a componential analysis of task‐switching deficits associated with lesions of left and right...

8/6/2019 A componential analysis of taskswitching deficits associated with lesions of left and right frontal cortex

1/13

A componential analysis of task-switching decitsassociated with lesions of left and right frontal

cortexAdam R. Aron, 1,4 Stephen Monsell, 3 Barbara J. Sahakian 1 and Trevor W. Robbins 2

1 Department of Psychiatry and 2 Department of Experimental Psychology, University of Cambridge,Cambridge, 3School of Psychology, University of Exeter,Exeter, UK and 4 Department of Psychology, University of California, Los Angeles, CA, USA

Correspondence to: Trevor W. Robbins, Department of Experimental Psychology, University of Cambridge, Downing Street, Cambridge CB2 3EB, UK E-mail: [email protected]

SummaryExecutive functions such as task-set switching arethought to depend on the frontal cortex. However, moreprecision is required in identifying which componentsof such high-level processes relate to which, if any, sub-regions of the brain. In a recent study of 19 patientswith focal right frontal (RF) lesions and 17 with leftfrontal (LF) lesions, we found that response inhibition,as measured by the stop-signal task, was specicallydisrupted by damage to the right inferior frontal gyrus(IFG). The present study examined task-switchingperformance in this same group of patients and inmatched controls on the grounds that inhibitory mech-anisms may also be required to switch task-set. Both

RF and LF patients showed signicantly larger switchcosts (the difference, in reaction time and errors,

between changing tasks and repeating the same task)than controls, but apparently for different reasons. ForRF patients, a part of the switch decit could beaccounted for by impaired inhibition of inappropriateresponses or task-sets triggered by stimuli, and onemeasure of the switch cost correlated reliably with dam-age to the IFG, specically the pars opercularis (POp).For LF patients, a part of the switch decit may havearisen from weak top-down control of task-set. Thedegree of top-down control correlated reliably with theextent of damage to the left middle frontal gyrus(MFG). This study localizes two components of thecomplex task-switching process (inhibition of task-sets

and/or responses and top-down control of task-set) tothe right IFG/POp and the left MFG respectively.

Keywords : inhibition; executive function; middle frontal gyrus; inferior frontal gyrus; pars opercularis

Abbreviations : C = congruent trial; CCNRP = Cambridge Cognitive Neuroscience Research Panel; exPOp = IFG regionsexcluding POp; IC = incongruent trial; IFG = inferior frontal gyrus; LF = left frontal; MED = medial; MFG = middlefrontal gyrus; N = neutral trial; NART = National Adult Reading Test; NS = non-switch trial; ORB = orbitofrontal; PFC =prefrontal cortex; POp = pars opercularis; RF = right frontal; ROI = region of interest; RSI = responsestimulus interval;RT = reaction time; SC = switch cost; SFG = superior frontal gyrus; SR = stimulusresponse; SSRT = stop-signalreaction time; SW = switch trial; ROI = region of interest

Received August 24, 2003. Revised November 15, 2003. Second revision February 18, 2004. Accepted February 21, 2004. Advance Access publication April 16, 2004

IntroductionTwo widely studied executive functions are response inhib-ition and task-set reconguration. The former can be studiedwith the stop-signal paradigm (Logan and Cowan, 1984), inwhich the stimulus species a rapid response which theparticipant makes on most trials, but must try to suppress if astop-signal is concurrently presented. Task-set recongura-tion has been studied with variants of a task-switchingparadigm (for review see Monsell, 2003) in which the

participant responds to a stimulus on each trial according tosome stimulusresponse (SR) task rule(s), and on some trialsmust change to a different task. To perform any such task, aparticipant must chain together and congure appropriately atask-set: an appropriate set of processes linking sensoryanalysis via the categorization or identication of thestimulus to the choice of a response and execution of motoroutput (Rogers and Monsell, 1995). To change tasks, one or

Brain Vol. 127 No. 7 Guarantors of Brain 2004; all rights reserved

DOI: 10.1093/brain/awh169 Brain (2004), 127, 15611573


2/13

more components of the task-set must be recongured.Behaviourally, the need to recongure the task-set results in asubstantial switch cost: longer reaction time (RT) and moreerrors on task-switch than on task-repeat trials (Fig. 1A). Anumber of authors have suggested that the switch cost arisesin part from the need to inhibit competing SR links speciedby the now inappropriate task (Rogers and Monsell, 1995), orto inhibit entire task-sets (Rogers and Monsell, 1995;Arbuthnott and Frank, 2000; Mayr and Keele, 2000; Mayr,2002; Schuch and Koch, 2003). A primary motive for thisstudy was to explore the neural correlates of these putativeinhibitory processes in task-switching by assessing beha-vioural performance in patients with unilateral lesions of thefrontal cortex whom we had also tested, in the same sitting, ona specic measure of response inhibition (c.f. Aron et al .,2003).

