Reward Reduces Conflict by Enhancing Attentional Controland Biasing Visual Cortical Processing

Srikanth Padmala1 and Luiz Pessoa2

Abstract

■ How does motivation interact with cognitive control duringchallenging behavioral conditions? Here, we investigated the in-teractions between motivation and cognition during a responseconflict task and tested a specific model of the effect of rewardon cognitive processing. Behaviorally, participants exhibited re-duced conflict during the reward versus no-reward condition. Brainimaging results revealed that a group of subcortical and fronto-parietal regions was robustly influenced by reward at cue process-ing and, importantly, that cue-related responses in fronto-parietalattentional regions were predictive of reduced conflict-related sig-nals in the medial pFC (MPFC)/ACC during the upcoming targetphase. Path analysis revealed that the relationship between cue

responses in the right intraparietal sulcus (IPS) and interference-related responses in the MPFC during the subsequent targetphase was mediated via signals in the left fusiform gyrus, whichwe linked to distractor-related processing. Finally, reward increasedfunctional connectivity between the right IPS and both bilateralputamen and bilateral nucleus accumbens during the cue phase,a relationship that covaried with across-individual sensitivity to re-ward in the case of the right nucleus accumbens. Taken together,our findings are consistent with a model in which motivationallysalient cues are employed to upregulate top–down control pro-cesses that bias the selection of visual information, thereby leadingtomore efficient stimulus processing during conflict conditions. ■

INTRODUCTION

Cognition interacts with motivation in important ways, anda significant amount of research in the past decade has at-tempted to elucidate the neural bases of these interactions(Pessoa, 2009; Braver, Gray, & Burgess, 2007; Watanabe,2002). Whereas it is clear that motivation affects cognitivefunction, the mechanisms by which cognition is altered re-main poorly understood. One possibility is that the effect isrelatively unspecific, such as an “energizing” function thatspeeds up performance. Contrariwise, motivation mayhave more specific effects on cognition, for instance, byenhancing executive function. In the latter scenario, moti-vation would be expected, for instance, to influence theselection of information pertinent to the task at hand.To probe this question, we investigated the effects of

reward during a response–conflict task (Figure 1A). Onthe basis of recent findings (Pessoa & Engelmann, 2010;Engelmann, Damaraju, Padmala, & Pessoa, 2009), weanticipated that motivation would enhance processing inattentional regions in fronto-parietal cortex. We reasonedthat these regions would then be better positioned toexert top–down control that favored the processing oftask-relevant information in visual cortex during the targetphase, for instance, by amplifying task-relevant informa-tion (Egner & Hirsch, 2005) and/or by improving filteringof task-irrelevant information (Polk, Drake, Jonides, Smith,

& Smith, 2008). Thus, behavioral conflict would be re-duced during the reward versus no-reward condition; like-wise, conflict-related brain responses inmedial pFC (MPFC;Carter et al., 1998) would be expected to be reduced. Acentral goal of the present study was to understand the linkbetween the processing of cues signaling potential rewardand interference-related activity during the subsequent tar-get phase (see Figure 1A). Critically, we sought to advanceour understanding of the effects of motivation on cogni-tion by evaluating potential network interactions thatmight have subserved these effects. Specifically, we em-ployed both mediation analysis and functional connectivityanalysis to test the model outlined in Figure 2 that sum-marizes the interactions hypothesized in the present study.

METHODS

Subjects

Fifty-four volunteers (22 ± 5 years old, 28 women) partici-pated in the study, which was approved by the institutionalreview board of Indiana University, Bloomington. All sub-jects were in good health with no past history of psychiat-ric or neurological disease. All participants had normal orcorrected-to-normal vision. All participants gave informedwritten consent. One male participantʼs data were excludedfrom the analysis because of excessive headmotion (greaterthan one voxel size), and data from three other participants(one man and two women) were removed because of poorperformance (less than 60% correct; two of the participants1Indiana University, 2University of Maryland

© 2011 Massachusetts Institute of Technology Journal of Cognitive Neuroscience 23:11, pp. 3419–3432

reported being unable to concentrate because of fatigue;the other responded only during reward trials).

Personality Questionnaire

Before the fMRI experiment, participants completed theBehavioral Activation System (BAS) scale, which assessesmultiple personality characteristics related to sensitivityto reward (Carver & White, 1994). As in our recent study(Engelmann et al., 2009), we employed the BAS-drive sub-

scale, which has the highest internal reliability ( Jorm et al.,1998; Carver & White, 1994), has been suggested to bethe strongest predictor of positive affective responses toreward (Beaver et al., 2006) and has been proposed to pro-vide a clearer measure of appetitive motivation and ap-proach behavior (Dawe, Gullo, & Loxton, 2004).

Stimuli and Behavioral Paradigm

Compound scene-plus-word stimuli were employed (Fig-ure 1A). These categories were chosen because they areassociated with category-related responses in visual cortex.Specifically, scenes strongly recruit the parahippocampalgyrus, bilaterally (Epstein, Harris, Stanley, & Kanwisher,1999), whereas words robustly recruit the left fusiformgyrus (McCandliss, Cohen, & Dehaene, 2003; Polk &Farah, 2002). Accordingly, images of houses and buildingsoverlaid with the words “HOUSE,” “BLDNG” were used tocreate congruent and incongruent trials; the letter string“XXXXX” was used to create a neutral trial type. Each trialbegun with a cue indicating the motivational condition(“$00” or “$20“) shown for 750 msec. Following a 2- to6-sec jittered interstimulus interval, a composite scene-plus-word target image was shown for 1000 msec. Partici-pants were instructed to press the index finger button toindicate that they saw a house image and the middle fin-ger button to indicate that they saw a building image ir-respective of the overlaid word (responses were alwaysmade with the right hand; the response button mappingwas counterbalanced across participants). Thus, imagestimuli were task relevant, and word stimuli were task ir-relevant. After 200 msec from the offset of the target, par-ticipants received visual feedback for 800 msec consistingof the total number of points won on the trial, as well astheir cumulative earnings (in points) until that moment

Figure 1. Response–conflictparadigm and behavioraldata. Subjects performed aresponse conflict task undertwo motivational contexts:(A) during the rewardcondition (shown here), acue stimulus (“$20”) signaledthat participants would berewarded for fast and correctperformance; during thecontrol condition (not shownhere), a cue stimulus (“$00”)signaled that no rewardwas involved. Following avariable-length delay, a targetstimulus containing a pictureof a house or building wasshown together with atask-irrelevant word (anincongruent condition isillustrated here). After the target stimulus, subjects were informed about the potential reward and about the total number of points accrued. Finally,a variable-length delay ended the trial. (B) Plot of interference (incongruent vs. neutral) and facilitation (congruent vs. neutral) RT scoresillustrating the interaction patterns. C = congruent trials; I = incongruent trials; N = neutral trials. Error bars represent the SEMs.

Figure 2. Hypothesized network interactions. (A) The relationshipbetween attentional control implemented in fronto-parietal cortexduring the cue phase and conflict-related activity in MPFC duringthe subsequent target phase was hypothesized to be mediated viathe amount of target/distractor processing in visual cortex. (B) Wealso hypothesized that the functional coupling between fronto-parietalcortex and subcortical regions involved in reward processing wouldbe affected by motivational context.

