Abstract

Genetic variability related to the catechol-O-methyltransferase (COMT) gene (Val158Met) has received increasing attention as a possible modulator of executive functioning and its neural correlates. However, this attention has generally centered on the prefrontal cortices because of the well-known direct impact of COMT enzyme on these cerebral regions. In this study, we were interested in the modulating effect of COMT genotype on anterior and posterior brain areas underlying interference resolution during a Stroop task. More specifically, we were interested in the functional connectivity between the right inferior frontal operculum (IFop), an area frequently associated with inhibitory efficiency, and posterior brain regions involved in reading/naming processes (the 2 main non-executive determinants of the Stroop effect). The Stroop task was administered during functional magnetic resonance imaging scanning to 3 groups of 15 young adults divided according to their COMT Val158Met genotype [Val/Val (VV), Val/Met (VM), and Met/Met (MM)]. Results indicate greater activity in the right IFop and the left middle temporal gyrus in homozygous VV individuals than in Met allele carriers. In addition, the VV group exhibited stronger positive functional connectivity between these 2 brain regions and stronger negative connectivity between the right IFop and left lingual gyrus. These results confirm the impact of COMT genotype on frontal functions. They also strongly suggest that differences in frontal activity influence posterior brain regions related to a non-executive component of the task. Particularly, changes in functional connectivity between anterior and posterior brain areas might correspond to compensatory processes for performing the task efficiently when the available dopamine level is low.

Introduction

Efficient inhibitory abilities are necessary to maintain an adequate level of adjustment to environmental demands. Inhibition is classically considered to consist of a set of processes that allow one to suppress the production of a predominant but inappropriate response in order to promote a more adapted one (e.g., Nigg 2000). The neural correlates of inhibition have been widely studied in the past 2 decades through perceptual, motor, and semantic paradigms. These studies showed the involvement of an extensive brain network including the cingulate, prefrontal, parietal, and temporal areas (Pardo et al. 1990; Bench et al. 1993; Larrue et al. 1994; Taylor et al. 1997; Bush et al. 1998; Garavan et al. 1999; Chee et al. 2000; Collette et al. 2001).

Among these brain areas, the anterior cingulate cortex (ACC) plays a central role in the detection and resolution of conflicting situations (Carter et al. 1998; Kerns et al. 2004). More specifically, in the context of the conflict monitoring hypothesis proposed by Botvinick et al. (2001), once a conflict is detected by the ACC, this area will recruit the dorsolateral prefrontal (DLPFC) cortices to adjust behavior to the conflicting situation. This adjustment will be made by biasing information processing in posterior brain areas toward the cognitive processes that are most relevant to task goal and context. Consequently, the DLPFC is considered to be the brain area specifically involved in implementing strategic adjustments in cognitive control when the ACC detects conflicting situations. Moreover, the activity in the DLPFC in response to conflicting situations is often associated with increased activity in the right inferior frontal gyrus (IFG; Garavan et al. 1999, 2002; Konishi et al. 1999; Rubia et al. 2003) and especially in the right pars triangularis of the IFG (for a review, see Aron et al. 2004, 2014).

A series of neuroimaging data support the conflict monitoring theory proposed by Botvinick et al. (2001, 2004), who argues that, following conflict detection, the ACC recruits the DLPFC, which in turn inhibits a response by biasing information processing in posterior brain areas. Kerns et al. (2004) showed that the activity in the ACC for conflicting trials predicted subsequent prefrontal cortex (PFC) activity and adjustments in behavior. Egner and Hirsch (2005) showed not only that response inhibition in situations of conflict adaptation was associated with increased activity in the PFC, but also that activity in this area was accompanied by increased functional integration with parietal and temporal gyri. Polk et al. (2008) observed that the presentation of interfering items during a Stroop task (Stroop 1935), which requires subjects to inhibit the overlearned reading process in favor of a color naming process, generated increased activity in brain areas previously associated with color processing (bilateral lingual gyrus), while brain activity in word processing areas (left fusiform gyrus) tended to decrease. As a whole, these studies demonstrated that inhibition of an irrelevant response or process involves the dynamic interplay of several anterior and posterior brain areas.

Several lines of evidence suggest that the neurotransmitter dopamine (DA) plays an important role in cognitive functions associated with prefrontal activity, such as executive processes (for a review, see Witte and Floel 2012). Catechol-O-methyltransferase (COMT) is the major enzyme involved in the metabolic degradation of released DA, accounting for >60% of DA degradation in the frontal cortex (Karoum et al. 1994). The human COMT gene, located on the long arm of chromosome 22q11 (Mannisto and Kaakkola 1999), contains a functional polymorphism in codon 158 (Val158Met), which affects the enzyme's activity (Lachman et al. 1996; Chen et al. 2004). A transition of guanine to adenine results in a valine-to-methionine substitution; consequently, there are 3 different COMT genotypes (GG, GA, and AA), corresponding, respectively, to Val158/Val158, Val158/ Met158, and Met158/ Met158. Each genotype is associated with different COMT enzymatic activity; the enzyme resulting from the Met158 variant is significantly less active than the Val158 enzyme, potentially resulting in a greater synaptic DA level (Lotta et al. 1995; Chen et al. 2004). Considering that DA plays an important role in human cognition (Kimberg et al. 1997; Mehta et al. 2000), COMT Val158Met polymorphism (rs4680) can be considered as a useful tool for investigating the modulating effect of DA on the brain areas associated with the conflict resolution processes.

Few studies so far have explored the influence of COMT polymorphism on the neural substrates of conflict/interference resolution processes, and most of them were interested in motor inhibition. For example, Congdon et al. (2009) used a stop-signal task and showed higher activity in the right inferior frontal operculum (IFop) during stop trials in carriers of the Met allele. Because all groups' behavioral performance was similar and several functional magnetic resonance imaging (fMRI) studies of the stop-signal task have demonstrated that enhanced responses relate to better inhibitory control in that task (Aron and Poldrack 2006; Li et al. 2006), the authors concluded that this result reflected better inhibitory control in Met allele carriers. Similarly, Stokes et al. (2011) found more activity in the right posterior cingulate gyrus in Met carriers for No-Go trials than for Go trials. However, no significant association was found between PFC activation and COMT genotype. In addition, the authors did not detect an effect of COMT genotype on functional connectivity between the posterior cingulate and anterior brain areas. Finally, Ettinger et al. (2008) reported a lower BOLD response in the ventromedial and dorsomedial PFC in Val carriers during an antisaccade task, again indicating more activity for Met carriers when oculomotor inhibition is required. Taken together, these results support the hypothesis that individuals homozygous for the Val allele are characterized by a less-efficient physiological response in cingulate and prefrontal areas during the motor inhibition processes. Nevertheless, it must be emphasized that fMRI effects in these studies were observed in the absence of any effect of genotype on behavioral efficiency. Interestingly, such an absence of effect was also observed when the procedure used was supposed to emphasize between-group differences. Indeed, Plewnia et al. (2013) observed no impact of COMT genotype on the accuracy of responses in a Go/No-Go task during transcranial direct current stimulation applied to the left DLPFC.

The influence of COMT polymorphism on perceptual inhibition was addressed by Blasi et al. (2005), who used a flanker task with 3 levels of attentional control. For the medium and high levels of attentional control, they showed a main effect of COMT genotype in the dorsal cingulate, with higher activity for VV individuals, followed by VM and then by MM. As MM individuals were also more accurate at the high attentional control level, this pattern of brain activity was interpreted as suggesting less-efficient cortical processing of stimuli and a less-efficient allocation of attentional resources in individuals who are homozygous for the Val allele. Finally, our group explored the effect of COMT polymorphism on cognitive control using a Stroop inhibition task (Jaspar et al. 2014). According to the Dual Mechanism of Control (DMC) theory (Braver et al. 2007), proactive control mechanisms, which are a sustained form of control, are specialized in interference prevention and anticipation, whereas reactive control mechanisms are dedicated to detecting and resolving interference when it occurs. Consequently, one strategy is favored over the other depending on whether there is a high or low occurrence of interfering events. In agreement with the DMC model, we observed that the neural substrates of proactive control are modulated by the level of DA available, with sustained increased activity in the ACC and decreased activity in the middle frontal gyrus in carriers of the Met allele. However, contrary to the model's predictions, we also observed an effect of DA in the reactive control condition, with individuals who are homozygous for the Val allele presenting consistently higher transient activity in the right IFop when interfering items were presented.