There are several reasons for supposing that the rightfrontal (RF) cortex may subserve inhibitory processesunderlying task-switching. First, many neuroimaging studies

of response inhibition (Konishi et al ., 1998; Garavan et al .,1999; Konishi et al ., 1999; Rubia et al ., 1999; Menon et al .,2001; Bunge et al ., 2002; Rubia et al ., 2003) and a number of neuroimaging studies of switching task, strategy or dimen-sions, including reversal-learning (Nagahama et al ., 2001;Cools et al ., 2002), the Wisconsin Card Sorting Test (WCST)(Monchi et al ., 2001; Nagahama et al ., 2001; Nakahara et al .,2002) and task-set switching (Dove et al ., 2000; Sohn et al .,2000; Dreher and Berman, 2002; Brass et al ., 2003) haveespecially, although not exclusively, reported activation,within the right hemisphere, of the inferior frontal gyrus(IFG) (Fig. 2). Secondly, a direct neuroimaging comparisonof a form of switching (the WCST) and response inhibitiondemonstrated a common locus in the right IFG (Konishi et al .,1999). Thirdly, a combined EEG/functional MRI studyinvestigating a Go versus Wait factor and a Switch versusNon-switch factor suggested that the right IFG locus wasrelated neither to pure switching nor to pure responseinhibition, but was instead responsible for `switching into asuppression mode' (Swainson et al ., 2003). Although theseneuroimaging ndings indicate neural substrates activated bythese executive functions, human lesion studies can providemore denitive proof that a given brain region is necessary.Recently, we established that the right IFG is indeednecessary for response inhibition by using a region of interest(ROI) MRI-based method in a sample of patients withexcisions of the frontal cortex: the greater the damage to thatROI alone, the longer it took the patients to inhibit their

Fig. 1 The task design and the run-position effect. ( A) Data for thethree subject groups from this experiment illustrate the run-position effect (Rogers and Monsell, 1995). Run-position 1 (theswitch) has elevated RT relative to non-switch positions 2 and 3.(B) An AAABBB design was employed whereby the relevant task (arrow or word) changed every three trials. The position labels `1',`2' and `3' were not shown to the subject, but are depicted herefor explanatory purposes. The position of the thick bar denotes theposition of the switch (run position 1). On each trial a cue(`arrow' or `word') appeared, followed by the stimulus. On thesubsequent trial the cue (and stimulus) appeared at the next,clockwise, position. Stimuli could be congruent (e.g. a left arrowwith the word `left', because it specied a left response for boththe Word and the Shape tasks), incongruent (e.g. a left arrow withthe word `right', because it specied a left response on the shape

task but a right response on the other task) or neutral (e.g. wherethe shape around the word was a rectangle or the letter stringwithin an arrow was `xxxx', so that the irrelevant attribute wasassociated with no response).

Fig. 2 Talairach coordinates plotted from six neuroimaging studiesof switching/sorting/reversing, and boundaries of inferior frontalgyrus (IFG). 3D-rendered sagittal view showing that severalreported task-set switching, Wisconsin Card Sorting Test andreversal foci fall within the inferior frontal gyrus, dened caudallyby the precentral sulcus, ventrally by the sylvian ssure androstrally by the inferior frontal sulcus. All lateral right frontalcoordinates are shown as colour codes: yellow, Cools et al .(2002); blue, Dreher and Berman (2002); purple, Sohn et al .(2000); red, Nakahara et al . (2002); black, Dove et al . (2000);brown, Monchi et al . (2001); green, Brass et al ., (2003).

1562 A. R. Aron et al .


3/13

responses (Aron et al ., 2003). The present study aimed toextend this approach, in the same group of patients, to task-setswitching in order to test the hypothesis of a common rightIFG substrate underlying both task-set switching andresponse inhibition executive functions, by assessing whether

any switching decit may plausibly relate to (i) inhibition of an inappropriate task-set and/or (ii) inhibition of responsetendencies activated via an inappropriate SR rule.

How a task-set switching experiment may be used toisolate these putative inhibitory factors requires explanation.First, it appears that a stimulus evokes tendencies to performboth a task recently or frequently associated with it andspecic responses associated with the stimulus (Rogers andMonsell, 1995; Allport and Wylie, 1999). Switching betweentasks may require inhibition at the level of the task-set and atthe level of individual responses. The degree to which suchinhibition is effective at both levels may be revealed bycomparisons of congruent, incongruent and neutral stimuli in

a task-switching design (Fig. 1B). As congruent andincongruent stimuli are equally associated with the twotasks, any difference in response latencies to them (Stroop-like interference) must reect competition due to activation of the irrelevant response. It is also sometimes observed (Rogersand Monsell, 1995) that congruent stimuli (which bydenition create no competition at a response level, but areassociated with both tasks) are responded to more slowly thanneutral stimuli (which are associated with only one task). Thiscan only reect competition from activation of the irrelevanttask-set. Hence a slower RT for congruent than for neutraltrials indexes difculty in inhibiting the irrelevant task-set,while a slower RT for incongruent than congruent trialsindexes difculty in inhibiting response tendencies activatedvia the irrelevant task-set. As interference from a recentlyperformed task may be observed even in a pure (i.e. non-switching) task block (Wylie and Allport, 2000), we exam-ined two measures that capture this competition on non-switch trials alone:

CN = RT NS C RT NS NICC = RT NS IC RT NS C

where the subscript NS refers to non-switch trials and IC, Cand N refer to incongruent, congruent and neutral stimulirespectively. Furthermore, as interference effects are typic-ally magnied on switch trials relative to non-switch trials(Rogers and Monsell, 1995; Meiran, 2000; Meiran et al .,2000) we also examined two measures that capture this moretransient competition from the task being switched from:

SC CN = (RT SW C RT SW N) (RT NS C RT NS N)SC ICC = (RT SW IC RT SW C) (RT NS IC RT NS C)

where the subscripts SW and SC refer to switch trials and theswitch cost respectively. The CN and ICC measures are notindependent, as more activation of the competing task setenables more activation of corresponding response tenden-cies, and this may facilitate congruent responses. However, apositive SC CN implies greater activation of the competing

task set on switch relative to non-switch trials. A largerSC ICC implies greater difculty in suppressing responseactivation generated by the irrelevant SR mapping on switchrelative to non-switch trials. If RF damage affects aninhibitory mechanism generalizable to task-set switching,

RF patients may be expected to show impairments on eitheror both measures.We also varied the responsestimulus interval (RSI)

between 100 ms (short) and long 1500 ms (long) to assessdecits in voluntary reconguration. The standard observa-tion in normal participants is that, as the time available forpreparation for the next task (the RSI in the presentexperiment) increases, switch costs decline but are noteliminated. The reduction has been attributed to endogenousprocesses that enable the required task-set and/or suppress theprevious task-set and are carried out in anticipation of thestimulus (Rogers and Monsell, 1995; Meiran, 1996; De Jong,2000; Rubinstein et al ., 2001). Such preparation can be

indexed by subtracting the switch cost at long RSI from theswitch cost at short RSI (we refer to this difference asSC REDUCT ). Interpretation of the residual switch still seenafter a long preparation interval (here designated SC RESID andestimated simply as the switch cost at the long RSI) iscontroversial (see Monsell, 2003). Suggested sources includean exogenous control process that is required to completetask-set reconguration and triggered by the stimulus onset(Rogers and Monsell, 1995; Rubinstein et al ., 2001), post-stimulus completion, on a proportion of trials, of areconguration process that failed to engage endogenously(De Jong, 2000), interference with response selection due totask-set inertia (Allport et al ., 1994; Allport and Wylie, 1999)or, more specically, persisting task-set inhibition (Meuterand Allport, 1999; Goschke, 2000; Schuch and Koch, 2003).If switching decits in patients relate to endogenous control,they would be expected to interact further with RSI, whiledecits that are equally apparent at short and long RSI arelikely to derive from processes triggered by stimulus onsetand/or carry-over of interference from the pre-switch trial,which the participant cannot pre-empt by active preparation.

Another motive for the study was to investigate theconsequences of LF damage for task-set switching and tocompare these with the consequences of RF damage. Anumber of neuropsychological studies suggest that LFcortical damage may affect task-set switching (Stablumet al ., 1994; Rogers et al ., 1998; Mecklinger et al ., 1999;Keele and Rafal, 2000). Mecklinger et al . (1999) found thatpatients with speech and language disorders had the greatestswitch decits, suggesting left hemisphere linguistic regions(e.g. the IFG) may be critical. Rogers et al . (1998) found thatpatients with LF damage had switch decits only underconditions in which there was interference between the tasks(i.e. where the establishment of task-set was particularlyimportant). Neuroimaging studies suggest that maintenanceand establishment of the task-set are functions attributable tothe left dorsolateral prefrontal cortex (PFC) (especially themiddle frontal gyrus, MFG) (MacDonald et al ., 2000;

Frontal cortical lesions and task-switching 1563


4/13

Garavan et al ., 2002). Keele and Rafal (2000) reported thatdamage to this same area led to an inability to recover from atask switch with the normal reduction in RT on the second orthird trial following the switch in a single, extensively tested,LF patient. In brief, the role of the LF cortex in task-switchingis not particularly clear, perhaps owing to the fact that none of the neuropsychological studies employed particularly size-able samples, nor could the locus of damage be specied withmuch precision. We therefore examined the same measures inLF as in RF patients. Specic difculties inhibiting theinappropriate task-set would be expected to show up in theCN and ICC measures, while more general difcultiesimposing a changed task-set would be expected to result in anabnormally large switch cost or other indices of recoveryfrom a change of tasks. Additionally, the putative importanceof the left IFG, MFG or other ROIs might be revealed byreliable correlations between extent of damage to theseregions and indices of switching and task control.