3420 Journal of Cognitive Neuroscience Volume 23, Number 11

in time (so as to minimize the “cognitive load” related tokeeping track of earnings). During reward trials, partici-pants won 2000 points per trial if they were fast and accu-rate in their response and they won zero points duringerror or slow trials. The RT threshold to determine “fast”responses was set at 800 msec based on behavioral pilotdata. During no-reward trials, participants won zero pointsirrespective of their performance. During correct trials,points won were displayed in white color, and during errortrials, points were displayed in blue color, thus providingperformance feedback for both motivational conditions.Finally, a 2- to 6-sec intertrial interval containing a blankscreen terminated the trial.Both interstimulus and intertrial intervals were selected

from an exponential distribution favoring shorter intervalsand helped in the estimation of separate cue- and target-related responses (see below). Before the start of the ex-periment, participants were informed that they could earn2000 points during each reward trial if performance wasfast and accurate and that at the end of the experimentthe points would be converted to cash, such that they couldearn an additional $20 based on their performance. At theend of the experiment, each participantʼs base pay ($25)was potentially increased such that for each point theyearned 0.01 cents (average reward-based earning = $18).We used the Presentation software (Neurobehavioral

Systems, Albany, CA) for the presentation of visual stimuli,and responses were recorded using an MRI-compatibleresponse box inside the scanner room.Each participant performed six runs of the conflict

task. Each run consisted of 36 trials, resulting in a totalof 216 trials and 36 trials per condition. Trial order wasbalanced, such that each trial type was preceded by theother trial types with the same frequency. For each con-dition, trials were equally divided between those contain-ing house and building scenes, and different images ofhouses and buildings were used during the control andreward conditions (counterbalanced across participants).Within condition, each image was used only once duringcongruent, incongruent, or neutral trial types. No sceneimages appeared in consecutive trials to minimize poten-tial priming effects (Mayr, Awh, & Laurey, 2003). Finally,each run ended with a 10-sec fixation cross allowing usto record the hemodynamic response for the final trial ofthe run.

Functional Localizer

At the end of the main experimental runs, subjects par-ticipated in an additional functional “localizer” run duringwhich they performed a simple 1-back working memorytask administered in a alternating blocked fashion andcontaining novel word and scene stimuli. Five blocks wereperformed per condition, each with 12 trials during whicha neutral word or scene (house or building) stimulus waspresented for 1000 msec and followed by a 250-msec

blank screen. Blocks lasted 15 sec and were separated bya 15-sec rest block during which participants passivelyviewed a white fixation cross on the screen.

MR Data Acquisition

MRI data were collected using a 3-T Siemens TRIO scanner(Siemens Medical Systems, Erlangen, Germany) witha 32-channel head coil (without parallel imaging). Eachscanning session began with a high-resolution MPRAGEanatomical scan (TR = 1900 msec, TE = 4.15 msec,TI = 1100 msec, 1-mm isotropic voxels, field of view =256 mm). Subsequently, in each functional run of themain experiment, 165 EPI volumes were acquired with aTR of 2500 and TE of 25 msec. Each volume consisted of44 oblique slices with a thickness of 3 mm and an in-planeresolution of 3 × 3 mm (field of view = 192 mm). Sliceswere positioned approximately 30° relative to the planedefined by the line connecting the anterior and poste-rior commissures. For the final functional localizer run,123 EPI volumes were collected with the same scanningparameters.

Behavioral Data Analysis

Mean RT data (excluding condition-specific extreme pointsthat were situated more than 3 SDs from the mean; 3.4%,on average) and mean accuracy data were initially analyzedusing a 2 Motivation (reward, no-reward) × 3 Congruency(neutral, congruent, incongruent) repeated measuresANOVA. Additional 2 × 2 repeatedmeasures ANOVAs wereevaluated to further characterize the results.

General fMRI Data Analysis

Preprocessing of the data was done using tools from theAFNI software package (Cox, 1996; afni.nimh.nih.gov/afni).The first three volumes of each functional run were dis-carded to account for equilibration effects. The remainingvolumes were slice time corrected using Fourier interpola-tion, such that all slices were realigned to the first slice toaccount for the timing offset between slices. Six-parameterrigid body motion correction within and across runs was per-formed using Fourier interpolation (Cox & Jesmanowicz,1999), such that all volumes were spatially registered to thevolume acquired closest in time to a particular subjectʼshigh-resolution anatomy. To normalize the functional datato Talairach space (Talairach & Tournoux, 1988), initiallyeach subjectʼs high-resolution MPRAGE anatomical volumewas spatially registered to the so-called TT_N27 template(in Talairach space) using a 12-parameter affine transfor-mation; the same transformation was then applied to thefunctional data. All volumes were spatially smoothed usinga Gaussian filter with an FWHM of 6 mm (i.e., twice thevoxel dimension). Finally, the signal intensity of each voxel

Padmala and Pessoa 3421

was scaled to a mean of 100 (on a per run basis), whichallowed the interpretation of the estimated regressioncoefficients in terms of percent signal change.

Voxelwise Analysis

Each participantʼs fMRI data were analyzed using a generallinear model framework in AFNI. There were eight experi-mental event types in the design matrix: no-reward and re-ward events during the cue phase and neutral, congruent,and incongruent events during the target phase, sepa-rately, for the no-reward and reward conditions. In ad-dition, both error and slow trials (in both cases pooledacross reward and no-reward conditions) were treatedseparately from correct trials and involved two additionalregressors, one accounting for cue-related activity andanother accounting for target-related activity. No assump-tions were made about the shape of the hemodynamicresponse function as responses were estimated via de-convolution. The lack of shape assumption was espe-cially pertinent given that cue-related responses could bemore or less sustained depending on brain region (andexperimental condition; see below). Responses were esti-mated starting from event onset to 15 sec after onset usingcubic spline basis functions. This method is closely relatedto the use of finite impulses (“stick functions”), the com-monly employed technique that can be considered thesimplest form of basis expansion. Cubic splines allow fora smoother approximation of the underlying responses,instead of the discrete approximation obtained by finiteimpulses. Constant, linear, and quadratic terms were in-cluded for each run separately (as covariates of no inter-est) to model baseline and drifts of the MRI signal. As anindex of activation, for each event type, we averaged theestimated responses at 5 and 7.5 sec after stimulus onset(as determined via the spline-based estimates). Exploratoryanalyses at earlier stages of data collection suggested thata similar pattern of results was obtained when the peakresponse was used as response index. Finally, note thatno separate estimate of the reward outcome phase wasdetermined, given its close temporal proximity to thetarget phase. In this manner, the estimates of target-relatedactivity included contributions from the outcome phase(see Supplementary Data). However, given our interestin evaluating interaction terms, the effect of the outcomephase was largely canceled out. Specifically, becauseboth incongruent and neutral trials during the rewardcondition had a reward–outcome phase, in our analyseswe focused on difference scores, that is (incongruent −neutral)REWARD and contrasted this difference to (incongru-ent − neutral)NO-REWARD (i.e., we evaluated the motivationby congruency interaction). However, as our design didnot allow the separation of outcome- from target-relatedactivity, we could not rule out the possibility that some dif-ference existed in outcome-related activity as a function ofcognitive trial type (incongruent, neutral), particularly inthe reward condition.

Estimation of Cue and Target Responses

Event-related designs allow the estimation of differentevent types when they occur in a randomized fashion.However, the present study, by design, required a fixedorder between the cue and target phases. In compara-ble situations, at times, a partial-trial design is employed(Ollinger, Shulman, & Corbetta, 2001). Here, instead, werandomized the delay between cue and target phases andthe intertrial interval. Because cue-related activation mayhave been sustained (at least in some brain regions),sequential dependencies may not have been fully elimi-nated (Ruge, Goschke, & Braver, 2009). We evaluated ourdesign in a set of computer simulations described in theSupplementary Data. As described, the contaminationin the estimation procedure produced by the sequentialnature of our design was negligible. In other words, target-related activity at 5 and 7.5 sec after stimulus onset was notcontaminated by cue-related responses. These results arealso consistent with the fact that the correlations betweendifferent regressors in our model were only modest anddid not exceed .37. Critically, our main objective was toinvestigate a potential motivation by congruency inter-action; accordingly, any spillover of motivation-related cuesignals would lead to a main effect type of contaminationand not one exhibiting an interaction pattern (note thatinformation about congruency was not provided at thetime of the cue).