Aim of the Study and A Priori Hypothesis

We had previously observed (Jaspar et al. 2014) that, in a Stroop task, Val individuals exhibit greater transient activity in the right IFop when they must deal with the relatively infrequent presentation of interfering items (i.e., in the task condition requiring them to implement reactive control). This increased activity can be interpreted as reflecting less-efficient cortical processing of the presented information in frontal areas (for a similar interpretation, see Blasi et al. 2005). As previous studies had shown that performance on the Stroop task is associated with a large fronto-temporo-parietal network (e.g., Laird et al. 2005; Nee et al. 2007), the aim of the present study is to complement our previous analyses and determine whether the COMT polymorphism's effect on inhibition abilities can also be expressed on posterior brain areas and/or modulate functional connectivity between anterior and posterior brain areas.

With that objective, the data set acquired to test the dopaminergic hypothesis of the DMC account (Braver et al. 2007) in our previous study (Jaspar et al. 2014) was reanalyzed with a focus on the reactive control condition only. As Stroop tasks used in past studies (Banich et al. 2000; Ruff et al. 2001; Milham et al. 2002) were generally composed of around 50% interfering events, we considered the reactive condition (composed of 17% interfering items) to be more appropriate than the proactive one for exploring the neural substrates of the interference effect. Moreover, our reactive task condition is very similar to some task designs (e.g., Leung, et al. 2000; Grandjean et al. 2012) showing the classical fronto-parieto-temporal pattern of brain activity following presentation of interfering Stroop items. In contrast to our previous study, we extended the fMRI analyses of COMT influence on the interference effect to the whole brain. We also assessed the presence of a differential effect of COMT polymorphism on functional connectivity between frontal and posterior brain areas.

Our predictions were the following. We expected a modulating effect of COMT polymorphism on posterior brain regions in addition to the previously reported effect in the right IFop. Indeed, in the dynamic interplay between anterior and posterior regions during performance of a Stroop task (Botvinick et al. 2001, 2004), changes in prefrontal activity depending on the presence of Val or Met alleles should impact activity in the parietal and temporal areas also involved in the processing of interfering items. More specifically, we expected to observe an influence of COMT genotype in areas associated with reading [occipito-temporal junction, basal temporal area, middle temporal, and inferior frontal gyri (Jobard et al. 2003)] and color processing [bilateral lingual and fusiform gyri (Polk et al. 2008)]. Psychophysiological interactions (PPIs) were also computed to test the hypothesis that the functional connectivity of the right IFop and the rest of the brain differs for interfering items depending on COMT polymorphism. In fact, we expected that individuals who were homozygous for the Val allele would present: (1) A smaller positive association between the right IFop and the brain areas that facilitate interference resolution [the ACC, associated with conflict detection (Carter et al. 1998); the bilateral DLPFC, the right inferior parietal lobule and the left precuneus, associated with response inhibition during a Stroop task (Nee et al. 2007); and the bilateral lingual and fusiform gyri, associated with color detection (Polk et al. 2008)]; (2) a stronger positive association between the right IFop and brain regions that impede interference resolution [the left lateralized cerebral network associated with word reading (occipito-temporal junction, basal temporal area, middle temporal, and inferior frontal gyri; Jobard et al. 2003)].

Methods

Ethics Statement

The study was approved by the Ethics Committee of the Faculty of Medicine of the University of Liège. In accordance with the Declaration of Helsinki, all participants gave their written informed consent prior to their inclusion in the study.

Participants

One hundred and six Caucasian right-handed native French-speaking young adults, aged from 18 to 30, with no diagnosed psychological or neurological disorders, were recruited from the university community and paid for their participation. All had normal color vision. Each participant was also screened for any physical or medical condition that could prevent an MRI session. Through a DNA screening, our sample was separated into 3 groups according to their COMT genotype: 30 homozygous Val/Val (VV), 27 homozygous Met/Met (MM), and 49 heterozygous Val/Met (VM). Fifteen subjects were selected from each group in order to match them for gender (F2 = 0.60, P = 0.55), age (F2 = 0.94, P = 0.40), and fluid intelligence level (F2 = 1.96, P = 0.15; see Supplementary Table 1 in Supplementary Materials). Fluid intelligence level was assessed using Raven's advanced progressive matrices test (Raven et al. 1983).

Genotyping

Genomic DNA was extracted from blood samples using a MagNA Pure LC Instrument (Roche Applied Science). The DNA sequence of interest was amplified by polymerase chain reaction in a final volume of 50 µL containing 0.6 µM of each primer (Thermo Scientific), 0.5 µL of Faststart Taq DNA Polymerase (Roche Diagnostics), 0.8 mM of each deoxynucleotide triphosphate (Roche Diagnostics), and 100 ng of genomic DNA. After 10 min of denaturation at 95 °C, samples underwent 35 cycles consisting of denaturation (95 °C, 30 s), annealing (60 °C, 40 s), and extension (72 °C, 30 s), followed by a final extension of 7 min at 72 °C. The amplified DNA samples then underwent the pyrosequencing reaction (Pyromark Q96 Vacuum Workstation, PSQ 96MA, Pyromark Gold Q96 Reagents, Qiagen). The sequences of primers that were used are available upon request.

Materials and Procedure

A modified form of the Stroop task (Grandjean et al. 2012) with 4 words presented on a white background (the French equivalents of Red, Blue, Black, and Green) was used for this experiment and is described in full details in Jaspar et al. (2014; see also Supplementary Fig. 1). Proportion congruency was manipulated using 3 different contexts of 12 items each: The mostly incongruent (MI) context, the mostly congruent (MC) context, and the mostly neutral (MN) context. Each MI block was composed of 8 incongruent items (IIs; e.g., the word red written in blue), 2 congruent items (CIs; e.g., the word blue in blue), and 2 neutral items (NIs), which were nonverbal stimuli (i.e., strings of 5 per cent signs %%%%%) presented in 1 of the 4 color possibilities. For the MC context, the proportions of CIs and IIs were reversed. Finally, 8 NIs, 2 CIs, and 2 IIs were presented during the MN context. Importantly, the first 4 items in each block were representative of the current task context (e.g., 4 incongruent trials at the beginning of each MI context) and were intended to induce the use of proactive or reactive control processes. The presentation order of the different blocks was pseudorandomized, with the use of 3 different presentation orders. Each of the 3 congruency conditions of 12 items (MI, MC, and MN contexts) was presented 15 times, for a total of 45 blocks and 540 items.

The participants were instructed that their task would be to select the color in which each item was printed by pressing the corresponding key on a keyboard. They were told that the items would be presented briefly, and that they would have to respond as fast and accurately as possible. Color words were presented on a screen that the participants viewed through a mirror located on the scanner's head coil. Each trial consisted of the presentation of a word in the center of the screen, with 4 response possibilities at the bottom of the screen (corresponding to the 4 color possibilities, always in the same order). Each item was presented until the participant responded (with a maximum presentation time of 2000 ms). If the participant responded before the deadline, a white screen was presented for the remaining period. If no response was provided, a white screen appeared after 2000 ms and an interstimulus interval of 500 ms occurred before the next item. A fixation cross was presented in the center of the screen for 5000 ms after every 2 or 3 contexts to provide breaks during the experiment.

Prior to the fMRI session, participants performed a practice session outside the scanner in which 40 items were presented to be sure that they understood the task instructions. In the fMRI scanner, 4 more examples were presented just before the test phase began. After the session, participants received a debriefing that explained the main objective of the experiment.