Material and methods ParticipantsThirty-six patients were recruited from the Cambridge CognitiveNeuroscience Research Panel (CCNRP) by referral fromAddenbrooke's Hospital, Cambridge. The study was approved bythe Cambridge Local Research Ethics Committee and all patientsgave informed consent prior to participation. Seventeen patients hada single focal lesion conned to the left PFC and 19 to the right PFC,veried by MRI in 33 of the patients and computer-assistedtomography (CAT) in three. Lesion aetiology was mostly tumourresection or cerebrovascular haemorrhage (Supplementary Table 1).We excluded patients with current or previous psychiatric diagnosis,colour blindness, or neurological disease other than that determininginclusion in the study. Twenty healthy control volunteers from theEast Anglia area were obtained either through advertisement orthrough the CCNRP and were paid. Controls were matched withpatients for age and estimated premorbid verbal IQ as assessed bythe National Adult Reading Test (NART) (Table 1). There were no

differences between groups in terms of age [ F (2,53) < 1, n.s.], NARTscore [ F (2,51) < 1, n.s.] or time since trauma for the two frontalpatients groups [ F (1,34) < 1, n.s].

Neuroradiological assessmentThirty-three of the 36 frontal excision patients received MRI scansof the brain, with 3D set acquisition in the coronal plane using aSPGR (spin gradient echo) T1-weighted sequence and a T2-weighted axial sequence (using the 1.5 T scanner at the MRISunit, Addenbrooke's Hospital, Cambridge). MRI scans wereinterpreted for the CCNRP by two neurologists blind to theexperimental results, and lesions were traced using MRIcro software(www.mricro.com). They were then normalized to a standardtemplate using SPM96 (Wellcome Department of CognitiveNeurology, London) using cost function masking (Brett et al .,2001). For the other three patients, MRI scans could not be obtainedand the lesion loci were estimated from CAT scans.

ROI method The frontal lobes of each hemisphere were divided into ve ROIs byDr P. C. Fletcher (Department of Psychiatry, University of Cambridge), who was blind to experimental results. These ROIswere: superior frontal gyrus (SFG), middle frontal gyrus (MFG),inferior frontal gyrus (IFG), orbitofrontal (ORB) and a medial area(MED) (for gure, see Aron et al ., 2003). MRIcro was used to tracethese areas onto the standard T1 template used by SPM99 (consistingof averaged scans from 152 healthy subjects). Because it was shownthat damage to a subregion of the right IFG, the pars opercularis(POp) [in Aron et al . (2003) we erroneously referred to the parsopercularis as the pars triangularis], was particularly critical forresponse inhibition (Aron et al ., 2003), an additional POp ROI wasused from the Automated Anatomical Labelling (AAL) map(Tzourio-Mazoyer et al ., 2002). For control purposes, non-POpregions of the IFG (i.e. pars triangularis and an orbital region fromthe AAL map) were combined into an exPOp region (IFG excludingPOp). For each hemisphere, the volumes of these ROIs (in cc) were104.1 (SFG), 68.8 (MFG), 49.2 (IFG), 39.7 (ORB), 56.1 (MED),12.5 (POp) and 33.1 (exPOp). The normalized lesion for each patient

Table 1 Demographic information on study participants

CT LF RF Statistic

Sample size 20 17 19Hand (L : R) 19 : 1 16 : 1 18 : 1Sex (M : F) 12 : 8 10 : 7 7 : 12Age 52.6 (10.8) 51.9 (10) 53.6 (10.0) F < 1, n.s.PV-IQ 115.9 (5.1) 116 (5.1) 113.2 (8.5) F = 1.1, n.s.Chronicity 3.1 (2.5) 3.2 (3.4) F < 1, n.s.Lesion size 31.2 (38.1) 63.5 (62.3) t = 1.6, n.s.SFG 6.8 (11.9) 14.7 (18.0) t = 1.5, n.s.MFG 7.9 (10.3) 15.9 (16.6) t = 1.6, n.s.IFG 6.5 (9.4) 8.7 (10.7) t < 1, n.s.ORB 4.3 (6.8) 5.9 (8.5) t < 1, n.s.MED 1.7 (2.8) 3.8 (6.3) t = 1.2, n.s.

PV-IQ = predicted verbal IQ from the National Adult Reading Test; chronicity = years since frontaltrauma; lesion size = total volume of lesion (cc) and average volume of damage to each of the veROIs (cc). There were no reliable differences between left and right frontal patients in the extent of damage to prefrontal regions of interest.



5/13

was superimposed onto each of these ROIs in order to compute thevolume of damaged grey matter. Although the RF group had moreoverall damage than the LF group, this was not a signicantdifference, nor were there signicant differences for any of the ROIs(Table 1).

Tasks and procedureThe experiment was run on a PC using ERTS (Frankfurt, Germany),an MS-DOS program with 0.6 ms timing resolution. Subjects sat 50cm from a computer screen on which was displayed a framework consisting of three lines radiating 10 cm from the centre at equalangles to form an inverted Y dening three sectors, in one of whichthe stimulus was displayed about 25 mm from the centre. Stimuli onsuccessive trials were displayed in successive sectors, clockwise(Fig. 1B). Immediately after the previous response, a task cue (theword `arrow' or `word') was displayed about 14 mm above theposition in which the next stimulus was then displayed after an RSIof 1500 or 100 ms, varied between blocks (as described below). Thecue word, and hence the task, changed every three trials. Theposition associated with a task switch in that block was additionallyindicated by the corresponding limb of the inverted Y being a thickerbar. Hence task switches were predictable (every three trials) and thetask was in addition redundantly indicated by the cue word and itslocation. The position of the switch trial was counterbalanced foreach participant (between blocks). Each stimulus remained on thescreen until the subject responded. If the participant made anincorrect response, a beep of 200 Hz sounded for 200 ms and the RSIwas extended by 2000 ms.

