACC activation we identified as responsive to conflict for both vocal and manual responses fell within what Picard and Strick would designate the rCZ. However, one might still wonder about the presence of somatotopy within the caudal cCZ. Picard and Strick and Paus have defined the cCZ somewhat differently. Paus has defined the border between the caudal and rostral ACC as Y = +10 mm, irrespective of the Z coordinate. In addition, Paus has suggested that activations associated with either simple or complex manual responses should segregate to the cCZ. The centroid for our ACC ROI demonstrating response conflict effects for both vocal and manual responses fell within the cCZ as defined by Paus. Thus, our results are not consistent with Paus' suggestion that activations associated with complex manual responses are localized to cCZ while activations associated with vocal and eye movement responses are located more rostrally. In contrast to Paus, Picard and Strick define the border between the rCZ and cCZ in terms of both the anterior/posterior (approximately Y = +5 mm) and inferior/superior (approximately Z = +45 mm) dimensions. Further, Picard and Strick suggested that the cCZ is primarily activated by simple manual responses. However, when we collapsed across conflict conditions (e.g. congruent, neutral and incongruent) and simply compared manual to verbal responses, we did not find selective activation of cCZ for manual responses, although we did find evidence for somatotopic mapping in primary motor cortex and SMA. Further, inspection of the activation maps computed separately for vocal and manual responses indicated activation of the cCZ in all task conditions for both vocal and manual responses. The activation of cCZ (as defined by Picard and Strick) for vocal responses is somewhat surprising, and suggests that further research is needed to determine whether this finding replicates and to clarify the functional role of cCZ activity in humans.

We also did not find any evidence for laterality differences in ACC activation as a function of processing domain (i.e. verbal versus non-verbal). As noted in the Introduction, some cortical regions, such as the inferior frontal cortex, have shown evidence of lateralization of activation as a function of processing domain (D'Esposito et al., 1998). However, other frontal regions, such as dorsolateral prefrontal cortex (DLPFC) show much less evidence for lateralization of activation as a function of processing domain (D'Esposito et al., 1998). Interestingly, the rostral ACC has known strong reciprocal connectivity with dorsolateral pre- frontal cortex (Devinsky et al., 1995). As such, our finding of common ACC regions activated by response conflict with both vocal and manual responses, and with both verbal and non-verbal processing may actually be consistent with our hypothesis regarding the role that ACC activation may play in cognitive control. Specifically, we have hypothesized that a conflict signal from the ACC may help recruit additional cognitive control functions that may be carried out by other brain regions or systems, such as the dorsolateral prefrontal cortex. Consistent with this view, many of the tasks that elicit ACC activation also elicit dorsolateral prefrontal cortex activation (Carter et al., 1995; Braver et al., 1997, 2001; Cohen et al., 1997; Corbetta et al., 1991). Thus, it may be that the ACC serves to help determine when regions such as the dorsolateral prefrontal cortex needs to come on line to provide needed biasing in favor of task-relevant processing. As noted above, the same regions of DLPFC appear to be active for processing in multiple different domain and with all response output modalities (D'Esposito et al., 1998). As such, we may not see evidence for somatotopic mapping of ACC regions responsive to conflict if their function is to engage control mechanisms supported by other cortical regions that are involved in multiple processing domains and with multiple output modalities.

The current study and the review of the literature in the Introduction was specifically designed to examine the relationship between response conflict and ACC activation. However, the review presented in the Introduction did not include studies examining a type of task that often elicits ACC activation, namely working memory tasks. It is possible that the ACC activations during such working memory studies reflects the presence of response conflict in such paradigms. However, since these working memory studies did not explicitly include conditions that varied the degree of response conflict, we did not feel that it was fair to include them in a review of ACC activations associated with response conflict. Further, the vast majority of working memory studies have used manual responses, making it difficult to compare the location of ACC activations as a function of response modality. Nonetheless, as reviewed by Petit et al. (Petit et al., 1998), many of the ACC activations in working memory studies fall very close to the ACC regions identified in the current study. More interestingly, Petit found that activity in rCZ during both spatial (X = 6, Y = 9, Z = 36) and face (X = 5, Y = 7, Z = 33) working memory tasks was sustained during the delay period in which participants had to hold information on line. One possible explanation for this sustained rCZ activity during the delay period, suggested by Petit, is that it reflects the fact that participants do not yet know what the final motor response will be, and thus are preparing to respond with either a right or left button press. Thus, in terms of our response conflict hypothesis, this simultaneous activation of motor programs associated with both a right and left button press may generate conflict between incompatible motor representations. However, at present this is a post hoc explanation for the presence of rCZ activity during working memory tasks, and additional research is needed to determine whether or not rCZ activation in working memory tasks reflects response conflict.

Notes

The authors thank Steve Petersen and David Donaldson for their thoughtful comments and helpful suggestions, Marcus Raichle for his support in carrying out this work, and David Molfese and Sarah Lageman for their help in conducting the study. This work was supported by a grant from a government agency.

Address correspondence to Deanna M. Barch, Department of Psychology, Washington University, Campus Box 1125, One Brookings Drive, St Louis, MO 63130, USA. Email: dbarch@artsci.wustl.edu.