Behavioral Data Analysis

All behavioral data analyses were conducted with a statistical level set at P < 0.05. Repeated-measures analyses of variance (ANOVAs) were run on the median reaction times (RTs) and accuracy data (proportion of correct responses), with task context (MC, MI, and MN) and item type (II, CI, and NI) as repeated-measures factors. Finally, planned comparisons were performed, also with a P-value of <0.05, using univariate tests of significance.

fMRI Acquisition and Analyses

Functional MRI time series were acquired on a 3-T head-only scanner (Magnetom Allegra, Siemens Medical Solutions, Erlangen, Germany) operated with the standard transmit-receive quadrature head coil. Structural images were obtained using a high-resolution T1-weighted sequence (3D Modified Driven Equilibrium Fournier Transform [MDEFT]; Deichmann et al. 2004; time repetition [TR] = 7.92 ms, time echo [TE] = 2.4 ms, time to inversion [TI] = 910 ms, FA = 15°, field of view [FOV] = 256 × 224 × 176 mm³, 1 mm isotropic spatial resolution). Multislice T2*-weighted functional images were acquired with a gradient echo-planar imaging (EPI) sequence using axial slice orientation and covering the whole brain (32 slices, FoV = 220 × 220 mm², voxel size 3.4 × 3.4 × 3 mm³, 30% interslice gap, matrix size 64 × 64 × 32, TR = 2130 ms, TE = 40 ms, flip angle [FA] = 90°). In each session, between 570 and 650 functional volumes were obtained. The first 3 volumes were discarded to account for T1 saturation. Stimuli were displayed on a screen positioned at the rear of the scanner, which the participant could see comfortably through a mirror mounted on the standard head coil.

Data were preprocessed and analyzed using SPM8 (Wellcome Trust Centre for Neuroimaging, http://www.fil.ion.ucl.ac.uk/spm) implemented in MATLAB 7.5.0 (Mathworks, Inc., Sherborn, MA, USA). Images of each individual participant were first realigned (motion-corrected). After this realignment, we spatially coregistered the mean EPI image to the anatomical MRI image, and coregistration parameters were applied to the realigned BOLD time series. Individual anatomical MRIs were spatially normalized in the MNI space (Montreal Neurological Institute, http://www.bic.mni.mcgill.ca), and the normalization parameters were subsequently applied to the individually coregistered BOLD times series, which was then smoothed using an isotropic 10-mm full-width at half-maximum Gaussian kernel.

For each participant, BOLD responses were modeled at each voxel, using a general linear model with events convolved with the canonical hemodynamic response function as regressors. Events were divided according to the 3 contexts (MI, MC, or MN) and the type of item (II, CI, or NI). These 9 regressors were modeled as event-related responses. Event durations corresponded to the presentation of the item until the subject's response, with a maximum duration of 2 s. Incorrect trials and non-responses were also modeled as separate regressors. The design matrix also included the realignment parameters to account for any residual movement-related effect. In addition, the first 4 items for each context were modeled separately in the design matrix. The rationale for excluding those items was that they did not fully reflect the cognitive control strategy postulated for the context in question (i.e., in the MI context, the first items served to establish the subsequent proactive control strategy by creating expectations associated with that context, and similarly in the MC context, the first items created a low expectation of incongruent trials). A high-pass filter was implemented using a cutoff period of 256 s in order to remove the low-frequency drifts from the time series. Linear contrasts assessed the simple main effect of each trial type. The resulting set of voxel values constituted a map of t-statistics, SPM[T]. The corresponding contrast images were entered into a second-level analysis, corresponding to a random-effect model.

At the second level (random-effect analysis), a 3 (context) × 3 (item type) whole-brain voxel-wise repeated-measures ANOVA was performed, which allowed us to examine the brain regions related to the comparisons of interest: (1) The general interference effect and (2) the interference effect in the MC context (involving reactive control). First, these individual contrast images were used to analyze neural activity common to the 3 genotype groups. These analyses were conducted using a correction for multiple comparisons at the voxel level with a conservative family-wise error (FWE) threshold of P < 0.05 corrected. Secondly, we focused on genotype-related differences in the neural correlates of interfering events in the whole task, but also when reactive control processes were specifically implemented. T-test comparisons between genotypes were performed at a P-value of <0.001 uncorrected. Only the analyses assessing between-group comparisons of the neural substrates of the interference effect are detailed in the main text. The extent threshold was set to >5 contiguous voxels. These analyses were first performed by comparing each genotype with the 2 others separately and next by grouping together participants carrying at least one Val or Met allele, respectively (VV and VM vs. MM, and VV vs. VM and MM). We will present and consider as relevant for the discussion only brain areas that are consistently found to be significant across these 2 analyses. The results of the following analyses are not reported in this paper, but are available in our previous work (Jaspar et al. 2014): General interference effect common to the 3 groups across the MC, MI and MN contexts; interference effects common to the 3 groups and specific to the MC context.

To assess the hypothesis that the IFop, involved in interference resolution, interacts differently with the ACC and posterior brain areas between our groups, we conducted PPI analyses. The logic underlying the choice of the right IFop as region of interest (ROI) for these analyses was that we had already shown that COMT genotype differently affects its involvement in response inhibition in conditions of reactive control (the MC context; Jaspar et al. 2014). The difference in cerebral activity between IIs and NIs for the selected ROI (right IFop: x y z = 54 12 −2) in the MC context was extracted using a spherical 10-mm radius for each volunteer. A general linear model was used to perform PPI analyses. At the first level of the analyses (fixed effect), 3 regressors were created (without taking account of realignment parameters). The first regressor represented the interference effect (II–NI in the MC context). The second one was the activity in the seed area. The third regressor represented the interaction of interest between the first (psychological) and the second (physiological) regressors. The contrast images obtained allowed us to determine, in each subject, the brain areas interacting significantly with the right IFop in respect of the psychological regressor. The contrast images were used at the second level (random-effect analysis) for between-group comparisons. Again, these analyses were performed first by comparing each genotype with the other 2 separately and then by grouping together participants carrying at least one Val or Met allele, respectively. All consistent PPI results for our 2 analysis methods are presented with a threshold of P < 0.001, uncorrected. The extent threshold was set to 10 contiguous voxels. As performing PPI analyses with the input from a single ROI (here, the right IFop) cannot provide definite evidence of effective connectivity (Friston et al. 1997), the results obtained will be discussed in terms of functional connectivity (a correlation of activity between different regions).

Some previous studies reported a gender influence on behavioral Stroop performance (e.g., von Kluge 1992) and also the presence of sexually dimorphic effects of COMT on brain activation during inhibition (White et al. 2014). To exclude from our interpretations a potential sex influence, we also conducted our behavioral and fMRI analyses adding sex as a covariate. Anticipating on the next section, consideration of sex did not modify the results.

Results

Behavioral Results

We conducted a repeated 3 (context) × 3 (item) ANOVA on median RTs for correct responses, with group as an independent variable. Significant item (F2,84 = 207.73; P < 0.0001) and context (F2,84 = 20.99; P < 0.0001) effects were observed, but no significant group effect (F2,42 = 0.65; P = 0.53). Planned comparisons showed that the item effect is characterized by slower RTs for IIs than for NIs (F1,42 = 200.70; P < 0.0001). This interference effect is observed in MI (F1,42 = 232.37; P < 0.0001), MC (F1,42 = 102.23; P < 0.0001), and MN (F1,42 = 160.56; P < 0.0001) contexts.

As with RT, a 3 (context) × 3 (item) ANOVA on item accuracy, with group as an independent variable, was conducted. The pattern of results in terms of item and context effects is the same as for RT. Planned comparisons showed that the item effect is characterized by less accurate responses for IIs than for NIs (F1,42 = 98.29; P < 0.0001). Again, this interference effect is observed in MI (F1,42 = 60.42; P < 0.0001), MC (F1,42 = 19.18; P < 0.0001), and MN (F1,42 = 36.93; P < 0.0001) contexts. With regard to the genotype, no significant group effect was found for item accuracy (F1,42 = 1.51; P = 0.23). However, a significant item × group interaction (F4,84 = 2.77; P = 0.03) was observed. Indeed, significant differences in performance for IIs by comparison to NIs (II − NI) are observed in the MM group, when compared with the VM (F1,42 = 4.85; P = 0.03) and VV (F1,42 = 6.37; P = 0.02) groups, with the MM group performing better. Interestingly, the same pattern of results is also observed specifically in the MC context, with better performance by the MM than by the VM (F1,42 = 5.90; P = 0.02) and VV groups (F1,42 = 6.92; P = 0.01; see Supplementary Fig. 2 in Supplementary Materials).