`Left' or `right' responses were made with the index and middlengers of the dominant hand. For the Word task, stimuli werecomposed of a word (`left' or `right') inside a shape (left arrow, rightarrow or rectangle). For the Arrow task, stimuli were composed of either a left or right arrow shape surrounding a letter string (`LEFT',`RIGHT' or `XXX'). For each task, each stimulus was used once in ablock in each run position in a random sequence, resulting in threeequally frequent congruency conditions: congruent (e.g. a left arrowwith the word `left'), incongruent (e.g. a left arrow with the word`right') and neutral (the shape around the word was a rectangle or theletter string was `XXX', so that the irrelevant attribute wasassociated with no left or right response).

Each block contained 36 experimental trials, one for eachcombination of task (Word, Arrow), run position (1, 2 or 3),response (left, right) and congruency (incongruent, neutral orcongruent). In addition, there were one, two or three warm-up trialsat the beginning of each block. Subjects were encouraged tominimize RT while avoiding errors. After each block, the computerdisplayed a feedback graph of mean RT and error rate so the subjectcould track performance relative to prior blocks. An instructionscreen was then displayed for the next block indicating the identityof the rst task, e.g. `Word', and the RSI (short or long), andreminding the subject to respond as quickly and accurately aspossible.

Subjects were initially given practice in four single-task blocks,two for the word task alternating with two for the arrow task, eachpreceded by an instruction screen explaining the relevant responsemappings for that task. In the practice blocks, cues and stimuli weredisplayed in the successive positions indicated by the inverted Yframework, but without the thick line which later demarcated theswitch position. The RSI was 1000 ms. If the subject made an error, abeep was sounded and the RSI was extended by 2000 ms. There

followed a demonstration of the switching concept, followed by oneblock of practice switching tasks for the long (1500 ms) and then theshort (100 ms) RSI. The eight blocks of the experiment properconsisted of alternating blocks with long and short RSIs, startingwith the long. The thickened limb of the inverted Y was initiallyupright, and moved one position clockwise after each block.

Analysis of performanceMean correct RTs and error rates were computed for each cell (task Q run position Q RSI Q congruence) excluding practice trials,warm-up trials, trials following an error on either of the precedingtwo trials, and RTs 4000 ms. Inspection of plots of RT ateach run position for each group (Fig. 1A) showed that reduction inRT for positions two and three followed roughly parallel lines foreach group [a test of the interaction of run position (for positions 2and 3) and group was non-signicant; F (2,53) < 1]. Therefore,switch costs for RT and errors were calculated by subtracting theaverage of trials at run positions two and three from the average of trials at run position one for each combination of RSI and

congruency (Table 2). Although patients responded more slowlyoverall than controls, this was not a signicant difference [ F (2,53) =1.7], nor was there a reliable difference at long RSI (when patientsappeared particularly slow) [ F (2,53) = 2.7, P > 0.05]. Although itthus appeared that patient and control groups could be contrasted forthe switching measures, without radical correction for overallslowing, conclusions were checked with respect to proportionalmeasures for crucial effects. Planned pairwise comparisons of controls, RF and LF patients were performed for the relevantmeasures (signicance threshold, P < 0.05, two-tailed). Reportedanalyses were no different from those performed with age as acovariate.

Results Behavioural resultsControl subjectsRT and error rates for control subjects were much asexpected. Comparing short and long RSI, controls achieveda substantial reduction (from 153 to 39 ms) in the RT switchcost (a 74% reduction in absolute cost, a 57% reduction incost expressed as a proportion of non-switch baseline) withpreparation (Fig. 3A). Most errors occurred on incongruentstimuli (Table 2), but switching caused only a modestincrease in error rate, and this was true at both short and longRSI (Fig. 3B, C).

Short RSI The difference between the LF and the RF groups wasapparent largely at the short RSI, when there was almost notime to recover/prepare between each response and the nextstimulus (Fig. 3B). Here the LF group, though not particularlyslow on non-switch trials (Table 2) was very slow on switchtrials, showing the largest RT cost both for absolute andproportionate switch cost. For absolute cost (SC short RSI ), theLF group had a signicant decit compared with controls[F (1,35) = 7.4, P < 0.01] and a marginally signicant decit



6/13

compared with the RF group [ F (1,34) = 3.0, P = 0.09]. Forproportional cost, the LF group had a signicant decitcompared with controls [ F (1,35) = 6.5, P < 0.05], and amarginally signicant decit compared with the RF group

[F (1,34) = 3.4, P = 0.07]. Although abnormally slow onswitch trials, the LF group was not particularly error prone onswitch trials; the error switch cost was about the same as forthe control group (Fig. 3AE, lower row).