Table 1

Stereotactic coordinates of activations in ACC for vocal responses

Reference Comparison ACLa 
  X Y Z 
aACL = Anterior cingulate location. 
(Barch et al., 1999) generate words–read words  27 38 
(Barch et al., 2000) generate words: high–low selection  30 14 
 generate words: weak–dominant response  28 29 
(Baker et al., 1997) generate words–repeat words  –4  18 36 
(Bench et al., 1993) Stroop task: incongruent–cross control  –4 20 
 Stroop task: word control–cross control  10  –4 24 
 Stroop task: incongruent–cross control  18  40 
 Stroop task: incongruent–cross control  20  42 
 Stroop task: incongruent–cross control  22  42 12 
(Brown et al., 1999) Stroop task: incongruent–color naming  23 35 
 Stroop task: incongruent–color naming  –4  14 35 
(Buckner et al., 1995) stem completion–fixation  20 30 
(Carter et al., 1995) Stroop task: incongruent–neutral  –8  22 28 
 Stroop task: incongruent–neutral  12  44 20 
(Carter et al., 2000) Stroop task: expectancy (mostly congruent, mostly incongruent)  trial type (congruent, incongruent) scan with trial  15 41 
(Derbyshire et al., 1998) Stroop task: incongruent–congruent  –2  14 40 
 Stroop task: incongruent–congruent 48 
(de Zubicaray et al., 19980 random letter generation–recite alphabet  14 42 
 random letter generation–recite alphabet  31 20 
(Dye et al., 1999) generate words–repeat words  26 40 
(Friston et al., 1993) generate words–repeat letters  –2  18 24 
(Frith et al., 1991) generate words–repeat words  23 36 
(George et al., 1994) Stroop task: incongruent–naming color bars –20 28 
 Stroop task: incongruent–naming color bars –22  24 32 
 Stroop task: incongruent–naming color bars  26 –10 28 
(George et al., 1997) Stroop task: incongruent–naming color bars –22 28 
(Grasby et al., 1993) word retrieval from memory–rest  22 28 
 Word retrieval from memory–rest  18 32 
(Pardo et al., 1990) Stroop task: incongruent–congruent trials  10  19 32 
 Stroop task: incongruent–trials  17 32 
 Stroop task: incongruent–congruent trials  17  25 30 
 Stroop task: incongruent–congruent trials  13  44 22 
(Paus et al., 1993) word association: reversal–rest  13 48 
 word association: reversal–rest  20 38 
 word association: reversal–fixation  34 13 
 word association: reversal–fixation  15 49 
 word association: reversal–overpracticed  30 17 
 word association: reversal–overpracticed  34 22 
 word association: reversal–overpracticed  22 49 
 word association: reversal–overpracticed  18 44 
 word association: reversal–overpracticed  20 36 
(Petersen et al., 1989) generate words–repeat words  14 41 
 generate words–repeat words  24 30 
 generate words–repeat words  18 41 
 generate words–repeat words  11  21 30 
(Peterson et al., 1999) Stroop task: incongruent–congruent  –7  26 27 
 Stroop task: incongruent–congruent  –7  18 36 
 Stroop task: incongruent–congruent  26 27 
(Phelps et al., 1997) generate words–repeat words  –4  20 40 
 generate words–repeat words  17 27 
 generate words–repeat words –12  17 27 
(Petrides et al., 1993) random number generation–counting  11  25 22 
(Raichle et al., 1994) naive generate words–practiced generate words  –4  28 36 
(Sergent et al., 1992) letter sound discrimination–object discrimination  18 31 
(Taylor et al., 1994) incongruent–congruent letter naming  10  14 43 
(Taylor et al., 1997) Stroop task: incongruent–neutral  –3  35 18 
(Thompson-Schill et al., 1997) generate words: high selection–low selection  –4  11 45 
(Vandenberghe et al., 1997) attend to two–attend to one feature –10  20 32 
(Warburton et al., 1996) word generation–word comparison –12 44 
 word generation–listening –12  14 48 
 word generation–listening  16 20 
Average coordinates verbal responses  20 31 
Reference Comparison ACLa 
  X Y Z 
aACL = Anterior cingulate location. 
(Barch et al., 1999) generate words–read words  27 38 
(Barch et al., 2000) generate words: high–low selection  30 14 
 generate words: weak–dominant response  28 29 
(Baker et al., 1997) generate words–repeat words  –4  18 36 
(Bench et al., 1993) Stroop task: incongruent–cross control  –4 20 
 Stroop task: word control–cross control  10  –4 24 
 Stroop task: incongruent–cross control  18  40 
 Stroop task: incongruent–cross control  20  42 
 Stroop task: incongruent–cross control  22  42 12 
(Brown et al., 1999) Stroop task: incongruent–color naming  23 35 
 Stroop task: incongruent–color naming  –4  14 35 
(Buckner et al., 1995) stem completion–fixation  20 30 
(Carter et al., 1995) Stroop task: incongruent–neutral  –8  22 28 
 Stroop task: incongruent–neutral  12  44 20 
(Carter et al., 2000) Stroop task: expectancy (mostly congruent, mostly incongruent)  trial type (congruent, incongruent) scan with trial  15 41 
(Derbyshire et al., 1998) Stroop task: incongruent–congruent  –2  14 40 
 Stroop task: incongruent–congruent 48 
(de Zubicaray et al., 19980 random letter generation–recite alphabet  14 42 
 random letter generation–recite alphabet  31 20 
(Dye et al., 1999) generate words–repeat words  26 40 
(Friston et al., 1993) generate words–repeat letters  –2  18 24 
(Frith et al., 1991) generate words–repeat words  23 36 
(George et al., 1994) Stroop task: incongruent–naming color bars –20 28 
 Stroop task: incongruent–naming color bars –22  24 32 
 Stroop task: incongruent–naming color bars  26 –10 28 
(George et al., 1997) Stroop task: incongruent–naming color bars –22 28 
(Grasby et al., 1993) word retrieval from memory–rest  22 28 
 Word retrieval from memory–rest  18 32 
(Pardo et al., 1990) Stroop task: incongruent–congruent trials  10  19 32 
 Stroop task: incongruent–trials  17 32 
 Stroop task: incongruent–congruent trials  17  25 30 
 Stroop task: incongruent–congruent trials  13  44 22 
(Paus et al., 1993) word association: reversal–rest  13 48 
 word association: reversal–rest  20 38 
 word association: reversal–fixation  34 13 
 word association: reversal–fixation  15 49 
 word association: reversal–overpracticed  30 17 
 word association: reversal–overpracticed  34 22 
 word association: reversal–overpracticed  22 49 
 word association: reversal–overpracticed  18 44 
 word association: reversal–overpracticed  20 36 
(Petersen et al., 1989) generate words–repeat words  14 41 
 generate words–repeat words  24 30 
 generate words–repeat words  18 41 
 generate words–repeat words  11  21 30 
(Peterson et al., 1999) Stroop task: incongruent–congruent  –7  26 27 
 Stroop task: incongruent–congruent  –7  18 36 
 Stroop task: incongruent–congruent  26 27 
(Phelps et al., 1997) generate words–repeat words  –4  20 40 
 generate words–repeat words  17 27 
 generate words–repeat words –12  17 27 
(Petrides et al., 1993) random number generation–counting  11  25 22 
(Raichle et al., 1994) naive generate words–practiced generate words  –4  28 36 
(Sergent et al., 1992) letter sound discrimination–object discrimination  18 31 
(Taylor et al., 1994) incongruent–congruent letter naming  10  14 43 
(Taylor et al., 1997) Stroop task: incongruent–neutral  –3  35 18 
(Thompson-Schill et al., 1997) generate words: high selection–low selection  –4  11 45 
(Vandenberghe et al., 1997) attend to two–attend to one feature –10  20 32 
(Warburton et al., 1996) word generation–word comparison –12 44 
 word generation–listening –12  14 48 
 word generation–listening  16 20 
Average coordinates verbal responses  20 31 
Table 2