Behavioral results associated with the MC context and requiring reactive control (the focus on the present report) can be summarized as follows. A significant interference effect was observed for RTs and accuracy. This effect is of similar amplitude in all 3 groups for RTs, while the MM group performed better in terms of accuracy. These effects were not modified when sex was used as a covariate in the analyses.

fMRI Results

As indicated in the Methods section, fMRI analyses were first performed by comparing each genotype group with the other 2 separately and then by grouping together participants carrying at least one Val or Met allele, respectively. In the text and the results tables, we will report only on the brain areas that were initially found in the first analysis (comparison of each genotype with the 2 others) and confirmed by the second one (comparisons of allelic groups). A complete description of the results obtained by these 2 approaches can be found in Supplementary Materials (see Supplementary Tables 2–7).

The Neural Substrates of the Interference Effect for the Task as a Whole

First of all, the general interference effect (i.e., IIs vs. NIs) in the whole sample of participants revealed a large map of activation corresponding to the extensive fronto-parieto-temporal network typically associated with interference resolution in the Stroop task (Jaspar et al. 2014).

Interestingly, when the transient pattern of cerebral activity for interfering items (compared with neutral ones) in the contexts that most induce the reading process (the MI and MC contexts, composed of 75% interfering or facilitator items) was compared between the VV, VM, and MM participants, we observed greater brain activity in the left superior temporal gyrus (STG) in VV and VM groups than in the MM group. This result was confirmed in the analysis conducted by grouping the Val allele carriers together and comparing them with the homozygous MM group (Table 1). These effects were not modified when sex was used as a covariate in the analyses.

Table 1

General interference effect—common features between comparisons by genotype and comparisons by allele

Hemisphere Anatomical region MNI coordinates
 
Cluster size Z-score P-value 
x y z 
Comparisons by genotype 
 VV > MM 
  Superior temporal gyrus −62 −54 16 3.36 <0.001 
 VM > MM 
  Superior temporal gyrus −60 −52 14 22 3.41 <0.001 
Comparisons by allele 
 VV and VM > MM 
  Superior temporal gyrus −62 −54 16 70 3.85 <0.0001 
−64 −44 16 70 3.48 <0.001 
Hemisphere Anatomical region MNI coordinates
 
Cluster size Z-score P-value 
x y z 
Comparisons by genotype 
 VV > MM 
  Superior temporal gyrus −62 −54 16 3.36 <0.001 
 VM > MM 
  Superior temporal gyrus −60 −52 14 22 3.41 <0.001 
Comparisons by allele 
 VV and VM > MM 
  Superior temporal gyrus −62 −54 16 70 3.85 <0.0001 
−64 −44 16 70 3.48 <0.001 

Note: Local maxima of brain regions showing more transient brain activity for the interference effect (IIs vs. NIs) during MC blocks at a voxel P-value of <0.001 uncorrected.

L/R, left or right; xyz, coordinates (mm) in the stereotactic space defined by the Montreal Neurological Institute (MNI).

The Neural Substrates of the Interference Effect in the Reactive Control Condition

The interference effect (i.e., IIs vs. NIs) in the whole sample of participants during the MC task context (i.e., when only 17% of interfering items were presented) also revealed a network of activation in fronto-parietal areas and increased activity in the insula and the cerebellum (Jaspar et al. 2014).

The transient activity in the whole brain for interfering items (compared with neutral ones) during the MC context was compared for our 3 groups of volunteers (VV, VM, and MM) and by grouping together participants carrying at least one Val or Met allele. Interestingly, we again observed increased cerebral activity in the left STG (Fig. 1A) in Val allele carriers by comparison to homozygous MM individuals (VV > MM; VM > MM; VV and VM > MM). Moreover, the left middle temporal gyrus (MTG), the right IFop, and the left precentral gyrus (PcG; Fig. 1B) appeared to be less activated in Met allele carriers (VV > VM; VV > MM; VV > VM and MM; Table 2). These effects were not modified when sex was used as a covariate in the analyses.

Table 2

Interference effect in the reactive control condition—common features between comparisons by genotype and comparisons by allele

Hemisphere Anatomical region MNI coordinates
 
Cluster size Z-score P-value 
x y z 
Comparisons by genotype 
 VV > MM 
  Inferior frontal operculum 60 −4 10 362 3.95 <0.0001 
  Precentral gyrus −62 14 10 3.38 <0.001 
  Middle temporal gyrus −54 −34 14 3.39 <0.001 
  Superior temporal gyrus −62 −46 18 68 3.66 <0.001 
 VV > VM 
  Inferior frontal operculum 54 16 −2 117 3.90 <0.0001 
  Precentral gyrus −62 16 11 3.46 <0.001 
  Middle temporal gyrus −54 −42 20 3.32 <0.001 
 VM > MM 
  Superior temporal gyrus −58 −56 12 3.27 <0.001 
Comparisons by allele 
 VV and VM > MM 
  Superior temporal gyrus −60 −56 12 264 3.64 <0.0001 
 VV > VM and MM 
  Inferior frontal operculum 56 10 −4 378 4.12 <0.0001 
  Precentral gyrus −62 16 95 3.58 <0.001 
  Middle temporal gyrus −54 −34 27 3.39 <0.001 
Hemisphere Anatomical region MNI coordinates
 
Cluster size Z-score P-value 
x y z 
Comparisons by genotype 
 VV > MM 
  Inferior frontal operculum 60 −4 10 362 3.95 <0.0001 
  Precentral gyrus −62 14 10 3.38 <0.001 
  Middle temporal gyrus −54 −34 14 3.39 <0.001 
  Superior temporal gyrus −62 −46 18 68 3.66 <0.001 
 VV > VM 
  Inferior frontal operculum 54 16 −2 117 3.90 <0.0001 
  Precentral gyrus −62 16 11 3.46 <0.001 
  Middle temporal gyrus −54 −42 20 3.32 <0.001 
 VM > MM 
  Superior temporal gyrus −58 −56 12 3.27 <0.001 
Comparisons by allele 
 VV and VM > MM 
  Superior temporal gyrus −60 −56 12 264 3.64 <0.0001 
 VV > VM and MM 
  Inferior frontal operculum 56 10 −4 378 4.12 <0.0001 
  Precentral gyrus −62 16 95 3.58 <0.001 
  Middle temporal gyrus −54 −34 27 3.39 <0.001 

Note: Local maxima of brain regions showing more transient brain activity for the interference effect (IIs vs. NIs) during MC blocks at a voxel P-value of <0.001 uncorrected.

L/R, left or right; xyz, coordinates (mm) in the stereotactic space defined by the Montreal Neurological Institute (MNI).

Figure 1.

Brain areas involved in interference resolution and differently affected by the COMT genotype during the MC context. (A) Brain area showing a larger difference in activity between IIs and NIs for the Val allele carriers compared with homozygous Met carriers during the MC contexts; (B) Brain area showing a smaller difference in activity between IIs and NIs for the Met allele carriers compared with homozygous Val carriers during MC contexts. The regions are displayed on the T1 canonical image implemented in SPM8. See Table 2 for coordinates. IFop, inferior frontal operculum; PcG, precentral gyrus; MTG, middle temporal gyrus; STG, superior temporal gyrus; MM, Met/Met participants; VM, Val/Met participants; VV, Val/Val participants.

Figure 1.

Brain areas involved in interference resolution and differently affected by the COMT genotype during the MC context. (A) Brain area showing a larger difference in activity between IIs and NIs for the Val allele carriers compared with homozygous Met carriers during the MC contexts; (B) Brain area showing a smaller difference in activity between IIs and NIs for the Met allele carriers compared with homozygous Val carriers during MC contexts. The regions are displayed on the T1 canonical image implemented in SPM8. See Table 2 for coordinates. IFop, inferior frontal operculum; PcG, precentral gyrus; MTG, middle temporal gyrus; STG, superior temporal gyrus; MM, Met/Met participants; VM, Val/Met participants; VV, Val/Val participants.