Group Analysis

Whole-brain voxelwise random-effects analyses were con-ducted separately for the cue and target phases and wererestricted to gray matter voxels based on the FSL auto-mated segmentation tool (www.fmrib.ox.ac.uk/fsl/). Forthe cue phase, a paired t test was run to compare theactivations between reward and no-reward conditions.For the target phase, a separate 2 × 2 repeated mea-sures ANOVA was run to probe the interaction betweenMotivation (reward, no-reward) and Interference (in-congruent vs. neutral). Given the working hypotheses ofthis study, when reporting interactions (see Table 2),we focused on those regions that also exhibited a maineffect of Interference (incongruent vs. neutral pooledover reward and no-reward conditions). Note that nobias was associated with this approach as main effectsand interactions are orthogonal to each other. Finally,all voxelwise statistical tests were corrected for multi-ple comparisons using the false discovery rate approach(Genovese, Lazar, & Nichols, 2002) at p < .05. Given thatthis method corrects only at the voxel level and not interms of topological features such as peaks or regions ofactivation, we report here activations with a minimumcluster extent of 10 contiguous voxels. As illustrated byChumbley and Friston (2009), this approach reduces thelikelihood that false discovery rate correction will revealspurious activations.

3422 Journal of Cognitive Neuroscience Volume 23, Number 11

Relationship between Cue- andTarget-related Activity

An important goal of the present study was to understandthe link between cue-related responses, which were antici-pated to vary as a function of reward, and target-relatedresponses, which were anticipated to vary as a functionof both congruency and reward. Accordingly, we cor-related cue- and target-related responses. For the cuephase, we considered differential reward versus no-rewardresponses. For the target phase, we employed an in-teraction index that contrasted the differential responseincongruent versus neutral during the reward and no-reward conditions: [(incongruent − neutral)REWARD −(incongruent − neutral)NO-REWARD]. As outlined in the nextparagraph, this interaction index was investigated in theMPFC/ACC given this regionʼs purported role in conflictmonitoring (Botvinick, Braver, Barch, Carter, & Cohen,2001). In the article, we refer to this site as “MPFC.”Given our focused hypotheses, we performed Pearson

correlation analysis at the ROI level. Cue-related ROIs wereselected among fronto-parietal regions involved in atten-tional processing, namely, the intraparietal sulcus (IPS),FEF, middle frontal gyrus (MFG), and pre-SMA/SMA (seeTable 1 for locations). A 5-mm radius sphere was utilizedand centered on the peak voxel of the reward versus no-reward contrast at the group level. In a similar fashion,target-related activity was obtained from a 5-mm radiussphere centered on the peak voxel of the interactionin the MPFC, as revealed by the 2 Motivation (reward,no-reward) × 2 Congruency (neutral, incongruent) re-peated measures ANOVA at the group level. Because thisinteraction-based selection criterion is orthogonal to amain effect of Motivation, which was used to define thecue-related regions, no selection bias was incurred in thecreation of our ROIs (as stated, no congruency informationwas available at the time of the cue).

Network Analysis

To probe potential network interactions, we performedmediation analysis (Baron & Kenny, 1986), with the goal ofassessing the model shown in Figure 2A (for recent appli-cations to neuroimaging data, see Atlas, Bolger, Lindquist,& Wager, 2010; Lim, Padmala, & Pessoa, 2009; Wager,Davidson, Hughes, Lindquist, & Ochsner, 2008). This modelformalizes the hypothesis that attentional control duringthe cue phase is linked to the amount of target/distractorprocessing in visual cortex during the target phase, therebydetermining the extent of conflict-related responses duringthe target phase. A standard statistical approach wasadopted, which involved evaluating the following com-ponents (Baron & Kenny, 1986): (1) total effect c (initialvariable → outcome, which can also be written in terms ofthe indirect effect ab plus the direct effect c0), (2) indirectpath a (initial variable → intervening variable), (3) indirectpath b (intervening variable → outcome, after controlling

Table 1. Voxelwise Analysis at Cue Phase (Peak TalairachCoordinates and t Values)

Peak Location

Reward versus No-reward

x y z t(49)*

Occipital

Calcarine sulcus R/L 0 −80 3 4.39

Middle occipital gyrus R 26 −88 3 7.27

L −22 −88 3 6.09

Inferior occipital gyrus R 34 −80 −5 4.19

L −34 −78 −4 4.23

Precuneus R 25 −69 24 3.4

L −23 −71 30 2.85

Parietal

IPS R 24 −54 40 2.56

L −27 −52 41 3.54

Inferior parietal lobe L −28 −42 41 2.92

Frontal

Rostral ACC R 13 39 8 2.83

ACC R 6 8 39 6.38

L −8 7 39 5.03

SMA/pre-SMA R/L 0 −6 57 5.72

Frontal eye field R 34 −11 48 5.86

L −31 −12 50 4.6

Precentral gyrus L −48 −4 37 5.03

MFG R 26 46 25 3.74

L −28 35 29 4.41

Anterior insula R 31 17 11 4.03

L −35 26 5 3.7

Subcortical

Midbrain R 7 −15 −8 4.63

L −10 −18 −8 5.3

Putamen R 17 9 −2 5.34

L −19 9 2 5.7

Caudate R 10 9 2 5.66

L −10 9 2 5.77

Nucleus accumbens R 13 6 −7 5.03

L −13 6 −7 4.77

*p< .05 (false discovery rate corrected; minimum cluster extent: 10 voxels).

Padmala and Pessoa 3423

for the initial variable), and (4) direct effect c0 (initial vari-able → outcome, after controlling for the intervening vari-able). The final mediation effect was tested by assessingthe significance of the product of paths a and b (see Fig-ure 6) by using Sobelʼs test (Sobel, 1982).

Cue-related responses from the right IPS that weremodu-lated by reward and correlated with conflict-related activityin the MPFC at the target phase were employed as the initialpredictor in the mediation analysis (see Results below). Forthe mediator variable, responses evoked in the left fusiformgyrus during the target phase were employed, and MPFCtarget-related activity comprised the outcome variable. Be-cause we were interested in evaluating how reward affectedthese interactions, the analyses employed indices involvingdifferential responses, as indicated in Figure 6. In particular,an interaction-type index was employed during the targetphase. In the left fusiform gyrus, we reasoned that this pro-vided a measure of word-related processing (incongruentvs. neutral; note that “XXXX” was employed in the neutralcondition) and how this was affected by reward. For theMPFC, we assumed that the interaction term measuredhow conflict-related processing (incongruent vs. neutral)varied with reward.

For the network analysis, the fronto-parietal ROIs werethe same as the ones in the previous correlation analysisbetween cue and target activity. As stated, the interaction-based selection criterion used to select the MPFC ROI isorthogonal to a main effect of Motivation, which was usedto define the fronto-parietal ROIs. Accordingly, no selectionbias was incurred in the creation of our ROIs. In addition,the ROI for the left fusiform gyrus was obtained from sepa-rate localizer data, as described in the next section.