Stereotactic coordinates of activations in ACC for manual responses

Reference Comparison ACC region of interest 
  X Y Z 
(Botvinick et al., 1999) Eriksen flanker task: incongruent–congruent  –2 28 31 
(Bush et al., 1998) counting Stroop task: incongruent–neutral  12 34 
(Carter et al., 1998) CPT-AX: high–low conflict trials 25 43 
(Corbetta et al., 1991) divided attention–passive control  –7 23 34 
 divided attention–passive control –11 45 24 
(Deiber et al., 1991) random–fixed joystick movements 34 32 
(Frith et al., 1991) random–directed finger movements 16 34 
(Garavan et al., 1999) Go/No-go task: correct no-go trials–fixation 16 42 
(Hyder et al., 1997) random–directed finger movements  12 16 23 
 random–directed finger movements  10 19 36 
(Jenkins et al., 1994) new key press sequence–rest 22 28 
 new–learned key press sequence 30 28 
(Jueptner et al., 1997) random–fixed key press sequence  –2 48 
 random–fixed key press sequence 16 32 
 new–fixed key press sequence 16 36 
(Jueptner et al., 1997) new–learned key press sequence 20 28 
 new–attend to learned key press sequence  –2 16 44 
 new–attend to learned key press sequence  18 34 –4 
(Kawashima et al., 1996) Go/No-go–response selection 42 
 Go/No-go–response selection  –4 38 
 Go/No-go–response selection  –4 30 
(Kiehl et al., 2001) Go/No-go: errors–fixation 22 40 
 Go/No-go: errors–fixation  12 36 12 
 Go/No-go: correct rejections–fixation 45 
 Go/No-go: errors–correct rejections 22 40 
 Go/No-go: errors–correct rejections  –8 45 15 
 Go/No-go: correct hits–fixation 20 40 
(Klingberg, 1998) auditory/visual dual task–control 16 44 
(Klingberg and Roland, 1997) Go/No-go–passive control  –2 16 45 
(Paus et al., 1993) condition key press: reversal–fixation 49 
 condition key press: reversal–overpracticed  15 49 
 opposite finger movement–rest –4 48 
 opposite-directed finger movement 22 38 
 opposite-directed finger movement 10 45 
(Petrides et al., 1993) self-ordered pointing–directed pointing 34 26. 
 self-ordered pointing–directed pointing 29 29 
 self-ordered pointing–directed pointing 24 40 
 conditional pointing–directed pointing 30 21 
(Playford et al., 1992) random joystick movement–rest 27 24 
 random joystick movement–rest 25 28 
(Samuel et al., 1998) random joystick movement–rest  –4 44 
(Vandenberghe et al., 1999) reversed–recognized stimulus/response associations –18 20 32 
 reversed–recognized stimulus/response associations –12 22 48 
(Whalen et al.,1998) counting Stroop task: neutral–fixation –6 46 
 counting Stroop task: neutral–fixation  –3 –6 50 
(Zatorre et al., 1992) alternate key press–rest 42 
(Zatorre et al., 1994) conditional key press–repetitive key press 18 29 
 conditional key press–repetitive key press 24 42 
 conditional key press–repetitive key press 36 26 
Average coordinates manual responses 19 35 
Reference Comparison ACC region of interest 
  X Y Z 
(Botvinick et al., 1999) Eriksen flanker task: incongruent–congruent  –2 28 31 
(Bush et al., 1998) counting Stroop task: incongruent–neutral  12 34 
(Carter et al., 1998) CPT-AX: high–low conflict trials 25 43 
(Corbetta et al., 1991) divided attention–passive control  –7 23 34 
 divided attention–passive control –11 45 24 
(Deiber et al., 1991) random–fixed joystick movements 34 32 
(Frith et al., 1991) random–directed finger movements 16 34 
(Garavan et al., 1999) Go/No-go task: correct no-go trials–fixation 16 42 
(Hyder et al., 1997) random–directed finger movements  12 16 23 
 random–directed finger movements  10 19 36 
(Jenkins et al., 1994) new key press sequence–rest 22 28 
 new–learned key press sequence 30 28 
(Jueptner et al., 1997) random–fixed key press sequence  –2 48 
 random–fixed key press sequence 16 32 
 new–fixed key press sequence 16 36 
(Jueptner et al., 1997) new–learned key press sequence 20 28 
 new–attend to learned key press sequence  –2 16 44 
 new–attend to learned key press sequence  18 34 –4 
(Kawashima et al., 1996) Go/No-go–response selection 42 
 Go/No-go–response selection  –4 38 
 Go/No-go–response selection  –4 30 
(Kiehl et al., 2001) Go/No-go: errors–fixation 22 40 
 Go/No-go: errors–fixation  12 36 12 
 Go/No-go: correct rejections–fixation 45 
 Go/No-go: errors–correct rejections 22 40 
 Go/No-go: errors–correct rejections  –8 45 15 
 Go/No-go: correct hits–fixation 20 40 
(Klingberg, 1998) auditory/visual dual task–control 16 44 
(Klingberg and Roland, 1997) Go/No-go–passive control  –2 16 45 
(Paus et al., 1993) condition key press: reversal–fixation 49 
 condition key press: reversal–overpracticed  15 49 
 opposite finger movement–rest –4 48 
 opposite-directed finger movement 22 38 
 opposite-directed finger movement 10 45 
(Petrides et al., 1993) self-ordered pointing–directed pointing 34 26. 
 self-ordered pointing–directed pointing 29 29 
 self-ordered pointing–directed pointing 24 40 
 conditional pointing–directed pointing 30 21 
(Playford et al., 1992) random joystick movement–rest 27 24 
 random joystick movement–rest 25 28 
(Samuel et al., 1998) random joystick movement–rest  –4 44 
(Vandenberghe et al., 1999) reversed–recognized stimulus/response associations –18 20 32 
 reversed–recognized stimulus/response associations –12 22 48 
(Whalen et al.,1998) counting Stroop task: neutral–fixation –6 46 
 counting Stroop task: neutral–fixation  –3 –6 50 
(Zatorre et al., 1992) alternate key press–rest 42 
(Zatorre et al., 1994) conditional key press–repetitive key press 18 29 
 conditional key press–repetitive key press 24 42 
 conditional key press–repetitive key press 36 26 
Average coordinates manual responses 19 35 
Table 3