Psychophysiological Interactions in the Reactive Control Condition

As the right frontal operculum has frequently been associated with inhibitory processes and was more activated by VV individuals when reactive control was required (Jaspar et al. 2014), we compared the functional connectivity between that area and the rest of the brain in our 3 groups of participants.

We observed that individuals carrying at least one Val allele have consistently greater positive functional connectivity in the right IFop with the right cingulate gyrus and right STG than homozygous Met individuals (VV > MM, VM > MM, VV and VM > MM). We also observed that activity in the IFop of homozygous Val carriers was more associated with the right superior frontal gyrus (SFG), left mid-cingulate gyrus, left MTG extending to the STG, and left lingual gyrus than in homozygous and heterozygous carriers of the Met allele (VV > MM, VV > VM, VV > VM and MM; Table 3). Note that the MTG observed here overlaps the left MTG area reported for the interference effect in the reactive condition (Table 2). Interestingly, the right IFop is positively associated with the first 3 regions but negatively with the lingual gyrus (Fig. 2). Finally, no increased functional connectivity was observed between the right IFop and the rest of the brain for carriers of the Met allele by comparison to homozygous and heterozygous Val participants (MM > VV, MM > VM, VM > VV, MM > VV and VM, MM and VM > VV). Similar results were observed when sex was used as a covariate in the PPI analyses (see Supplementary Table 8).

Table 3

PPI analyses—common features between comparisons by genotype and comparisons by allele

Hemisphere Anatomical region MNI coordinates
 
Cluster size Z-score P-value 
x y z 
Comparisons by genotype 
 VV > MM 
  Superior frontal gyrus 10 12 56 13 3.41 <0.001 
  Cingulate gyrus 18 −6 40 23 3.73 <0.001 
  Mid-cingulum −8 −4 34 18 3.44 <0.001 
  Middle temporal gyrus −58 −40 236 3.61 <0.001 
  Superior temporal gyrus 52 −24 86 3.36 <0.001 
  Lingual gyrus −18 −84 −10 124 3.89 <0.0001 
 VV > VM 
  Superior frontal gyrus 12 56 12 3.24 <0.001 
  Mid-cingulum −12 34 15 3.24 <0.001 
  Middle temporal gyrus −56 −42 196 3.96 <0.0001 
−58 −32 196 3.31 <0.001 
  Lingual gyrus −32 −72 −4 100 3.52 <0.001 
 VM > MM 
  Cingulate gyrus 22 38 26 3.62 <0.001 
  Superior temporal gyrus 52 −24 11 3.68 <0.001 
Comparisons by allele 
 VV and VM > MM 
  Cingulate gyrus 20 −2 38 21 3.47 <0.0001 
  Superior temporal gyrus 52 −24 185 4.14 <0.0001 
 VV > VM and MM 
  Superior frontal gyrus 14 56 42 3.59 <0.001 
  Mid-cingulum −8 −4 34 27 3.51 <0.001 
  Middle temporal gyrus −58 −40 364 4.14 <0.0001 
−52 −32 364 3.71 <0.001 
  Lingual gyrus −30 −70 −6 246 3.75 <0.0001 
−22 −80 −10 82 3.56 <0.001 
Hemisphere Anatomical region MNI coordinates
 
Cluster size Z-score P-value 
x y z 
Comparisons by genotype 
 VV > MM 
  Superior frontal gyrus 10 12 56 13 3.41 <0.001 
  Cingulate gyrus 18 −6 40 23 3.73 <0.001 
  Mid-cingulum −8 −4 34 18 3.44 <0.001 
  Middle temporal gyrus −58 −40 236 3.61 <0.001 
  Superior temporal gyrus 52 −24 86 3.36 <0.001 
  Lingual gyrus −18 −84 −10 124 3.89 <0.0001 
 VV > VM 
  Superior frontal gyrus 12 56 12 3.24 <0.001 
  Mid-cingulum −12 34 15 3.24 <0.001 
  Middle temporal gyrus −56 −42 196 3.96 <0.0001 
−58 −32 196 3.31 <0.001 
  Lingual gyrus −32 −72 −4 100 3.52 <0.001 
 VM > MM 
  Cingulate gyrus 22 38 26 3.62 <0.001 
  Superior temporal gyrus 52 −24 11 3.68 <0.001 
Comparisons by allele 
 VV and VM > MM 
  Cingulate gyrus 20 −2 38 21 3.47 <0.0001 
  Superior temporal gyrus 52 −24 185 4.14 <0.0001 
 VV > VM and MM 
  Superior frontal gyrus 14 56 42 3.59 <0.001 
  Mid-cingulum −8 −4 34 27 3.51 <0.001 
  Middle temporal gyrus −58 −40 364 4.14 <0.0001 
−52 −32 364 3.71 <0.001 
  Lingual gyrus −30 −70 −6 246 3.75 <0.0001 
−22 −80 −10 82 3.56 <0.001 

Note: Local maxima of brain regions showing more transient brain activity for the interference effect (IIs vs. NIs) during MC blocks at a voxel P-value of <0.001 uncorrected.

L/R, left or right; xyz, coordinates (mm) in the stereotactic space defined by the Montreal Neurological Institute (MNI); Sup., superior.

Figure 2.

PPIs between the interference effect (II–NI) and the right IFop. Brain areas showing a significantly greater PPI with the right IFop in the homozygous Val individuals. The regions are displayed on the T1 canonical image implemented in SPM8. See Table 3 for coordinates. SFG, superior frontal gyrus; mid-cing., mid-cingulum; STG, superior temporal gyrus; ling. gy., lingual gyrus; MM, Met/Met participants; VM, Val/Met participants; VV, Val/Val participants.

Figure 2.

PPIs between the interference effect (II–NI) and the right IFop. Brain areas showing a significantly greater PPI with the right IFop in the homozygous Val individuals. The regions are displayed on the T1 canonical image implemented in SPM8. See Table 3 for coordinates. SFG, superior frontal gyrus; mid-cing., mid-cingulum; STG, superior temporal gyrus; ling. gy., lingual gyrus; MM, Met/Met participants; VM, Val/Met participants; VV, Val/Val participants.

Discussion

The objective of this study was to determine whether the COMT polymorphism's effect on inhibition previously observed in frontal areas (Blasi et al. 2005; Congdon et al. 2009; Jaspar et al. 2014) is also expressed in posterior brain areas, and whether this polymorphism modulates the functional connectivity between anterior and posterior brain areas.

From a behavioral point of view, it appeared that VV and VM participants were more sensitive to interference than MM participants. Indeed, these volunteers were less efficient at processing interfering events (compared with neutral ones), as revealed in their response accuracy. It is noteworthy that the decreased accuracy for incongruent events in the Val allele carrier groups was observed even though the groups of participants were matched for age, IQ, and other demographic factors. In parallel to these observations, we also showed that the same individuals (VV and VM) presented greater brain activity in the left STG when they had to resolve interference. We also reproduced our previous data showing an increase in activity in the right IFop in the VV group by comparison to the 2 other groups when interfering items were presented. Additionally, we observed that VV individuals presented higher brain activity in the left PcG, a region regularly involved in Stroop inhibitory tasks (Laird et al. 2005), and in the MTG. Now, with regard to functional connectivity analyses, we showed that activity in the right IFop is more positively linked to that in the right cingulate and superior temporal gyri in individuals carrying at least one Val allele (compared with homozygous Met individuals). Moreover, brain activity in the right IFop in homozygous Val carriers is more positively associated than in Met allele carriers with the right SFG, the left mid-cingulate, and middle temporal gyri, but more negatively with the lingual gyrus. Considering that the right IFop presented a higher level of activity in homozygous VV participants only (compared with the other 2 groups), we will focus our discussion of PPI differences on the comparison of VV individuals with Met allele carriers. Finally, as some studies reported evidence of sexually dimorphic effects of COMT on behavioral performance (Soeiro-De-Souza et al. 2013) and brain activity (White et al. 2014) during executive tasks, we replicated our analyses using sex as a covariate. These last analyses highlighted that genotypic differences in behavior and brain activity observed here are independent of any sex effect.