In addition to the right IPS ROI, cue-related responsesfrom the pre-SMA/SMA and right FEF ROIs also exhibitedsignificant correlations with conflict-related activity in theMPFC ROI at the target phase (see Results below). There-fore, for completeness, we ran two additional mediationanalyses: one employing pre-SMA/SMA cue-related re-sponses as the initial variable and another employing rightFEF cue-related responses as the initial variable.

Definition of ROIs in Visual Cortex

ROIs in visual cortex, namely, bilateral parahippocampalgyrus and left fusiform gyrus, were defined based on datafrom the localizer run. These ROIs were defined at the in-dividual level via the contrast of word versus scene blocks.Specifically, for the left fusiform gyrus, voxels were con-sidered that exhibited stronger response for word relativeto scene blocks ( p < .005, uncorrected); for the parahip-pocampal gyrus, the reverse contrast was employed. Inboth cases, a 5-mm radius sphere centered on the peakvoxel was used. Mean individual peak coordinates wereas follows: left fusiform gyrus: x = −39, y = −46, z =−14; left parahippocampal gyrus: x = −24, y = −44,z = −6; right parahippocampal gyrus: x = 25, y = −44,z = −7.

For the left fusiform gyrus ROI, we probed interferenceand facilitation effects. Accordingly, for target-related re-sponses, two separate 2 × 2 repeated measures ANOVAswere run to probe the interactions between Motivation(reward, no-reward) and Interference (incongruent, neu-tral) and between Motivation (reward, no-reward) andFacilitation (congruent, neutral). In the parahippocampalgyrus, surprisingly, we observed reduced responses duringthe reward relative to the no-reward condition irrespectiveof congruency, potentially because of a contamination ofthe responses linked to reward outcome (see Supplemen-tary Data). Accordingly, we did not perform ROI analyseson parahippocampal gyrus data.

Functional Connectivity Analysis at Cue Phase

We investigated potential interactions between the rightIPS ROI and subcortical reward-related ROIs by perform-ing a functional connectivity analysis based on the trial-by-trial methods proposed by DʼEsposito and colleagues(Rissman, Gazzaley, & DʼEsposito, 2004). Application ofthese methods to connectivity analysis has been reportedin several previous studies (Camara, Rodriguez-Fornells, &Munte, 2008; Nee, Jonides, & Berman, 2007; Daselaar,Fleck, Dobbins, Madden, & Cabeza, 2006; Buchsbaum,Olsen, Koch, & Berman, 2005). Here, for each partici-pant, trial-based responses were estimated during both thecue and target phases (as a function of trial type). Responsesto individual trials were estimated simultaneously. Specifi-cally, for each participant, a design matrix was set up suchthat each trialʼs cue and target phase was coded as a sepa-rate event. To provide response estimates for the cue andtarget phases of each trial, a hemodynamic response wasassumed (Cohen, 1997). Although in so doing we assumedthat evoked responses were transient (which may nothave been the optimal assumption for some brain regions,overall, trial-based estimates provided an excellent fit tothe data as illustrated in Supplementary Figure 2 (for anevaluation of this method in the context of functional con-nectivity analysis, see Zhou, Thompson, & Siegle, 2009).Note, however, that without assuming a fixed shape, theestimation of single-trial responses during relatively fast-paced event-related designs is poor and possibly unfeasible.The right IPS ROI and subcortical reward-related ROIs

were based on 5-mm radius spheres centered on the peakvoxel of the contrast reward versus no-reward defined atthe group level. Trial-based responses were averagedacross all voxels within the seed ROI as well as withinreward-related ROIs to create representative estimates.We then calculated the trial-by-trial Pearson correlationsbetween right IPS responses and subcortical reward-relatedROIs, separately for the reward and no-reward conditions.To contrast the correlations at the group level, they wereinitially transformed (via Fisherʼs Z transform) and thencompared via a paired t test. Because correlation valuesare orthogonal from differential activation scores, our pro-cedure did not incur circularities in the selection process.

3424 Journal of Cognitive Neuroscience Volume 23, Number 11

RESULTS

Behavioral Results

A 2 Motivation (reward, no-reward) × 3 Congruency (neu-tral, congruent, incongruent) repeated measures ANOVAon mean RT data (Figure 1B) revealed main effects ofMotivation (F(1, 49) = 99.24, p < .001) and Congruency(F(2, 98) = 128.18, p < .001), as well as a Motivation ×Congruency interaction (F(2, 98) = 12.11, p < .001).Further 2 × 2 ANOVAs revealed that reward decreasedboth interference (incongruent vs. neutral) and facilitationeffects (congruent vs. neutral), consistent with the notionthat motivation enhanced attentional filtering, thereby re-ducing the influence of the task-irrelevant word item (seeSupplementary Data for additional behavioral results; seealso Supplementary Figure 1B for error data).

Functional MRI Results

Cue and Target Phase

Cue-related responses were probed by contrasting thereward versus no-reward conditions. Increased evoked re-sponses during the reward condition were observed inseveral brain regions, including dorsal and ventral stria-tum, subcortically, and fronto-parietal regions, cortically(Figure 3; Table 1). The target phase involved severaltask components, including attentional selection, conflict-related processing (during incongruent trials), and rewardoutcome processing (during rewarded trials). Our maingoal was to evaluate interference-related responses inMPFC and how these were influenced by reward. Accord-ingly, as with behavioral data (see Supplementary Data),we performed a 2 Motivation (reward, no-reward) × 2Congruency (neutral, incongruent) repeated measuresANOVA, which revealed a significant main effect and, criti-cally, a significant interaction in MPFC (Figure 4A; Table 2),such that interference-related activity (incongruent −neutral) was reduced during the reward compared withno-reward condition (Figure 4B). No significant effectof Motivation was observed on Facilitation (congruent −neutral)-related activity in the MPFC. On the basis of these

voxelwise analyses at the cue and target periods, wedefined a series of ROIs that allowed us to evaluate themodel shown in Figure 2 in a focused manner, as describedin the subsequent sections (note also that all ROI analysesavoided problems of “circularity”; Kriegeskorte, Simmons,Bellgowan, & Baker, 2009; see Methods).

Relationship between Cue- and Target-related Activity

To probe the link between cue processing and interference-related activity during the target phase and how these wereinfluenced by reward, we performed correlation analy-ses. Cue-related ROIs were selected among fronto-parietalregions involved in attentional processing, namely, the IPS,FEF, MFG, and pre-SMA/SMA (see Table 1 for locations).The target-related ROI focused on the MPFC, which was de-fined based on the Motivation × Congruency interactiondetermined in the voxelwise analysis (see Table 2 for loca-tion). Significant Pearson correlations were obtained withthe right IPS (r(49) = −.39, p < .005; Figure 5) , pre-SMA/SMA (r(49) = −.34, p < .05) and right FEF (r(49) = −.39,p < .05), such that increased cue-related responses infronto-parietal regions during the reward condition werecorrelated with reduced interference in MPFC during thesubsequent target phase.

Visual Responses during the Target Phase inCategory-responsive Regions

We investigated category-related responses in inferior tem-poral cortex, including those in the left fusiform gyrus,which responds to word stimuli, and in the parahippo-campal gyrus, bilaterally, which responds to scene stimuli.We probed target-related responses in these ROIs (definedbased on separate localizer runs), according to 2 × 2repeated measures ANOVAs. In the left fusiform gyrus,significant Motivation × Congruency interactions weredetected both in the case of interference (incongruentvs. neutral: F(1, 49) = 11.39, p< .001) and facilitation (con-gruent vs. neutral: F(1, 49) = 4.44, p< .05), such that both

Figure 3. Cue-relatedresponses. Increased evokedresponses to reward relativeto no-reward were observedacross fronto-parietal cortexand subcortical regions.The color scale representsp values corrected formultiple comparisons viafalse discovery rate.