Stereotactic coordinates of activations in ACC for oculomotor responses

Reference Comparison Anterior cingulate location 
  X Y Z 
(Anderson et al., 1994) remembered saccades–reflexive saccades 44 
(Doricchi et al., 1997) antisaccade–fixation 32 20 
 antisaccade–fixation –2 26 16 
 antisaccade–prosaccade –6 26 12 
 antisaccade–prosaccade 26 16 
(O'Driscoll et al., 1995) antisaccade–prosaccade –2 10 44 
(Paus et al., 1993) conditional saccades: reversal–overpracticed 27 29 
 conditional saccades: reversal–fixation 29 22 
 conditional saccades: reversal–fixation 32 12 
 opposite–targeted saccades 10 42 
(Sweeney et al., 1996) antisaccade–prosaccade –4 28 –4 
Average coordinates oculomotor responses 24 23 
Reference Comparison Anterior cingulate location 
  X Y Z 
(Anderson et al., 1994) remembered saccades–reflexive saccades 44 
(Doricchi et al., 1997) antisaccade–fixation 32 20 
 antisaccade–fixation –2 26 16 
 antisaccade–prosaccade –6 26 12 
 antisaccade–prosaccade 26 16 
(O'Driscoll et al., 1995) antisaccade–prosaccade –2 10 44 
(Paus et al., 1993) conditional saccades: reversal–overpracticed 27 29 
 conditional saccades: reversal–fixation 29 22 
 conditional saccades: reversal–fixation 32 12 
 opposite–targeted saccades 10 42 
(Sweeney et al., 1996) antisaccade–prosaccade –4 28 –4 
Average coordinates oculomotor responses 24 23 
Table 4

Regions demonstrating significant response conflict effects in conjunction analyses

Regions of interest Brodmann area(s) Xa Ya Za No. of voxels 
aX, Y and Z are coordinates in a standard stereotactic space (Talairach and Tournoux, 1988) in which positive values refer to regions right of (X), anterior to (Y) and superior to (Z) the anterior commissure (AC). 
Both vocal and manual responses      
 Anterior cingulate 24/32  7.5 42 29 
Location–manual and word–vocal      
 Anterior cingulate 24  1.5 18 24 
 Anterior cingulate 24/32  7.5 42 28 
 SMA –13.5 54 
Both word and location tasks      
 Anterior cingulate 24  7.5 21 24 
 Anterior cingulate 24/32  10.5 42 11 
 Anterior cingulate 32  4.5 15 39 14 
 SMA –13.5 54 
 Pre-SMA  –4.5 21 48 
Regions of interest Brodmann area(s) Xa Ya Za No. of voxels 
aX, Y and Z are coordinates in a standard stereotactic space (Talairach and Tournoux, 1988) in which positive values refer to regions right of (X), anterior to (Y) and superior to (Z) the anterior commissure (AC). 
Both vocal and manual responses      
 Anterior cingulate 24/32  7.5 42 29 
Location–manual and word–vocal      
 Anterior cingulate 24  1.5 18 24 
 Anterior cingulate 24/32  7.5 42 28 
 SMA –13.5 54 
Both word and location tasks      
 Anterior cingulate 24  7.5 21 24 
 Anterior cingulate 24/32  10.5 42 11 
 Anterior cingulate 32  4.5 15 39 14 
 SMA –13.5 54 
 Pre-SMA  –4.5 21 48 
Table 5

Regions demonstrating selective response conflict effects in disjunction analyses