Fig. 3 Mean (and standard error) of the correct RT and error switch cost variables for controls (light grey), left frontals (darker grey) andright frontals (black). Numbers above bars represent proportional measures in percentages. The top row shows RT indices; the bottom rowshows proportion of errors. ( A) SC REDUCT (switch cost at short RSI minus switch cost at long RSI). ( B) SC shortRSI (switch cost at shortRSI); the proportional measure is SC shortRSI /non-switch_short_RSI. ( C ) SC RESID (residual switch cost: switch cost at long RSI); theproportional measure is SC shortRSI /non-switch_long_RSI. ( D) SC CN (switch cost for congruent trials minus switch cost for neutral trials).(E ) SC ICC, shortRSI (switch cost for incongruent trials minus switch cost for congruent trials at short RSI only).

Table 2 Arithmetic means of RT data and proportion of errors for the task-set switching experiment

CT LF RFNS SW SC NS SW SC NS SW SC

Reaction timeShort RSI

Incongruent 1205.4 1379.6 174.3 1279.4 1575.2 295.8 1258.8 1526.8 268Congruent 1043.3 1197.7 154.4 1064.5 1399.6 335.1 1101.9 1287.7 185.7Neutral 1002.7 1132.9 130.2 1011.8 1327.9 316.1 1125.6 1295.6 170.1Average 152.9 315.6 207.9

Long RSIIncongruent 738.7 782.9 44.2 958.3 1107.6 149.3 884.7 1001.8 117.1Congruent 600.7 620.3 19.5 737 849.3 112.3 731.8 814.1 82.3Neutral 584.2 637.5 53.3 719.9 820.2 100.3 726.7 894.9 168.2Average 39 120.6 122.5

Error rateShort RSI

Incongruent 0.103 0.138 0.035 0.168 0.216 0.047 0.152 0.302 0.149Congruent 0.009 0.017 0.009 0.009 0.015 0.005 0.013 0.017 0.004Neutral 0.004 0.009 0.005 0.008 0.009 0.001 0.012 0.042 0.029

Average 0.016 0.018 0.061Long RSIIncongruent 0.109 0.151 0.042 0.164 0.175 0.012 0.161 0.216 0.055Congruent 0.004 0.008 0.004 0.002 0.02 0.017 0.008 0.01 0.002Neutral 0.013 0.006 0.007 0.012 0.015 0.003 0.014 0.017 0.004Average 0.013 0.011 0.02

NS = non-switch; SW = switch; SC = switch cost (SW NS).



7/13

At short RSI, the RF group, though showing somewhatgreater (but not reliably so) RT costs than controls [ F (1,37) =1.7, n.s.], exhibited a dramatically elevated error rate onswitch trials and hence a large switch cost in error rate(Fig. 3B). This error decit was reliable compared with both

controls [ F (1,37) = 6.3, P < 0.05] and the LF group [ F (1,34) =5.2, P < 0.05]. This resulted primarily from incongruentstimuli; the RF group made errors on 30% of incongruentstimuli on switch trails, twice as many as on non-switch trials(Table 2). By contrast, both controls and the LF group, whilemaking the majority of their errors on incongruent stimuli,showed only a modest (and similar) increase in these errors onswitch trials (Table 2). The unusual amplication of theswitch cost for incongruent stimuli in the RF group wasparticularly severe at short RSI, as captured by the contrastSC ICC(short RSI) (Fig. 3E). For the error measure, this wassignicantly greater than for controls [ F (1,37) = 5.4, P SC longRSI . Further inspection of the raw data(Supplementary Table 2) suggested that subjects with thegreatest POp damage made little improvement across runpositions at the short RSI, but did improve for positions 2 and3, relative to position 1, at long RSI. This pattern of results isdiscussed below.

LF groupFor this group, there were no reliable correlations for any of the switch cost indices (for RT or errors) for any ROI, just asthere were no reliable correlations with SSRT in the study by

Aron et al . (2003). The only measures correlated with damageto a ROI were the CN and (marginally) ICC contrasts (fornon-switch trials), indicative of task-set cueing and Stroop-like interference. However, as scatter plots showed thesecorrelations to derive largely from one patient, we re-rancorrelations for both these indices, at both RSIs, excludingthis outlier. There were a number of reliable correlationsbetween these indices and MFG damage at the strict alphalevel (Fig. 5). MFG damage correlated reliably with ICC longRSI (r = 0.68, P < 0.007) and with CN shortRSI (r = 0.75, P< 0.002). Therefore, the greater the MFG damage in the LFgroup, the greater was the tendency for stimuli to activate thecompeting task and the incongruent response even on non-

switch trials, suggesting weaker endogenous task-set control.

ValidationTo verify that POp damage in the RF group was critical forthe behavioural measures, we divided the RF group into thosewith POp damage (11 patients) and those without (sevenpatients); one RF patient had no MRI scan. There weresignicant differences ( P < 0.05, one-tailed) for SC REDUCT ,and SC CN for RT, and SC RESID and SC ICC(short RSI) forerrors. There were marginally signicant effects on severalother measures (Supplementary Table 3). This conrms that

the presence of POp damage is critical for producingbehavioural decits for many of the measures used here.