Padmala and Pessoa 3425

incongruent versus neutral and congruent versus neutraldifferential activity was reduced during the reward relativeto the no-reward condition (Figure 4C). Given that parallelbehavioral results were observed, namely, decreased in-terference and facilitation RT effects, visual responses wereconsistent with the notion that word items underwent re-duced processing during the reward condition, possiblybecause of increased attentional filtering.

Mediation Analysis

As described above, differential reward versus no-rewardresponses in fronto-parietal regions to the cue were in-versely correlated with interference-related activity duringthe target phase in the MPFC (see Figure 5). On the basisof connectivity analysis, Wang and colleagues recently sug-gested that the IPS plays an important role in coordinatingsites in medial and lateral pFC during conflict resolution(Wang et al., 2010). The IPS is also important for top–down attentional control and can bias the selection of vi-sual information (Greenberg, Esterman, Wilson, Serences,& Yantis, 2010; Hopfinger, Buonocore, & Mangun, 2000;Kastner, Pinsk, De Weerd, Desimone, & Ungerleider,1999). We thus reasoned that, in our task, the relation-ship between the IPS and MPFC might be mediated viavisual cortical regions involved in processing target anddistractor stimuli (see Figure 2A). We tested this hypothe-sis by carrying out a mediation analysis by focusing on theleft fusiform gyrus, the region identified in the previoussection. A statistically significant mediation was observed,

such that the total effect between the right IPS and theMPFC (see Figure 5) was significantly reduced once thecontribution of the left fusiform gyrus was taken intoaccount—a relationship evaluated by assessing the signifi-cance of the product of path weights a and b (ab = −.29,SE = 0.14, t(48) = 2.02, p < .05; Figure 6).We interpreted the relationship between responses

in the right IPS and left fusiform gyrus in terms of a filter-ing mechanism. If filtering was indeed involved, responsesin the right IPS should be correlated with reduced re-sponses in the left fusiform gyrus not only during interfer-ence processing as established, but also during facilitation(behaviorally, facilitation decreased, too). Accordingly,across participants, we correlated differential responses inthe right IPS during the cue phase (reward vs. no-reward)and facilitation-related responses in the left fusiformgyrus at the target phase [(congruent − neutral)REWARD −(congruent − neutral)NO-REWARD]. Stronger responses inthe right IPS were associated with reduced scores in the leftfusiform gyrus (r(50) = −.47, p < .005), consistent withthe filtering proposal.

Functional Connectivity Analysis at Cue Phase andIndividual Differences

We reasoned that, if motivationally salient cues engagefronto-parietal regions more robustly during the cue phase,they would exhibit increased “functional coupling” withreward-related brain regions that are capable of evaluatingthe motivational significance of the cues (see Figure 2B).

Figure 4. Target-relatedresponses. (A) Regions thatshowed significant Motivation(reward, no-reward) ×Interference (incongruent,neutral) interactions. AI =anterior insula. (B) Plot ofInterference (incongruentvs. neutral) and Facilitation(congruent vs. neutral) effectsin the MPFC ROI illustratingthe interaction patterns.(C) Plot of Interference(incongruent vs. neutral)and Facilitation (congruentvs. neutral) effects in theleft fusiform gyrus ROIillustrating the interactionpatterns. C = congruenttrials; I = incongruent trials;N = neutral trials. Errorbars in B and C representthe SEMs.

3426 Journal of Cognitive Neuroscience Volume 23, Number 11

Accordingly, we investigated the functional connectivitybetween the right IPS ROI and subcortical regions thatwere sensitive to the reward manipulation. Increased cor-relation with the right IPS during reward (vs. no-reward)was observed in the following subcortical ROIs: bilateralputamen (right: t(49) = 2.63, p < .05; left: t(49) = 2.28,p < .05) and bilateral nucleus accumbens (right: t(49) =2.37, p < .05; left: t(49) = 2.67, p < .05; Figure 7A and B).Do individual differences in reward sensitivity influence

the strength of coupling between the right IPS and thesesubcortical ROIs? To evaluate this possibility, BAS-drivescores, which have been proposed to be a good measureof appetitive motivation and approach behavior (Daweet al., 2004), were correlated with functional connectivityscores. Positive correlations were observed between right

IPS–right accumbens connectivity and BAS-drive scores(r(50) = .32, p < .05; Figure 7C).

DISCUSSION

How does motivation interact with cognitive control duringchallenging behavioral conditions? We hypothesized thata motivationally salient cue stimulus would allow partici-pants to upregulate control and thereby handle a response–conflict inducing stimulus in an improved manner. Our goalwas not only to probe how evoked responses were influ-enced by cognitive condition and reward but to advanceour understanding of the network interactions subserv-ing the impact of reward on cognitive control and visualprocessing. Behaviorally, participants exhibited reduced

Table 2. Voxelwise Analysis at Target Phase (Peak Talairach Coordinates and F Values)

Peak Location

Reward × Interference Interference Reward

x y z F(1, 49)* x y z F(1, 49)* x y z F(1, 49)*

Occipital

Middle occipital gyrus R 41 −70 17 18.96

Inferior occipital gyrus L −37 −76 −11 20.75

Parietal

IPS R 35 −57 35 16.89 35 −53 34 18.35 38 −61 41 25.03

L −32 −59 34 12.2 −31 −55 38 15.64

Inferior parietal lobe R 41 −43 38 30.81

Frontal

Rostral ACC R 8 26 20 14.11 5 32 14 23.82

Dorsal ACC R 8 8 38 23.35

MPFC R/L 0 19 45 15.05 0 19 45 14.82

SMA/pre-SMA R/L 0 12 50 7.52 0 12 50 23.86

Frontal eye field R 35 −7 47 18.34

L −35 −6 51 16.24

Precentral gyrus R 42 7 31 11.39 40 2 29 16.98

L −41 2 31 12.2 −42 5 29 29.39

Superior frontal gyrus R 31 46 14 8.97 21 45 24 17.32

MFG R 46 14 32 11.98 44 27 30 14.84 39 53 3 18.49

L −48 24 21 17.59

Anterior insula R 31 19 8 17.45 31 19 8 15.7 35 17 −4 31.96

L −31 19 4 14.09

Subcortical

Caudate R 9 6 11 17.57 12 6 14 10.72

*p < .05 (False Discovery Rate corrected; minimum cluster extent: 10 voxels).

Padmala and Pessoa 3427

conflict during the reward versus no-reward condition. Brainimaging results are discussed next.

Stronger cue-related responses during the reward con-dition were observed in a network of fronto-parietal re-gions involved in attention, several of which have beenshown to generate preparatory cue-related activity during

attentional tasks (Corbetta & Shulman, 2002; Kastner &Ungerleider, 2000). Previously, we observed that responsesin fronto-parietal attentional regions increased as a func-tion of the absolute magnitude of an incentive (Engelmannet al., 2009). In the present study, several subcortical sitesalso exhibited stronger reward-related signals at the cue

Figure 5. Relationshipbetween cue- and target-phaseresponses. The scatter plotillustrates the relationshipbetween cue-related responsesin the right IPS ROI andconflict-related responsesin the MPFC ROI. Increaseddifferential responses duringthe cue phase were linkedto decreased conflict-relatedresponses during thesubsequent target phase.The linear fit is shown forillustration purposes.Rewd = reward condition;No-Rewd = no-rewardcondition.