Regions of interest Brodmann area(s) Xa Ya Za No. of voxels 
aX, Y, and Z are coordinates in a standard stereotactic space (Talairach and Tournoux, 1988) in which positive values refer to regions right of (X), anterior to (Y), and superior to (Z) the anterior commissure (AC). 
Manual but not vocal responses      
 Cerebellum  –  –1.5 –66 –15 
 Inferior frontal cortex 47 –28.5  15  –3 
 Inferior frontal cortex 44/6  49.5 
 Superior parietal cortex 40  31.5 –36  45 
 SMA  10.5 –21  54 
Vocal but not manual responses      
 Middle frontal cortex  25.5  18  39 
 Location–manual but not word–vocal 24  1.5  18  21 
 Primary motor cortex  1/2  46.5 –24  30 
 Inferior frontal cortex 45 –22.5  30 
 Middle frontal cortex 46/9 –40.5  36  24 
Word–vocal but not location–manual      
 Visual cortex 18  –7.5 –78 
 Visual cortex 18  10.5 –81 
 Temporal cortex 21  58.5 –21 –21 
 Temporal cortex 21/22 –64.5 –51 
 Inferior frontal cortex 47 –49.5  24  –6 
 Middle frontal cortex  13.5  60  27 
Word but not location task      
 Visual cortex 17/18  –7.5 –81  –3 38 
 Visual cortex 17  –7.5 –90 –12 
 Visual cortex 17/18  13.5 –81 28 
 Visual cortex 18  1.5 –81  12 11 
 Cerebellum  – –16.5 –57  –9 
 Parietal cortex 40 –61.5 –24  24 
Regions of interest Brodmann area(s) Xa Ya Za No. of voxels 
aX, Y, and Z are coordinates in a standard stereotactic space (Talairach and Tournoux, 1988) in which positive values refer to regions right of (X), anterior to (Y), and superior to (Z) the anterior commissure (AC). 
Manual but not vocal responses      
 Cerebellum  –  –1.5 –66 –15 
 Inferior frontal cortex 47 –28.5  15  –3 
 Inferior frontal cortex 44/6  49.5 
 Superior parietal cortex 40  31.5 –36  45 
 SMA  10.5 –21  54 
Vocal but not manual responses      
 Middle frontal cortex  25.5  18  39 
 Location–manual but not word–vocal 24  1.5  18  21 
 Primary motor cortex  1/2  46.5 –24  30 
 Inferior frontal cortex 45 –22.5  30 
 Middle frontal cortex 46/9 –40.5  36  24 
Word–vocal but not location–manual      
 Visual cortex 18  –7.5 –78 
 Visual cortex 18  10.5 –81 
 Temporal cortex 21  58.5 –21 –21 
 Temporal cortex 21/22 –64.5 –51 
 Inferior frontal cortex 47 –49.5  24  –6 
 Middle frontal cortex  13.5  60  27 
Word but not location task      
 Visual cortex 17/18  –7.5 –81  –3 38 
 Visual cortex 17  –7.5 –90 –12 
 Visual cortex 17/18  13.5 –81 28 
 Visual cortex 18  1.5 –81  12 11 
 Cerebellum  – –16.5 –57  –9 
 Parietal cortex 40 –61.5 –24  24 
Figure 1.

 Plot of ACC activations associated with different response modalities in Talairach space (see Table 1 for coordinates).

 Plot of ACC activations associated with different response modalities in Talairach space (see Table 1 for coordinates).

Figure 2.

 Graph plotting reaction times for behavioral data acquired during fMRI scanning. Error rates are shown at the bottom of the graph bar for each condition.

Figure 2.

 Graph plotting reaction times for behavioral data acquired during fMRI scanning. Error rates are shown at the bottom of the graph bar for each condition.

Figure 3.

 ACC regions exhibiting significantly response conflict effects (greater event-related activation to incongruent than neutral trials) for both manual and vocal responses. Insets plot percent signal change (averaged across all voxels within a ROI) for the seven time points following the onset of the stimulus, separately for neutral and incongruent trials.

Figure 3.

 ACC regions exhibiting significantly response conflict effects (greater event-related activation to incongruent than neutral trials) for both manual and vocal responses. Insets plot percent signal change (averaged across all voxels within a ROI) for the seven time points following the onset of the stimulus, separately for neutral and incongruent trials.

Figure 4.

 ACC regions exhibiting significantly response conflict effects (greater event-related activation to incongruent than neutral trials) for both the word and location tasks. The insets plot percent signal change (averaged across all voxels within a ROI) for the seven time points following the onset of the stimulus, separately for neutral and incongruent trials.

Figure 4.

 ACC regions exhibiting significantly response conflict effects (greater event-related activation to incongruent than neutral trials) for both the word and location tasks. The insets plot percent signal change (averaged across all voxels within a ROI) for the seven time points following the onset of the stimulus, separately for neutral and incongruent trials.

Figure 5

. Regions within primary motor and supplementary motor cortex, demonstrating differential activation as a function of response modality.

Figure 5

. Regions within primary motor and supplementary motor cortex, demonstrating differential activation as a function of response modality.

References

Akbudak E, Gusnard DA, Snyder AZ, Rosen HJ, Raichle ME, Conturo TE (1999) fMRI techniques for general auditory stimulation and vocal response monitoring. International Society for Magnetic Resonance in Medicine Meeting, Philadelphia, PA, p. 1663.
Anderson TJ, Jenkins IH, Brooks DJ, Hawken MB, Frackowiak RSJ, Kennard C (
1994
) Cortical control of saccades and fixation in man: a PET study.
Brain
 
117
:
1073
–1084.
Baker SC, Frith CD, Dolan RJ (
1997
) The interaction between mood and cognitive function studied with PET.
Psychol Med
 
27
:
565
–578.
Barch DM, Carter CS, Braver TS, Sabb FW, Noll DC, Cohen JC (
1999
) Overt verbal responding during fMRI scanning: empirical investigations of problems and potential solutions.
Neuroimage
 
10
:
642
–657.
Barch DM, Sabb FW, Noll DC (
2000
) The anterior cingulate cortex and response competition: evidence from an fMRI study of overt verb generation.
J Cogn Neurosci
 
12
:
298
–305.
Bench CJ, Frith CD, Grasby PM, Friston KJ, Paulesu E, Frackowiak RSJ, Dolan RJ (
1993
) Investigations of the functional anatomy of attention using the Stroop test.
Neuropsychologia
 
31
:
907
–922.
Botvinick MM, Nystrom L, Fissel K, Carter CS, Cohen JD (
1999
) Conflict monitoring versus selection-for-action in anterior cingulate cortex.
Nature
 
402
:
179
–181.
Botvinick MM, Braver TS, Barch DM, Carter CS, Cohen JC (2001) Evaluating the demand for control: anterior cingulate cortex and conflict monitoring. Psychol Rev (in press).
Boynton GM, Engel SA, Glover GH, Heeger DJ (
1996
) Linear systems analysis of functional magnetic resonance imaging in human V1.
J Neurosci
 
16
:
4207
–4221.
Braver TS, Cohen JD, Nystrom LE, Jonides J, Smith EE, Noll DC (
1997
) A parametric study of prefrontal cortex involvement in human working memory.
Neuroimage
 