DiscussionThirty-six patients with unilateral lesions to the right or leftPFC were assessed on a test of task-set switching. Aneuroradiological ROI method was used to correlate per-formance on the switch measures with extent of damage tospecic subregions. As a group, RF patients had particulardifculty suppressing Stroop-like interference from the just-performed task on switch trials (but not post-switch trials) atshort RSIs. In contrast, as a group, the LF patients showedabnormally slow, but not error-prone, switching at short RSIs.Both LF and RF groups showed exaggerated time costs of aswitch at the long RSI (large residual costs). There may be anumber of reasons for this, but there is evidence that thedecit is different for the LF and RF groups. First, althoughthe LF group did not show the RF pattern of high error rateson incongruent stimuli on short-RSI switch trials, theyshowed greater ICC interference overall, which correlatedreliably with MFG damage. Secondly, there was an elevatedCN effect for the LF group, which correlated with MFGdamage, and this was increased on switch trials. This patternof data is compatible with the LF group exerting weaker

Fig. 5 The greater the damage to the left middle frontal gyrus(MFG), the weaker the endogenous task control. Correlations areshown for the degree of MFG damage against measures of endogenous task control such as CN and IC at short and longRSI. ( A) CN shortRSI (r = 0.68). ( B) CN longRSI . (C ) ICC shortRSI .(D) ICC longRSI (r = 0.75).



10/13

endogenous control consequent upon damage to MFG,resulting in a greater inuence on task-set activation of stimuli associated with the competing task set. This extendsthe specicity of prior neuropsychological data on left PFCcontributions to task-set reconguration. By contrast, the

residual switch cost of the RF group was reliably correlatedwith more specic POp damage, was the measure of responseinhibition (SSRT). The residual switch cost and SSRT werethemselves reliably correlated. It is plausible therefore thatright POp damage disrupts an inhibitory mechanism respon-sible for suppression of inappropriate responses and/or task sets in both stop-signal (no-go) and task-switching contexts.These results have implications for understanding: (i) theseparate contributions of the left and right frontal cortex toexecutive control, (ii) the specic role of the right POp, (iii)the fractionation of task-switching into component mechan-isms, and (iv) the neuropsychological sequelae of frontalcortical damage.

Components of task-setA person's task-set at any particular moment results from aninteraction of task-set inertia, exogenous task-set activationand endogenous control input. Task-set inertia arises from thepersistence of activation/inhibition from the previous trial(s)(Allport and Wylie, 1999; Yeung and Monsell, 2003), andleads to biasing and/or interference on the current trial,especially when it is a switch trial. Exogenous task-setactivation is generated by the stimulus itself because stimuliactivate task-sets associated with them (Lhermitte, 1983;Monsell et al ., 2001), and this is especially so on switch trials(Rogers and Monsell, 1995). To overcome both the inertia ona switch trial and the inappropriately activated task-set,endogenous control is required. Endogenous control consistsin top-down input, which biases a task-set by directingattention to a particular attribute, selecting a SR rule, and soforth (Norman and Shallice, 1986; Gilbert and Shallice, 2002;Yeung and Monsell, 2003). However, it cannot be the casethat endogenous control completely suppresses the inappro-priate task-set because we usually see ICC interference dueto activation of the inappropriate response tendency.Furthermore, several authors have documented that prepar-ation (endogenous input) can substantially reduce switchcosts without reducing ICC interference (Rogers andMonsell, 1995; Meiran, 2000; Meiran et al ., 2000; Monsellet al ., 2001), though other studies have seen such a reduction(e.g. Goschke, 2000). Therefore, effective imposition of endogenous control does not prevent some activation of responses afforded by the current stimulus; arguably thisactivation can only be dealt with reactively, upon detection of conict (Monsell et al ., 2003).

Right frontal cortex and task-switchingThe present data suggest that the right PFC may be crucial tothis reactive suppression of inappropriately activated re-

sponses, especially when the task set is weakly established(i.e. at short RSI). On this account, damage to the right PFCcompromises this function, thus producing a striking switch-ing decit. For error rate, the RF group had a signicantlygreater switch cost for incongruent than congruent trials

(SC ICC(short RSI) ) than both controls and the LF group (and asignicant effect compared with the LF group for RT). ForRT, at short RSI, the RF group had a larger (although non-signicant) switch cost than controls, while at long RSI thiswas a reliable switch decit.

One explanation for this pattern of results, consistent withour theoretical framework concerning inhibitory mechanismsin task-switching, is that, at short RSI, when endogenouscontrol is not strongly established, difculty in reactivelysuppressing inappropriate responses (or task-sets) means thatthe RF group frequently makes the wrong response. This mayalso explain the fact that those subjects with the greatest POpdamage had larger switch costs at long than at short RSI. At

short RSI, the patients with greatest right POp damage hadinated RTs at position 1 of the run (the switch position) andthey did not recover much at positions 2 and 3 of the run (non-switch positions), so the switch cost was small. However, atlong RSI, we posit that they could recover much better atpositions 2 and 3 of the run (relative to position 1) becauseendogenous control was stronger and could help compensatefor weak reactive suppression; hence the switch cost wassubstantial.