Figure 6. Network analysis.The relationship betweencue-related responses in theright IPS (X, initial variable)and target-related responsesin the MPFC (Y, outcomevariable) was mediated viathe left fusiform gyrus (M,mediator/intervening variable).The X, Y, and M variablesemployed in the analysisinvolved contrast or interactionterms, as indicated in redfont. The letters a, b, c, c0,refer to estimated pathcoefficients (see SupplementaryData). FG = fusiform gyrus;Rewd = reward condition;No-Rewd = no-rewardcondition; I = incongruenttrial; N = neutral trial. *p <.05, **p < .01, ***p < .005,two-tailed. The dotted lineindicates a path that wasnot statistically significantafter the mediator was takeninto account.

3428 Journal of Cognitive Neuroscience Volume 23, Number 11

phase, including the nucleus accumbens, caudate, puta-men, and a site in the midbrain that appeared to includethe ventral tegmental area, and to some extent the sub-stantia nigra. All of these subcortical structures are involvedin reward processing (Haber & Knutson, 2010; Camara,Rodriguez-Fornells, Ye, & Munte, 2009; Delgado, 2007;Schultz, 2000).During the target phase, paralleling the behavioral data,

we were interested in probing conflict-related responses(i.e., incongruent vs. neutral). In particular, we targetedresponses in the MPFC, as this region has been implicatedin conflict processing (Botvinick et al., 2001) and otherexecutive functions (Aarts, Roelofs, & van Turennout,2008; Brown & Braver, 2005; Weissman, Gopalakrishnan,Hazlett, & Woldorff, 2005). Regardless of the precise func-tions carried out by the MPFC, in specific paradigms it ap-pears to provide an index of the amount of conflict (Yeung& Nieuwenhuis, 2009; Kerns et al., 2004; van Veen, Cohen,Botvinick, Stenger, & Carter, 2001; MacDonald, Cohen,Stenger, & Carter, 2000). Here, conflict-related responsesin the MPFC decreased during the reward versus no-reward condition (i.e., an interference by reward interac-tion was observed).To investigate the relationship between cue- and target-

related responses, we correlated differential (i.e., reward vs.no-reward) responses at the cue phase with interference-related responses (i.e., (incongruent − neutral)REWARD −(incongruent − neutral)NO-REWARD) at the target phase.We observed an inverse linear relationship between cueand target responses, such that as differential cue-relatedresponses increased in fronto-parietal cortical regions in-volved in attentional control, conflict-related responses inMPFC decreased at the subsequent target phase. We in-terpret these findings as suggesting that motivationallysalient cues enhanced top–down control, thereby reducing

conflict-related responses (as well as reducing behavioralconflict)—and thus increasing the likelihood of reward.

In this study, we tested a model of how increased top–down control may affect conflict processing, namely bybiasing information processing in visual cortex in favorof task-relevant information (Egner & Hirsch, 2005) andagainst task-irrelevant information (only the latter wastested because our design did not allow us to isolatescene-related processing; see Supplementary Data). To testthis model, we performed a path analysis, which revealedthat the relationship between the right IPS (during cue)and the MPFC (during target) was mediated via responsesin the left fusiform gyrus (during target). For the latterregion, we reasoned that the contrast of incongruent andneutral trials provided an index of word-related process-ing, as only the former included a word stimulus (but notethat this approach does not assume that the left fusiformgyrus is strictly dedicated to word processing; see Xue &Poldrack, 2007). According to the path analysis, strongerdifferential responses in the right IPS were observed inconjunction with decreased word-related responses inthe left fusiform gyrus in the reward versus no-reward con-dition (path a), consistent with the idea that enhancedattentional control biased processing in a way that de-creased the influence of word stimuli. Put another way,reward improved the filtering of task-irrelevant stimuli.At the same time, decreased word-related processing inthe left fusiform gyrus was observed in conjunction withreduced conflict-related responses in the MPFC after con-trolling for right IPS cue responses (both during the targetphase; path b), strengthening the notion that the filteringof word stimuli reduced the amount of conflict registeredin MPFC. This interpretation is consonant with the be-havioral data where both interference and facilitation RT ef-fects were reduced with reward. In particular, the fact that

Figure 7. Functional connectivity. (A) Regions exhibiting stronger functional connectivity with the right IPS ROI during the cue phase (rewardvs. no-reward). (B) The scatter plot illustrates the trial-by-trial relationship between right IPS and right nucleus accumbens signals during thereward (red dots and line) and no-reward (green dots and line) conditions. Data are illustrated for one individual. The lines are included forillustration purposes and are linear fits to the data. NAcc = nucleus accumbens. (C) The scatter plot illustrates the relationship between BAS-drivescores and functional connectivity between right IPS and right nucleus accumbens. Participants with higher BAS-drive scores exhibited increasedfunctional coupling. The linear fit to the data is included for illustration purposes.

Padmala and Pessoa 3429

facilitation also decreased speaks in favor of the filteringnotion, as the potentially advantageous effect of a congru-ent word stimulus was reduced. It is worth pointing outthat we did not find evidence for enhancement effects invisual cortex because, as stated above, our design did notallow us to investigate this scenario. It is quite possible,however, that task-relevant information (i.e., scenes) repre-sented in the parahippocampal gyrus was also amplified(Polk et al., 2008; Egner &Hirsch, 2005) in the present task.

The role of the parietal cortex, in general, and the IPS,in particular, in controlling attention is well documented(Knudsen, 2007; Serences & Yantis, 2006; Shomstein &Behrmann, 2006; Bisley & Goldberg, 2003; Corbetta &Shulman, 2002). Furthermore, the parietal cortex is impor-tant for the filtering of distractor information (Friedman-Hill, Robertson, Desimone, & Ungerleider, 2003). Althoughgreat emphasis is often put on the role of pFC in conflictprocessing, interestingly, the parietal cortex also appearsto be involved (Egner, Delano, & Hirsch, 2007). Recently,it was proposed that conflict resolution relies on atten-tional processes carried out by the IPS via interactions withthemedial and lateral pFC (Wang et al., 2010). Our findingsare consistent with these reports and showed that the IPSis involved in attentional control during conflict processing.

We also observed effects of reward during the cue phaseon lateral pFC (e.g., bilateral MFG). This effect is note-worthy given the importance of the lateral pFC in the im-plementation of attentional control (Banich et al., 2009;Kerns et al., 2004). However, the network interactions thatwere observed involving the IPS were not detected in thecase of the MFG. It is possible that, in the present case,the IPS was involved more specifically given its role inattentional filtering (Friedman-Hill et al., 2003) and thatthe MFG would be involved in other types of control set-tings (but see Egner & Hirsch, 2005).

A final goal of our study was to advance the understand-ing of the mechanisms by which motivation influencestop–down control. In particular, we anticipated that thecoupling between subcortical regions important in evaluat-ing the motivational significance of the cue stimulus andfronto-parietal attentional regions important for attentionwould be enhanced during the reward condition. Indeed,increased functional coupling with the right IPS was ob-served with the bilateral nucleus accumbens and bilateralputamen (note that directionality cannot be inferred fromthe analysis performed).

Personality traits related to appetitive motivation andapproach behavior influence both how participants reactto incentives behaviorally (Savine, Beck, Edwards, Chiew,& Braver, 2010; van Steenbergen, Band, & Hommel,2009) and, importantly, the strength of the modulatoryeffects of reward on brain activity (Engelmann et al.,2009; Locke & Braver, 2008). Here, the functional con-nectivity between right IPS and right nucleus accumbenswas also correlated with BAS-drive scores, which suggeststhat reward sensitivity influences how regions interact witheach other during motivated conditions.