5
:
49
–62.
Brown GG, Kindermann SS, Siegle GJ, Granholm E, Wong EC, Buxton RD (
1999
) Brain activation and pupil response during covert performance of the stroop color word task.
J Int Neuropsychol Soc
 
5
:
308
–319.
Buckner RL, Petersen SE, Ojemann JG, Miezin FM, Squire LR, Raichle ME (
1995
) Functional anatomical studies of explicit and implicit memory retrieval tasks.
J Neurosci
 
15
:
12
–29.
Buckner RL, Koutstaal W, Schacter DL, Dale AM, Rotte M, Rosen BR (
1998
) Functional–anatomic study of episodic retrieval. II. Selective averaging of event-related fMRI trials to test the retrieval success hypothesis.
NeuroImage
 
7
:
163
–175.
Bush G, Whalen PJ, Rosen BR, Jenike MA, McInerney SC, Rauch SL (
1998
) The counting Stroop: an interference task specialized for functional neuroimaging – validation study with functional MRI.
Hum Brain Mapp
 
6
:
270
–282.
Carter CS, Mintun M, Cohen JD (
1995
) Interference and facilitation effects during selective attention: an [15O]-H2O PET study of Stroop task performance.
NeuroImage
 
2
:
264
–272.
Carter CS, Braver TS, Barch DM, Botvinick MM, Noll DC, Cohen JD (
1998
) Anterior cingulate cortex, error detection, and the online monitoring of performance.
Science
 
280
:
747
–749.
Carter CS, Macdonald AM, Botvinick M, Ross LL, Stenger A, Noll D, Cohen JD (
2000
) Parsing executive processes: strategic versus evaluative functions of the anterior cingulate cortex.
Proc Natl Acad Sci
 
97
:
1944
–1948.
Casey BJ, Trainor RJ, Orendi JL, Schubert AB, Nystrom LE, Geidd JN, Castellanos FX, Haxby JV, Noll DC, Cohen JD, Forman SD, Dahl RE, Rapoport JL (
1997
) A developmental functional MRI study of pre- frontal activation during performance of a go-no-go task.
J Cogn Neurosci
 
9
:
835
–847.
Cohen JD, MacWhinney B, Flatt MR, Provost J (
1993
) PsyScope: a new graphic interactive environment for designing psychology experiments.
Behav Res Meth Instruments Comput
 
25
:
257
–271.
Cohen JD, Perstein WM, Braver TS, Nystrom LE, Noll DC, Jonides J, Smith EE (
1997
) Temporal dynamics of brain activation during a working memory task.
Nature
 
386
:
604
–608.
Conturo TE, McKinstry RC, Akbudak E, Snyder AZ, Yang T, Raichle ME (1996) Sensitivity optimization and experimental design in functional magnetic resonance imaging. In: Society for neuroscience. Washington, DC: Society for Neuroscience, p. 7.
Corbetta M, Miezin FM, Dobmeyer S, Shulman GL, Petersen SE (
1991
) Selective and divided attention during visual discriminations of shape, color, and speed: functional anatomy by positron emission tomography.
J Neurosci
 
11
:
2383
–2402.
Crosson B, Sadek JR, Bobholz JA, Gokcay D, Mohr CM, Leonard CM, Maron L, Auerbach EJ, Browd SR, Freeman AJ, Briggs RW (
1999
) Activity in the paracingulate and cingulate sulci during word generation: an fMRI study of functional anatomy.
Cereb Cortex
 
9
:
307
–316.
Dahaene S, Posner MI, Tucker DM (
1994
) Localization of a neural system for error detection and compensation.
Psychol Sci
 
5
:
303
–305.
Dale AM, Buckner RL (
1997
) Selective averaging of rapidly presented individual trials using fMRI.
Hum Brain Mapp
 
5
:
329
–340.
Deiber MP, Passingham RE, Colebatch JG, Friston KJ, Nixon PD, Frackowiak RSJ (
1991
) Cortical areas and the selection of movement: a study with positron emission tomography.
Exp Brain Res
 
84
:
393
–402.
D'Esposito M, Aguirre GK, Zarahn E, Ballard D, Shin RK, Lease J (
1998
) Functional MRI studies of spatial and nonspatial working memory.
Cogn Brain Res
 
7
:
1
–13.
de Zubicaray GI, Williams SCR, Wilson SJ, Rose SE, Brammer MJ, Bullmore ET, Simmons A, Chalk JB, Semple J, Brown AP, Smith GA, Ashton R, Doddrell DM (
1998
) Prefrontal cortex involvement in selective letter generation: a functional magnetic resonance imaging study.
Cortex
 
34
:
389
–401.
Derbyshire SWG, Vogt BA, Jones AKP (
1998
) Pain and Stroop interference tasks activate separate processing modules in anterior cingulate cortex.
Exp Brain Res
 
118
:
52
–60.
Devinsky O, Morrell MJ, Vogt B (
1995
) Contributions of anterior cingulate cortex to behavior.
Brain
 
118
:
279
–306.
Doricchi F, Perani D, Incoccia C, Grassi F, Cappa SF, Bettinardi V, Galati G, Pizzamiglio L, Fazio F (
1997
) Neural control of fast-regular saccades and antisaccades: an investigation using positron emission tomography.
Exp Brain Res
 
116
:
50
–62.
Dye SM, Spence A, Bench CJ, Hirsch SR, Stefan MD, Sharma T, Grasby PM (
1999
) No evidence for left superior temporal dysfunction in asymptomatic schizophrenia and bipolar disorder.
Br J Psychiatry
 
175
:
367
–374.
Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC (
1995
) Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold.
Magn Reson Med
 
33
:
636
–647.
Friston KJ, Frith CD, Liddle PF, Frackowiak RSJ (
1993
) Functional connectivity: the principal-component analysis of large (PET) data sets.
J Cereb Blood Flow Metab
 