What Is the Role of Posterior Brain Regions Observed to be Differently Modulated by COMT Polymorphisms During the Stroop Task?

Two posterior areas appeared differently affected between COMT allelic groups during the Stroop task. First, participants with at least one Val allele exhibited less deactivation in the left STG for interfering items than for NIs. Secondly, the homozygous Val group presented greater activity in the left MTG for interfering than for neutral stimuli. As it is also commonly accepted that language abilities, and specifically reading processes, are subserved by temporal areas (Jobard et al. 2003), the involvement of these 2 regions could plausibly be related to the reading induced by the Stroop task. More specifically, based on the dual-route model of reading (Coltheart et al. 1993), the left MTG and STG may be related to 2 different aspects of word reading processes. Indeed, this model postulates that, after preliminary visual analyses, words can be read by 2 different routes. The direct one, the lexico-semantic route, allows a direct association between the visual form of the word and its meaning. The indirect, or grapho-phonological, route involves a grapheme-to-phoneme conversion to transform the word to its auditory form and access its meaning. A meta-analysis by Jobard et al. (2003) associated activity in the left STG to the indirect route while the left MTG seems to be involved in both reading routes.

The part of the left STG that we observed to be directly affected by the COMT genotype was specifically associated with the phonological analysis of words, a process that is based on sublexical mechanisms (Simos et al. 2000). Considering the very frequent occurrence in French of the words used during the task (red, green, black, and blue), differences in regions associated with this indirect reading route were not expected. However, these differences represent changes in deactivation patterns driven mainly by the processing of NIs. So, COMT-related changes in activity in that area do not seem to be specifically related to the Stroop effect and will not be discussed further. On the other hand, the left MTG appeared to be very close to the cerebral network associated with semantic access processes in the direct reading route (Fiebach et al. 2002; Jobard et al. 2003). Consequently, this region might be considered as part of the brain network that must be inhibited to perform the Stroop task efficiently.

How Can the Impact of COMT Genotype on Posterior Brain Areas and Their Functional Connectivity with the Right IFop in VV Participants Be Explained and Interpreted?

In a previous study, we showed that COMT genotypes had different effects on transient activity in anterior brain areas (VV > VM and MM in the right IFop and SFG; VV and VM > MM in the right IFop and cingulate areas) for interfering events in reactive control contexts (Jaspar et al. 2014). Here, we also show that posterior brain regions, especially temporal areas, are also affected by COMT Val158Met polymorphism (VV > VM and MM in the left MTG; VV and VM > MM in the left STG). These different patterns of brain activation were obtained with analyses focusing on accurately processed stimuli. Thus, all group discrepancies observed and discussed at the brain activation level can be considered to be compensatory mechanisms set-up by homozygous VV individuals to perform the task efficiently. In addition, we assume that these cerebral compensatory mechanisms can also be expressed by changes in brain functional connectivity (Cabeza and Dennis 2012).

The observation of an influence of COMT polymorphism on temporal areas seems surprising at first glance. Indeed, the action of the COMT gene is mainly characterized by the degradation of DA in the frontal cortex (Karoum et al. 1994), but not really in posterior brain regions. On the basis of our psychophysiological analyses, we hypothesized that dopaminergic-mediated gating signals (influencing frontal activity and affected by the COMT genotype) would impact the temporal areas associated with word processing via the right IFop. Indeed, we observed stronger positive functional connectivity between the right IFop and the left MTG in the homozygous Val carriers during the processing of interfering items. This pattern of results indicates that these participants recruit more of an area classically associated with interference resolution, the right IFop, than Met allele carriers (Garavan et al. 1999, 2002; Konishi et al. 1999; Rubia et al. 2003), when an area associated with reading processes, the left MTG (Simos et al. 2000), is also highly activated. Interestingly, this observation is coupled with a stronger negative interaction between the right IFop and the lingual gyrus for the same individuals; the latter region is involved in color identification (Polk et al. 2008). Although these patterns of PPIs cannot be directly interpreted in terms of causal effects by one area on another, we assumed that the lower DA level in the PFC for the homozygous VV group requires more involvement in PFC areas (right IFop and left PcG) to perform the task efficiently. As the right IFop is considered to have a braking function that can be turned on to pause an automatic response or to stop it outright (Aron et al. 2014), we propose that the stronger positive relationship observed between the right IFop and the left MTG reflects the right IFop's impact on temporal areas associated with reading; that influence may be direct or indirect. On the other hand, the negative relationship with the lingual gyrus cannot arise from a direct influence between these 2 areas if we consider the braking function attributed to the right IFop. Indeed, cognitive processes specific to the Stroop task and supported by the lingual gyrus (i.e., color processing) are supposed to be enhanced, and not inhibited, during task performance. So, we can suppose that the interaction between the right IFop and left lingual gyrus is mediated by a third (anterior or posterior) area that was not directly revealed by our PPI analysis.

As a whole, the results of this study indicate that the balance between the implementation of word reading and color naming processes to perform the Stroop task correctly is specifically associated with activity in the right IFop in homozygous Val carriers. As activity in that area was previously associated with a less-efficient physiological response in that population, this pattern of results can be interpreted as indicating less-efficient regulation of processes that control color naming and word reading. In agreement with this control hypothesis, activity in the right IFop is also positively associated with that in the right superior frontal area. The right SFG is an area that has classically been linked to the Stroop effect (Laird et al. 2005) and is known to be a key area for rapid inhibitory control (Floden and Stuss 2006). However, given our experimental design, the exact interrelationships between these anterior and posterior brain areas according to available DA level in frontal areas cannot be determined. It would be particularly interesting to investigate this question using dynamic causal modeling (DCM) analyses. Such analyses would allow a direct assessment of how the coupling parameters of the cerebral network observed here (right IFop, right STG, left MTG, and right lingual gyrus) are modulated by the COMT genotype (Kahan and Foltynie 2013). In that context, the neural network model of dual control mechanisms developed by De Pisapia and Braver (2006) seems a particularly relevant starting point to assess the (direct and indirect) excitatory and inhibitory influences of these 4 areas on each other, and how it depends of DA availability.

The main outcome of this study is the presence, during an inhibitory task, of an influence of COMT polymorphism on brain circuitry, as assessed by changes in anterior–posterior functional connectivity, and not just on PFC. Similarly, 2 other studies showed decreased functional connectivity at rest between prefrontal regions and the posterior cingulate/retrosplenial cortices in homozygous Val individuals (Liu et al. 2010; Tian et al. 2013). Moreover, in the memory domain, Bertolino et al. (2006) showed that the number of Met alleles predicted the strength of relationship between the hippocampal formation and the ventrolateral PFC during retrieval and (to a lesser extent) encoding of information. So, the influence of DA degradation in frontal areas on posterior brain networks could be a relatively general mechanism. Consequently, the exploration of the functional (and also structural) connectivity between anterior and posterior brain areas deserves a special attention in populations with DA-depleted states, especially when a disconnection process was proposed as an explanation to cognitive deficits [i.e., in normal aging (Sullivan et al. 2001) or in schizophrenia (Friston 1999)].

Conclusion

In this study, we first demonstrated that variations in DA signaling mediated by COMT Val158Met polymorphism influence anterior and posterior brain areas recruited to resolve interference. Indeed, homozygous VV individuals recruited more regions associated with executive functioning (right IFop and left PcG) but also with non-executive functioning (left MTG and STG) than homozygous MM individuals when performing the Stroop task. In addition, PPI analyses demonstrated that COMT genotype also impacts functional connectivity between the right IFop (whose the function is braking automatic responses) and posterior brain areas associated with color processing and reading. These genotype-related changes in cerebral activity and brain functional connectivity can be considered to be compensatory mechanisms developed by homozygous VV individuals so they can efficiently perform the task. Nevertheless, further studies are necessary to really understand the functional significance and generality of changes in brain connectivity evidenced here with an inhibitory task. The exploration of large samples of healthy participants using DCM analyses as well as functional and structural brain connectivity analyses in populations with DA-depleted states appears particularly relevant to answer these questions.