Conflict-related tasks have generated a healthy databaseof behavioral and neuroimaging findings in the past coupleof decades. Our goal in the present study was to capitalizeupon this rich paradigm to probe interactions betweenmotivation and executive function. Our findings suggestthat participants are able to employ motivationally sa-lient cues to upregulate top–down control processes thatbias the selection of visual information in a way that re-duces both behavioral conflict and conflict-related brainresponses. It is noteworthy that, in the case of congruenttrials, this strategy entailed less use of the task-irrelevantinformation and, accordingly, reduced facilitation scores.It should be pointed out, however, that in most trials thisstrategy did not compromise the likelihood of obtaining areward given that RTs were already fast in the congruentcondition. Overall, given that participants were not pro-vided information about the upcoming cognitive trial type(congruent, neutral, or incongruent) at the time of the cue,the overall best strategy might have been to attempt toreduce the impact of the task-irrelevant words. Finally, itappears that the ability to upregulate control is associatedwith how reward-related brain regions process motivation-ally salient cues and interact with fronto-parietal regionsimportant for the control of attention.

Acknowledgments

Support for this work was provided in part by the NationalInstitute of Mental Health (R01 MH071589) and the IndianaMETACyt Initiative of Indiana University, funded in part througha major grant from the Lilly Endowment, Inc. We would like tothank Morta Lapkus for assistance with summarizing personalityscores, Andrew Bauer for assistance with data collection, andJong-Moon Choi for discussions.

Reprint requests should be sent to Srikanth Padmala, Depart-ment of Psychological and Brain Sciences, Indiana University,Bloomington, IN, or via e-mail: spadmala@indiana.edu or LuizPessoa, Department of Psychology, University of Maryland, CollegePark, MD, or via e-mail: pessoa@umd.edu.

REFERENCES

Aarts, E., Roelofs, A., & van Turennout, M. (2008). Anticipatoryactivity in anterior cingulate cortex can be independent ofconflict and error likelihood. Journal of Neuroscience, 28,4671–4678.

Atlas, L. Y., Bolger, N., Lindquist, M. A., & Wager, T. D. (2010).Brain mediators of predictive cue effects on perceivedpain. Journal of Neuroscience, 30, 12964–12977.

Banich, M. T., Mackiewicz, K. L., Depue, B. E., Whitmer,A. J., Miller, G. A., & Heller, W. (2009). Cognitive controlmechanisms, emotion and memory: A neural perspectivewith implications for psychopathology. Neuroscienceand Biobehavioral Reviews, 33, 613–630.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychologicalresearch: Conceptual, strategic, and statisticalconsiderations. Journal of Personality and SocialPsychology, 51, 1173–1182.

Beaver, J. D., Lawrence, A. D., van Ditzhuijzen, J., Davis, M. H.,Woods, A., & Calder, A. J. (2006). Individual differences in

3430 Journal of Cognitive Neuroscience Volume 23, Number 11

reward drive predict neural responses to images of food.Journal of Neuroscience, 26, 5160–5166.

Bisley, J. W., & Goldberg, M. E. (2003). Neuronal activity inthe lateral intraparietal area and spatial attention. Science,299, 81–86.

Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S.,& Cohen, J. D. (2001). Conflict monitoring and cognitivecontrol. Psychological Review, 108, 624–652.

Braver, T. S., Gray, J. R., & Burgess, G. C. (2007). Explainingthe many varieties of working memory variation: Dualmechanisms of cognitive control. In A. R. A. Conway,C. Jarrold, M. J. Kane, A. Miyake, & J. N. Towse (Eds.),Variation in working memory (pp. 76–106). Oxford,UK: Oxford University Press.

Brown, J. W., & Braver, T. S. (2005). Learned predictions oferror likelihood in the anterior cingulate cortex. Science,307, 1118–1121.

Buchsbaum, B. R., Olsen, R. K., Koch, P., & Berman, K. F.(2005). Human dorsal and ventral auditory streamssubserve rehearsal-based and echoic processes duringverbal working memory. Neuron, 48, 687–697.

Camara, E., Rodriguez-Fornells, A., & Munte, T. F. (2008).Functional connectivity of reward processing in thebrain. Frontiers in Human Neuroscience, 2, 19.

Camara, E., Rodriguez-Fornells, A., Ye, Z., & Munte, T. F. (2009).Reward networks in the brain as captured by connectivitymeasures. Frontiers in Neuroscience, 3, 350–362.

Carter, C. S., Braver, T. S., Barch, D. M., Botvinick, M. M.,Noll, D., & Cohen, J. D. (1998). Anterior cingulate cortex,error detection, and the online monitoring of performance.Science, 280, 747–749.

Carver, C. S., & White, T. L. (1994). Behavioral inhibition,behavioral activation, and affective responses to impendingreward and punishment: The BIS/BAS scales. Journal ofPersonality and Social Psychology, 67, 319–333.

Chumbley, J. S., & Friston, K. J. (2009). False discovery raterevisited: FDR and topological inference using Gaussianrandom fields. Neuroimage, 44, 62–70.

Cohen, M. S. (1997). Parametric analysis of fMRI data usinglinear systems methods. Neuroimage, 6, 93–103.

Corbetta, M., & Shulman, G. L. (2002). Control of goal-directedand stimulus-driven attention in the brain. Nature ReviewsNeuroscience, 3, 201–215.

Cox, R., & Jesmanowicz, A. (1999). Real-time 3D imageregistration for functional MRI. Magnetic Resonancein Medicine, 42, 1014–1018.

Cox, R. W. (1996). AFNI: Software for analysis and visualizationof functional magnetic resonance neuroimages. Computersand Biomedical Research, 29, 162–173.

Daselaar, S. M., Fleck, M. S., Dobbins, I. G., Madden, D. J., &Cabeza, R. (2006). Effects of healthy aging on hippocampaland rhinal memory functions: An event-related fMRI study.Cerebral Cortex, 16, 1771–1782.

Dawe, S., Gullo, M. J., & Loxton, N. J. (2004). Reward drive andrash impulsiveness as dimensions of impulsivity: Implicationsfor substance misuse. Addictive Behaviors, 29, 1389–1405.

Delgado, M. R. (2007). Reward-related responses in thehuman striatum. Annals of the New York Academy ofSciences, 1104, 70–88.

Egner, T., Delano, M., & Hirsch, J. (2007). Separate conflict-specific cognitive control mechanisms in the human brain.Neuroimage, 35, 940–948.

Egner, T., & Hirsch, J. (2005). Cognitive control mechanismsresolve conflict through cortical amplification of task-relevantinformation. Nature Neuroscience, 8, 1784–1790.

Engelmann, J. B., Damaraju, E. C., Padmala, S., & Pessoa, L.(2009). Combined effects of attention and motivationon visual task performance: Transient and sustained

motivational effects. Frontiers in Human Neuroscience,3, doi:10.3389/neuro.3309.3004.2009.

Epstein, R., Harris, A., Stanley, D., & Kanwisher, N. (1999).The parahippocampal place area: Recognition, navigation,or encoding? Neuron, 23, 115–125.

Friedman-Hill, S., Robertson, L. C., Desimone, R., &Ungerleider, L. G. (2003). Posterior parietal cortex andthe filtering of distractors. Proceedings of the NationalAcademy of Sciences, U.S.A., 2003, 4263–4268.

Genovese, C. R., Lazar, N. A., & Nichols, T. (2002). Thresholdingof statistical maps in functional neuroimaging using thefalse discovery rate. Neuroimage, 15, 870–878.