13
:
5
–14.
Friston KJ, Jezzard P, Turner R (
1994
) The analysis of functional MRI time series.
Hum Brain Mapp
 ,
1
–45.
Friston KJ, Holmes AP, Price CJ, Buchel C, Worsley KJ (
1999
) Multisubject fMRI studies and conjunction analyses.
Neuroimage
 
10
:
385
–396.
Frith CD, Friston K, Liddle PF, Frackowiak RSJ (
1991
) Willed action and the prefrontal cortex in man: a study with PET.
Proc R Soc Lond
 
244
:
241
–246.
Frith CD, Friston KJ, Liddle PF, Frackowiak RSJ (
1993
) A PET study of word finding.
Neuropsychologia
 
29
:
1137
–1148.
Garavan H, Ross TJ, Stein EA (
1999
) Right hemispheric dominance of inhibitory control: an event-related functional MRI study.
Proc Natl Acad Sci USA
 
96
:
8301
–8306.
Gehring WJ, Coles MGH, Meyer DE, Donchin E (
1990
) The error- related negativity: an event-related potential accompanying errors.
Psychophysiology
 
27
:
S34
.
George MS, Ketter TA, Parekh PI, Rosinsky N, Ring H, Casey BJ, Trimble MR, Horwitz B, Herscovitch P, Post RM (
1994
) Regional brain activity when selecting a response despite interference: an O-15 PET study of the Stroop and an emotional Stroop.
Hum Brain Mapp
 
1
:
194
–209.
George MS, Ketter TA, Parekh PI, Rosinsky N, Ring HA, Pazzaglia PJ, Marangell LB, Callahan AM, Post RM (
1997
) Blunted left cingulate activation in mood disorder subjects during a response interference task (the Stroop).
J Neuropsychiatry
 
1
:
55
–63.
Grasby PM, Frith CD, Friston KJ, Bench C, Frackowiak RSJ, Dolan RJ (
1993
) Functional mapping of brain areas implicated in auditory– verbal memory function.
Brain
 
116
:
1
–20.
Hohnsbein J, Falkenstein M, Hoorman J (
1989
) Error processing in visual and auditory choice reaction tasks.
J Psychophysiol
 
3
:
32
.
Hyder F, Phelps EA, Wiggins CJ, Labar KS, Blamire AM, Shulman G (
1997
) ‘Willed action’: a functional MRI study of the human prefrontal cortex during a sensorimotor task.
Proc Natl Acad Science USA
 
94
:
6989
–6994.
Jenkins IH, Brooks DJ, Nixon PD, Frackowiak RSJ, Passingham RE (
1994
) Motor sequence learning: a study with positron emission tomography.
J Neurosci
 
14
:
3775
–3790.
Josephs O, Turner R, Friston KJ (
1997
) Event-related fMRI.
Hum Brain Mapp
 
5
:
243
–248.
Jueptner M, Frith CD, Brooks JD, Frackowiak RSJ, Passingham RE (
1997
) Anatomy of motor learning: II. Subcortical structures and learning by trial and error.
J Neurosci
 
77
:
1325
–1337.
Kawashima R, Satoh K, Itoh H, Ono S, Furumoto S, Gotoh R, Koyoma M, Yoshioka S, Takahashi T, Takahashi K, Yanagisawa T, Fukuda H (
1996
) Functional anatomy of GO/NO-GO discrimination and response selection: a PET study in man.
Brain Res
 
728
:
79
–89.
Kelley WM, Miezin FM, McDermott KB, Buckner RL, Raichle ME, Cohen NJ, Ollinger JM, Akbudak E, Conturo TE, Snyder AZ, Petersen SE (
1998
) Hemispheric specialization in human dorsal frontal cortex and medial temporal lobe for verbal and non-verbal memory encoding.
Neuron
 
20
:
927
–936.
Kiehl KA, Liddle PF, Hopfinger JB (
2000
) Error processing and the rostral anterior cingulate: an event-related fMRI study.
Psychophysiology
 
37
:
216
–223.
Klingberg T (
1998
) Concurrent performance of two working memory tasks: potential mechanisms of interference.
Cereb Cortex
 
8
:
593
–601.
Klingberg T, Roland PE (
1997
) Interference between two concurrent tasks is associated with activation of overlapping fields in the cortex.
Cogn Brain Res
 
6
:
1
–8.
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.
Miezin FM, Maccotta L, Ollinger JM, Petersen SE, Buckner RL (
2000
) Characterizing the hemodynamic response: effects of presentation rate, sampling procedure, and the possibility of ordering brain activity based on relative timing.
Neuroimage
 
11
:
735
–759.
Mugler JPI, Brookeman JR (
1990
) Three-dimensional magnetization- prepared rapid gradient-echo imaging (3D MP-RAGE).
Magn Reson Med
 
15
:
152
–157.
O'Driscoll GA, Alpert NM, Matthysse SW, Levy DL, Rauch SL, Holzman PS (
1995
) Functional neuroanatomy of antisaccade eye movements investigated with positron emission tomography.
Proc Natl Acad Sci USA
 
92
:
925
–929.
Pardo JV, Pardo PJ, Janer KW, Raichle ME (
1990
) The anterior cingulate cortex mediates processing selection in the stroop attentional conflict paradigm.
Proc Natl Acad Sci USA
 
87
:
256
–259.
Paus T, Petrides M, Evans AC, Meyer E (
1993
) Role of the human anterior cingulate cortex in the control of oculomotor, manual, and speech responses: a positron emission tomography study.
J Neurophysiol
 
70
:
453
–469.
Paus T, Koski L, Caramanos Z, Westbury C (
1998
) 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
 
9
:
R37
–R47.
Petersen SE, Fox PT, Posner MI, Mintun M, Raichle ME (
1989
) Positron emission tomographic studies of the processing of single words.
J Cogn Neurosci
 
2
:
153
–170.
Peterson BS, Skudlarski J, Gatenby C, Zhang H, Anderson AW, Gore JC (
1999
) An fMRI study of stroop color-word interference: evidence for cingulate subregions subserving multiple distributed attentional systems.
Biol Psychiatry
 