Supplementary Material

Supplementary material can be found at: http://www.cercor.oxfordjournals.org/

Funding

This research was supported by the Belgian National Fund for Scientific Research (FRS-FNRS to P.M. and F.C.), the University of Liège, the Léon Fredericq Foundation, the InterUniversity Attraction Pole (IUAP P7/11), the French Speaking Community Concerted Research Action (ARC 09/14-03), the Walloon Excellence in Life Sciences and Biotechnology (Welbio) program, and the Baron-Clerdent Foundation. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors declare no competing financial interests.

Notes

Conflict of Interest: None declared.

References

Aron
AR
Poldrack
RA
.
2006
.
Cortical and subcortical contributions to Stop signal response inhibition: role of the subthalamic nucleus
.
J Neurosci
 .
26
:
2424
2433
.
Aron
AR
Robbins
TW
Poldrack
RA
.
2014
.
Inhibition and the right inferior frontal cortex: one decade on
.
Trends Cogn Sci
 .
18
:
1777
1785
.
Aron
AR
Robbins
TW
Poldrack
RA
.
2004
.
Inhibition and the right inferior frontal cortex
.
Trends Cogn Sci
 .
8
:
170
177
.
Banich
MT
Milham
MP
Atchley
R
Cohen
NJ
Webb
A
Wszalek
T
Kramer
AF
Liang
ZP
Wright
A
Shenker
J
et al
2000
.
fMRI studies of Stroop tasks reveal unique roles of anterior and posterior brain systems in attentional selection
.
J Cogn Neurosci
 .
12
:
988
1000
.
Bench
CJ
Frith
CD
Grasby
PM
Friston
KJ
Paulesu
E
Frackowiak
RS
Dolan
RJ
.
1993
.
Investigations of the functional anatomy of attention using the Stroop test
.
Neuropsychologia
 .
31
:
907
922
.
Bertolino
A
Rubino
V
Sambataro
F
Blasi
G
Latorre
V
Fazio
L
Caforio
G
Petruzzella
V
Kolachana
B
Hariri
A
et al
2006
.
Prefrontal-hippocampal coupling during memory processing is modulated by COMT Val158Met genotype
.
Biol Psychiatry
 .
60
:
1250
1258
.
Blasi
G
Mattay
VS
Bertolino
A
Elvevag
B
Callicott
JH
Das
S
Kolachana
BS
Egan
MF
Goldberg
TE
Weinberger
DR
.
2005
.
Effect of catechol-O-methyltransferase Val158Met genotype on attentional control
.
J Neurosci
 .
25
:
5038
5045
.
Botvinick
MM
Braver
TS
Barch
DM
Carter
CS
Cohen
JD
.
2001
.
Conflict monitoring and cognitive control
.
Psychol Rev
 .
108
:
624
652
.
Botvinick
MM
Cohen
JD
Carter
CS
.
2004
.
Conflict monitoring and anterior cingulate cortex: an update
.
Trends Cogn Sci
 .
8
:
539
546
.
Braver
T
Gray
J
Burgess
GC
.
2007
.
Explaining the many varieties of working memory variation: dual mechanisms of cognitive control
. In:
Conway
AR
Jarrold
C
Kane
MJ
Miyake
A
Towse
JN
, editors.
Variation in working memory
 .
New-York
:
Oxford University Press
. p.
76
106
.
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
.
Cabeza
R
Dennis
NA
.
2012
.
Frontal lobes and aging
. In:
Stuss
DT
Knight
RT
, editors.
Principles of frontal lobe function
 .
2nd ed
.
New York
:
Oxford University Press
. p.
628
652
.
Carter
CS
Braver
TS
Barch
DM
Botvinick
MM
Noll
D
Cohen
JD
.
1998
.
Anterior cingulate cortex, error detection, and the online monitoring of performance
.
Science
 .
280
:
747
749
.
Chee
MW
Sriram
N
Soon
CS
Lee
KM
.
2000
.
Dorsolateral prefrontal cortex and the implicit association of concepts and attributes
.
Neuroreport
 .
11
:
135
140
.
Chen
J
Lipska
BK
Halim
N
Ma
QD
Matsumoto
M
Melhem
S
Kolachana
BS
Hyde
TM
Herman
MM
Apud
J
et al
2004
.
Functional analysis of genetic variation in catechol-O-methyltransferase (COMT): effects on mRNA, protein, and enzyme activity in postmortem human brain
.
Am J Hum Genet
 .
75
:
807
821
.
Collette
F
Van der Linden
M
Delfiore
G
Degueldre
C
Luxen
A
Salmon
E
.
2001
.
The functional anatomy of inhibition processes investigated with the Hayling task
.
Neuroimage
 .
14
:
258
267
.
Coltheart
M
Curtis
B
Atkins
P
Haller
M
.
1993
.
Models of reading aloud: dual-route and parallel-distributed-processing approaches
.
Psychol Review
 .
100
:
589
608
.
Congdon
E
Constable
RT
Lesch
KP
Canli
T
.
2009
.
Influence of SLC6A3 and COMT variation on neural activation during response inhibition
.
Biol Psychol
 .
81
:
144
152
.
Deichmann
R
Schwarzbauer
C
Turner
R
.
2004
.
Optimisation of the 3D MDEFT sequence for anatomical brain imaging: technical implications at 1.5 and 3T
.
Neuroimage
 .
21
:
757
767
.
De Pisapia
N
Braver
T
.
2006
.
A model of dual control mechanisms through anterior cingulate and prefrontal cortex interactions
.
Neurocomputing
 .
69
:
1322
1326
.
Egner
T
Hirsch
J
.
2005
.
The neural correlates and functional integration of cognitive control in a Stroop task
.
Neuroimage
 .
24
:
539
547
.
Ettinger
U
Kumari
V
Collier
DA
Powell
J
Luzi
S
Michel
TM
Zedomi
O
Williams
SC
.
2008
.
Catechol-O-methyltransferase (COMT) Val158Met genotype is associated with BOLD response as a function of task characteristic
.
Neuropsychopharmacology
 .
33
:
3046
3057
.
Fiebach
CJ
Friederici
AD
Muller
K
von Cramon
DY
.
2002
.
fMRI evidence for dual routes to the mental lexicon in visual word recognition
.
J Cogn Neurosci
 .
14
:
11
23
.
Floden
D
Stuss
DT
.
2006
.
Inhibitory control is slowed in patients with right superior medial frontal damage
.
J Cogn Neurosci
 .
18
:
1843
1849
.
Friston
KJ
.
1999
.
Schizophrenia and the disconnection hypothesis
.
Acta Psychiatr Scand Suppl
 .
395
:
68
79
.
Friston
KJ
Buechel
C
Fink
GR
Morris
J
Rolls
E
Dolan
RJ
.
1997
.
Psychophysiological and modulatory interactions in neuroimaging
.
Neuroimage
 .
6
:
218
229
.
Garavan
H
Ross
TJ
Murphy
K
Roche
RA
Stein
EA
.
2002
.
Dissociable executive functions in the dynamic control of behavior: inhibition, error detection, and correction
.
Neuroimage
 .
17
:
1820
1829
.
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
.
Grandjean
J
D'Ostilio
K
Phillips
C
Balteau
E
Degueldre
C
Luxen
A
Maquet
P
Salmon
E
Collette
F
.
2012
.
Modulation of brain activity during a Stroop inhibitory task by the kind of cognitive control required
.
PLoS ONE
 .
7
:
e41513
.
Jaspar
M
Genon
S
Muto
V
Meyer
C
Manard
M
Dideberg
V
Bours
V
Salmon
E
Maquet
P
Collette
F
.
2014
.
Modulating effect of COMT genotype on the brain regions underlying proactive control process during inhibition
.
Cortex
 .
50
:
148
161
.
Jobard
G
Crivello
F
Tzourio-Mazoyer
N
.
2003
.
Evaluation of the dual route theory of reading: a metanalysis of 35 neuroimaging studies
.
Neuroimage
 .
20
:
693
712
.
Kahan
J