Greenberg, A. S., Esterman, M., Wilson, D., Serences, J. T.,& Yantis, S. (2010). Control of spatial and feature-basedattention in frontoparietal cortex. Journal of Neuroscience,30, 14330–14339.

Haber, S. N., & Knutson, B. (2010). The reward circuit:Linking primate anatomy and human imaging.Neuropsychopharmacology, 35, 4–26.

Hopfinger, J. B., Buonocore, M. H., & Mangun, G. R. (2000).The neural mechanisms of top–down attentional control.Nature Neuroscience, 3, 284–291.

Jorm, A. F., Christensen, H., Henderson, A. S., Jacomb, P. A.,Korten, A. E., & Rodgers, B. (1998). Using the BIS/BAS scalesto measure behavioral inhibition and behavioral activation:Factor structure, validity and norms in a large communitysample. Personality and Individual Differences, 26, 49–58.

Kastner, S., Pinsk, M. A., De Weerd, P., Desimone, R., &Ungerleider, L. G. (1999). Increased activity in humanvisual cortex during directed attention in the absenceof visual stimulation. Neuron, 22, 751–761.

Kastner, S., & Ungerleider, L. G. (2000). Mechanisms ofvisual attention in the human cortex. Annual Review ofNeuroscience, 23, 315–341.

Kerns, J. G., Cohen, J. D., MacDonald, A. W., Cho, R. Y.,Stenger, V. A., & Carter, C. S. (2004). Anterior cingulateconflict monitoring and adjustments in control. Science,303, 1023–1026.

Knudsen, E. I. (2007). Fundamental components of attention.Annual Review of Neuroscience, 30, 57–78.

Kriegeskorte, N., Simmons, W. K., Bellgowan, P. S., & Baker, C. I.(2009). Circular analysis in systems neuroscience: The dangersof double dipping. Nature Neuroscience, 12, 535–540.

Lim, S. L., Padmala, S., & Pessoa, L. (2009). Segregating thesignificant from the mundane on a moment-to-momentbasis via direct and indirect amygdala contributions.Proceedings of the National Academy of Sciences, U.S.A.,106, 16841–16846.

Locke, H. S., & Braver, T. S. (2008). Motivational influenceson cognitive control: Behavior, brain activation, andindividual differences. Cognitive, Affective & BehavioralNeuroscience, 8, 99–112.

MacDonald, A. W., III, Cohen, J. D., Stenger, V. A., & Carter,C. S. (2000). Dissociating the role of the dorsolateralprefrontal and anterior cingulate cortex in cognitivecontrol. Science, 288, 1835–1838.

Mayr, U., Awh, E., & Laurey, P. (2003). Conflict adaptationeffects in the absence of executive control. NatureNeuroscience, 6, 450–452.

McCandliss, B. D., Cohen, L., & Dehaene, S. (2003). The visualword form area: Expertise for reading in the fusiformgyrus. Trends in Cognitive Sciences, 7, 293–299.

Nee, D. E., Jonides, J., & Berman, M. G. (2007). Neuralmechanisms of proactive interference resolution.Neuroimage, 38, 740–751.

Ollinger, J. M., Shulman, G. L., & Corbetta, M. (2001). Separatingprocesses within a trial in event-related functional MRI.Neuroimage, 13, 210–217.

Padmala and Pessoa 3431

Pessoa, L. (2009). How do emotion and motivation directexecutive function? Trends in Cognitive Sciences, 13, 160–166.

Pessoa, L., & Engelmann, J. B. (2010). Embedding rewardsignals into perception and cognition. Frontiers inNeuroscience, 4, 17.

Polk, T. A., Drake, R. M., Jonides, J. J., Smith, M. R., & Smith,E. E. (2008). Attention enhances the neural processingof relevant features and suppresses the processing ofirrelevant features in humans: A functional magneticresonance imaging study of the stroop task. Journalof Neuroscience, 28, 13786–13792.

Polk, T. A., & Farah, M. J. (2002). Functional MRI evidencefor an abstract, not perceptual, word-form area. Journalof Experimental Psychology: General, 131, 65–72.

Rissman, J., Gazzaley, A., & DʼEsposito, M. (2004). Measuringfunctional connectivity during distinct stages of a cognitivetask. Neuroimage, 23, 752–763.

Ruge, H., Goschke, T., & Braver, T. S. (2009). Separatingevent-related BOLD components within trials: Thepartial-trial design revisited. Neuroimage, 47, 501–513.

Savine, A. C., Beck, S. M., Edwards, B. G., Chiew, K. S., &Braver, T. S. (2010). Enhancement of cognitive control byapproach and avoidance motivational states. Cognition &Emotion, 24, 338–356.

Schultz, W. (2000). Multiple reward signals in the brain.Nature Reviews Neuroscience, 1, 199–207.

Serences, J. T., & Yantis, S. (2006). Selective visual attention andperceptual coherence. Trends in Cognitive Sciences, 10, 38–45.

Shomstein, S., & Behrmann, M. (2006). Cortical systemsmediating visual attention to both objects and spatiallocations. Proceedings of the National Academy ofSciences, U.S.A., 103, 11387–11392.

Sobel, M. E. (1982). Asymptotic confidence intervals for indirecteffects in structural equation models. In S. Leinhardt (Ed.),Sociological methodology (pp. 290–312). San Francisco, CA:Jossey Bass.

Talairach, J., & Tournoux, P. (1988). Co-planar stereotaxicatlas of the human brain. Stuttgart: Thieme.

van Steenbergen, H., Band, G. P., & Hommel, B. (2009).Reward counteracts conflict adaptation: Evidence for arole of affect in executive control. Psychological Science,20, 1473–1477.

van Veen, V., Cohen, J. D., Botvinick, M. M., Stenger, V. A.,& Carter, C. S. (2001). Anterior cingulate cortex, conflictmonitoring, and levels of processing. Neuroimage, 14,1302–1308.

Wager, T. D., Davidson, M. L., Hughes, B. L., Lindquist, M. A.,& Ochsner, K. N. (2008). Prefrontal-subcortical pathwaysmediating successful emotion regulation. Neuron, 59,1037–1050.

Wang, L., Liu, X., Guise, K. G., Knight, R. T., Ghajar, J., &Fan, J. (2010). Effective connectivity of the fronto-parietalnetwork during attentional control. Journal of CognitiveNeuroscience, 22, 543–553.

Watanabe, M. (2002). Integration across multiple cognitiveand motivational domains in monkey prefrontal cortex. InD. T. Stuss & R. T. Knight (Eds.), Principles of frontal lobefunction (pp. 326–337). New York: Oxford University Press.

Weissman, D. H., Gopalakrishnan, A., Hazlett, C. J., & Woldorff,M. G. (2005). Dorsal anterior cingulate cortex resolves conflictfrom distracting stimuli by boosting attention towardrelevant events. Cerebral Cortex, 15, 229–237.

Xue, G., & Poldrack, R. A. (2007). The neural substrates ofvisual perceptual learning of words: Implications for thevisual word form area hypothesis. Journal of CognitiveNeuroscience, 19, 1643–1655.

Yeung, N., & Nieuwenhuis, S. (2009). Dissociating responseconflict and error likelihood in anterior cingulate cortex.Journal of Neuroscience, 29, 14506–14510.

Zhou, D., Thompson, W. K., & Siegle, G. (2009). MATLABtoolbox for functional connectivity. Neuroimage, 47,1590–1607.

3432 Journal of Cognitive Neuroscience Volume 23, Number 11