45
:
1237
–1258.
Petit L, Courtney SM, Ungerleider LG, Haxby JV (
1998
) Sustained activity in the medial wall during working memory delays.
J Neurosci
 
18
:
9429
–9437.
Petrides ME, Alivisatos B, Evans AC, Meyer E (
1993
) Dissociation of human mid-dorsolateral from posterior dorsolateral frontal cortex in memory processing.
Proc Natl Acad Sci USA
 
90
:
873
–877.
Petrides ME, Alivisatos B, Meyer E, Evans AC (
1993
) Functional activation of the human frontal cortex during the performance of verbal working memory tasks.
Proc Natl Acad Sci USA
 
90
:
878
–882.
Phelps EA, Hyder F, Blamire AM, Shulman RG (
1997
) FMRI of the prefrontal cortex during overt verbal fluency.
NeuroReport
 
8
:
561
–565.
Picard N, Strick P (
1996
) Motor areas of the medial wall: a review of their location and functional activation.
Cereb Cortex
 
6
:
342
–353.
Playford ED, Jenkins IH, Passingham RE, Nutt J, Frackowiak RSJ, Brooks DJ (
1992
) Impaired medial frontal and putamen activation in Parkinson's Disease: a positron emission tomography study.
Ann Neurol
 
32
:
151
–161.
Price CJ, Friston KJ (
1997
) Cognitive conjunction: a new approach to brain activation experiments.
NeuroImage
 
5
:
261
–270.
Raichle ME, Fiez JA, Videen TO, MacCleod AK, Pardo JV, Fox PT, Petersen SE (
1994
) Practice-related changes in human brain functional anatomy during nonmotor learning.
Cereb Cortex
 
4
:
8
–26.
Rauch SL, Jenike MA, Alpert NM, Baer L, Breiter HCR, Savage CR, Fischman AJ (
1994
) Regional cerebral blood flow measured during symptom provocations in obsessive-compulsive disorder using oxygen 15-labeled carbon dioxide and positron emission tomography.
Arch Gen Psychiatry
 
51
:
62
–70.
Rauch SL, Savage CR, Alpert NM, Miguel EC, Baer L, Breiter HC, Fischman AJ, Manzo PA, Moretti C, Jenike MA (
1995
) A positron emission tomographic study of simple phobic symptom provocation.
Arch Gen Psychiatry
 
52
:
20
–28.
Samuel M, Williams SCR, Leigh PN, Simmons A, Chakraborti S, Andrew CM, Friston KJ, Goldstein LH, Brooks DJ (
1998
) Exploring the temporal nature of hemodynamic responses of cortical motor areas using functional MRI.
Neurology
 
51
:
1567
–1575.
Sergent J, Zuck E, Levesque M, MacDonald B (
1992
) Positron emission tomography study of letter and object processing: empirical findings and methodological considerations.
Cereb Cortex
 
80
:
68
–80.
Snyder AZ (1996) Difference image versus ratio image error function forms in PET–PET realignment. In: Quantification of brain function using PET (Bailer D and Jones T, eds), pp. 131–137. San Diego, CA: Academic Press.
Talairach J, Tournoux P (1988) Co-planar stereotaxic atlas of the human brain. New York: Thieme.
Taylor SF, Kornblum S, Minoshima S, Oliver LM, Koeppe RA (
1994
) Changes in medial cortical blood flow with a stimulus-response compatibility task.
Neuropsychologia
 
32
:
249
–255.
Taylor SF, Kornblum S, Lauber EJ, Minoshima S, Koeppe RA (
1997
) Isolation of specific interference processing in the stroop task: PET activation studies.
NeuroImage
 
6
:
81
–92.
Thompson-Schill SL, D'Esposito M, Aguire GK, Farah MJ (
1997
) Role of left inferior prefrontal cortex in retrieval of semantic knowledge: a re-evaluation.
Proc Natl Acad Sci
 
94
:
14792
–14797.
Turken AU, Swick D (
1999
) Response selection in the human anterior cingulate cortex.
Nat Neurosci
 
2
:
920
–924.
Vandenberge R, Duncan J, Dupont P, Ward R, Poline J, Bormans G, Michiels J, Mortelmans L, Orban GA (
1997
) Attention to one or two features in left or right visual field: a positron emission tomography study.
J Neurosci
 
17
:
3739
–3750.
Vandenberghe R, Dupont P, Bormans G, Mortelmans L, Orban GA (
1999
) Brain activity underlying stereotyped and non-stereotyped retrieval of learned stimulus–response associations.
Eur J Neurosci
 
11
:
4037
–4050.
Warburton E, Wise RJS, Price CJ, Weiller C, Hadar U, Ramsay S, Frackowiak RSJ (
1996
) Noun and verb retrieval by normal subjects: studies with PET.
Brain
 
119
:
159
–179.
Whalen PJ, Bush G, McNally RJ, Wilhelm S, McInerney SC, Jenike MA, Rauch SL (
1998
) The emotional counting Stroop paradigm: a functional magnetic resonance imaging probe of the anterior cingulate affective division.
Biol Psychiatry
 
44
:
1219
–1228.
Worsley KJ, Friston KJ (
1995
) Analysis of fMRI time-series revisited — again.
Neuroimage
 
2
:
173
–181.
Yetkin FZ, Hammeke TA, Swanson SJ, Morris GL, Mueller WM, McAuliffe TL (
1995
) A comparison of functional MR activation patterns during silent and audible language tasks.
Am J Neuroradiol
 
16
:
1087
–1092.
Zarahn E, Aguirre GK, D'Esposito M (
1997
) Empirical analyses of BOLD fMRI statistics: I. Spatially unsmoothed data collected under null- hypothesis conditions.
NeuroImage
 
5
:
179
–197.
Zatorre RJ, Evans AC, Meyer E, Gjedde A (
1992
) Lateralization of phonetic and pitch discrimination in speech processing.
Science
 
256
:
846
–849.
Zatorre RJ, Evans AC, Meyer E (
1994
) Neural mechanisms underlying melodic perception and memory for pitch.
J Neurosci
 
14
:
1908
–1919.