Foltynie
T
.
2013
.
Understanding DCM: ten simple rules for the clinician
.
Neuroimage
 .
83
:
542
549
.
Karoum
F
Chrapusta
SJ
Egan
MF
.
1994
.
3-Methoxytyramine is the major metabolite of released dopamine in the rat frontal cortex: reassessment of the effects of antipsychotics on the dynamics of dopamine release and metabolism in the frontal cortex, nucleus accumbens, and striatum by a simple two pool model
.
J Neurochem
 .
63
:
972
979
.
Kerns
JG
Cohen
JD
MacDonald
AW
III
Cho
RY
Stenger
VA
Carter
CS
.
2004
.
Anterior cingulate conflict monitoring and adjustments in control
.
Science
 .
303
:
1023
1026
.
Kimberg
DY
D'Esposito
M
Farah
MJ
.
1997
.
Effects of bromocriptine on human subjects depend on working memory capacity
.
Neuroreport
 .
8
:
3581
3585
.
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
.
Lachman
HM
Papolos
DF
Saito
T
Yu
YM
Szumlanski
CL
Weinshilboum
RM
.
1996
.
Human catechol-O-methyltransferase pharmacogenetics: description of a functional polymorphism and its potential application to neuropsychiatric disorders
.
Pharmacogenetics
 .
6
:
243
250
.
Laird
AR
McMillan
KM
Lancaster
JL
Kochunov
P
Turkeltaub
PE
Pardo
JV
Fox
PT
.
2005
.
A comparison of label-based review and ALE meta-analysis in the Stroop task
.
Hum Brain Mapp
 .
25
:
6
21
.
Larrue
V
Celsis
P
Bes
A
Marc-Vergnes
JP
.
1994
.
The functional anatomy of attention in humans: cerebral blood flow changes induced by reading, naming, and the Stroop effect
.
J Cereb Blood Flow Metab
 .
14
:
958
962
.
Leung
HC
Skudlarski
P
Gatenby
JC
Peterson
BS
Gore
JC
.
2000
.
An event-related functional MRI study of the Stroop color word interference task
.
Cereb Cortex
 .
10
:
552
560
.
Li
CS
Huang
C
Constable
RT
Sinha
R
.
2006
.
Imaging response inhibition in a stop-signal task: neural correlates independent of signal monitoring and post-response processing
.
J Neurosci
 .
26
:
186
192
.
Liu
B
Song
M
Li
J
Liu
Y
Li
K
Yu
C
Jiang
T
.
2010
.
Prefrontal-related functional connectivities within the default network are modulated by COMT Val158Met in healthy young adults
.
J Neurosci
 .
30
:
64
69
.
Lotta
T
Vidgren
J
Tilgmann
C
Ulmanen
I
Melen
K
Julkunen
I
Taskinen
J
.
1995
.
Kinetics of human soluble and membrane-bound catechol-O-methyltransferase: a revised mechanism and description of the thermolabile variant of the enzyme
.
Biochemistry
 .
34
:
4202
4210
.
Mannisto
PT
Kaakkola
S
.
1999
.
Catechol-O-methyltransferase (COMT): biochemistry, molecular biology, pharmacology, and clinical efficacy of the new selective COMT inhibitors
.
Pharmacol Rev
 .
51
:
593
628
.
Mehta
MA
Calloway
P
Sahakian
BJ
.
2000
.
Amelioration of specific working memory deficits by methylphenidate in a case of adult attention deficit/hyperactivity disorder
.
J Psychopharmacol
 .
14
:
299
302
.
Milham
MP
Erickson
KI
Banich
MT
Kramer
AF
Webb
A
Wszalek
T
Cohen
NJ
.
2002
.
Attentional control in the aging brain: insights from an fMRI study of the Stroop task
.
Brain Cogn
 .
49
:
277
296
.
Nee
DE
Wager
TD
Jonides
J
.
2007
.
Interference resolution: insights from a meta-analysis of neuroimaging tasks
.
Cogn Affect Behav Neurosci
 .
7
:
1
17
.
Nigg
JT
.
2000
.
On inhibition/disinhibition in developmental psychopathology: views from cognitive and personality psychology and a working inhibition taxonomy
.
Psychol Bull
 .
126
:
220
246
.
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
.
Plewnia
C
Zwissler
B
Langst
I
Maurer
B
Giel
K
Kruger
R
.
2013
.
Effects of transcranial direct current stimulation (tDCS) on executive functions: influence of COMT Val/Met polymorphism
.
Cortex
 .
49
:
1801
1807
.
Polk
TA
Drake
RM
Jonides
JJ
Smith
MR
Smith
EE
.
2008
.
Attention enhances the neural processing of relevant features and suppresses the processing of irrelevant features in humans: a functional magnetic resonance imaging study of the Stroop task
.
J Neurosci
 .
28
:
13786
13792
.
Raven
JC
Court
JH
Raven
J
.
1983
.
Manual for Raven's progressive matrices and vocabulary scales: advanced progressive matrices sets I and II
 .
London
:
H.K. Lewis
.
Rubia
K
Smith
AB
Brammer
MJ
Taylor
E
.
2003
.
Right inferior prefrontal cortex mediates response inhibition while mesial prefrontal cortex is responsible for error detection
.
Neuroimage
 .
20
:
351
358
.
Ruff
CC
Woodward
TS
Laurens
KR
Liddle
PF
.
2001
.
The role of the anterior cingulate cortex in conflict processing: evidence from reverse Stroop interference
.
Neuroimage
 .
14
:
1150
1158
.
Simos
PG
Breier
JI
Wheless
JW
Maggio
WW
Fletcher
JM
Castillo
EM
Papanicolaou
AC
.
2000
.
Brain mechanisms for reading: the role of the superior temporal gyrus in word and pseudoword naming
.
Neuroreport
 .
11
:
2443
2447
.
Soeiro-De-Souza
MG
Bio
DS
David
DP
Missio
G
Lima
B
Fernandes
F
Machado-Vieira
R
Moreno
RA
.
2013
.
Gender effects of the COMT Val158Met genotype on verbal fluency in healthy adults
.
Mol Med Rep
 .
8
:
837
844
.
Stokes
PR
Rhodes
RA
Grasby
PM
Mehta
MA
.
2011
.
The effects of the COMT Val108/158Met polymorphism on BOLD activation during working memory, planning, and response inhibition: a role for the posterior cingulate cortex?
Neuropsychopharmacology
 .
36
:
763
771
.
Stroop
J
.
1935
.
Studies of interference in serial verbal reactions
.
J Exp Psychol
 .
18
:
643
662
.
Sullivan
M
Jones
DK
Summers
PE
Morris
RG
Williams
SC
Markus
HS
.
2001
.
Evidence for cortical "disconnection" as a mechanism of age-related cognitive decline
.
Neurology
 .
57
:
632
638
.
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
.
Tian
T
Qin
W
Liu
B
Wang
D
Wang
J
Jiang
T
Yu
C
.
2013
.
Catechol-O-methyltransferase Val158Met polymorphism modulates gray matter volume and functional connectivity of the default mode network
.
PLoS ONE
 .
8
:
e786997
.
Von Kluge
S
.
1992
.
Trading accuracy for speed: gender differences on a Stroop task under mild performance anxiety
.
Percept Mot Skills
 .
75
:
651
657
.
White
TP
Loth
E
Krabbendam
L
Rubia
K
Whelan
R
Banaschewski
T
Barker
GJ
Bokde
AL
Büchel
C
Conrod
P
et al
2014
.
Sex differences in COMT polymorphism effects on prefrontal inhibitory control in adolescence
.
Neuropsychopharmacology
 . .
Witte
AV
Floel
A
.
2012
.
Effects of COMT polymorphisms on brain function and behavior in health and disease
.
Brain Res Bull
 .
88
:
418
428
.