Although this interpretation of the pattern of RF switch-costs is speculative [especially considering the multipletheoretical frameworks surrounding task-switching (for re-view see Monsell, 2003), as well as the multiple cognitivecomponents that doubtless interact to effect a task-switch],we nd it highly plausible in accounting for at least part of thedecit. This is because the RF cortex, and the POp region inparticular, has been repeatedly implicated as a focus forinhibitory mechanisms (for review see Aron et al ., 2004).With respect to the present study, we note that SC RESID wassignicantly correlated with damage to POp, as well as withSSRT, a measure of response inhibition. The nding that theamount of damage to POp affected measures from twoindependent tasks increases condence that POp implementsan inhibitory mechanism required for suppressing responsesand/or task sets. The requirement to inhibit a prepotentresponse upon presentation of a stop-signal and the require-ment to inhibit activation of an irrelevant response by abivalent-stimulus in a task-switching experiment have strongprima facie similarity.

A plausible neural model for the generation of thesuppression effect is that once the anterior cingulate cortex(ACC) detects conict in a task-setting, the right POp isrecruited to inhibit the irrelevant response activation (Gehringand Knight, 2000). That neither the error nor the RT measureof SC I-C(short RSI) correlated reliably with damage to the rightPOp (although the correlations were in the right direction)may be because these measures had much less of a range(hence less variability) than SC RESID . Independent evidence



11/13

from the Eriksen-Flanker task (effectively an ICC compari-son) indicates signicant right IFG/POp activation change infMRI experiments (Hazeltine et al ., 2000; Bunge et al .,2002).

The present results clearly show that RF damage can

indeed lead to a switching decit. It is possible that the failureof prior neuropsychological studies (Rogers et al ., 1998;Bedard and Richer, 1999; Mecklinger et al ., 1999; Keele andRafal, 2000) to detect switch decits in RF patients may havebeen due to under-representation within those samples of IFGdamage, or damage to other foci critical for switching(Rushworth et al ., 2002). It is also noteworthy that the RFdecit in the present study was engendered by specicconditions (particularly the congruency manipulation) whichwere not always present in prior studies.

Left frontal cortex and task-switchingIn accordance with three prior lesion studies of task-setswitching (Rogers et al ., 1998; Mecklinger et al ., 1999; Keeleand Rafal, 2000) and one using patients with closed headdamage (Stablum et al ., 1994), we found that LF damageproduced a substantial increase in switch costs. Unlike the RFgroup, the LF group did not have particular difcultysuppressing the inappropriate response on switch trialsunder speed stress (i.e. short RSIs). Instead, they showed amore general difculty in imposing the appropriate task set,as indexed by larger switch costs at both short and long RSIs,a larger interference effect (IC) on all trials, and somesuggestion of greater activation of the competing task set (theCN contrasts). Although the correlations with extent of damage were generally weaker than for the RF patients, therewas a reliable correlation between the ICC contrast andMFG damage (at least at the long RSI) and between the CNeffect and MFG damage (at least at the short RSI). Thesendings are consistent with weaker endogenous control of task-set in the LF patients leading to amplication of theinuence of exogenous activation of task-set.

Mecklinger et al . (1999) found that that those LF patientswith the greatest switching difculty also had speech andlanguage disorders, and argued for a specically linguisticcomponent of endogenous control (Goschke, 2000) associ-ated with left PFC. However, there is also evidence suggest-ive of a specic role for the left MFG in implementing andmaintaining a task-set. Garavan et al . (2002) showed that theamount of activation in the left MFG (Brodmann area 9)preceding a no-go trial predicted a correct stop or an incorrectcommission error, thus indicating that greater activation (agreater role) for the left PFC meant the subject was bettercapable of keeping in mind the task-set [also see MacDonaldet al ., 2000, who also reported left MFG (Brodmann area 9)activity during task preparation]. Further research byM. Funnell and H. Garavan (submitted for publication) in asplit-brain patient showed that, in a go/no go task and twostop-signal tasks, the right hemisphere was better able toinhibit motor responses than the left. Conversely, the left

hemisphere was better able to stay on task when the task requirements were complicated. Consistent with the presentstudy, these results suggest that the left hemisphere is moreimportant for selection and maintenance of a task-set, whilethe right hemisphere is more important for the implementa-

tion of inhibitory control.

ConclusionsThe results suggest that a subregion of the right PFC, the parsopercularis (POp) of the IFG, plays an inhibitory role relatedto inhibition of responses and/or task-sets. The criticalimportance of this region for both response inhibition andswitching revealed in the present cohort conrms twoneuroimaging studies directly comparing response inhibitionwith switching (Konishi et al ., 1999; Swainson et al ., 2003),both of which found activity in the right IFG common to both

tasks. The IFG/POp has been repeatedly activated in a widerange of neuroimaging experiments (for review see Duncanand Owen, 2000) and in studies specically investigatingresponse inhibition (e.g. Garavan et al ., 1999; de Zubicarayet al ., 2000; Menon et al ., 2001; Bunge et al ., 2002; Garavanet al ., 2002; Aron et al ., 2003; Rubia et al ., 2003) andswitching/reversing/shifting (e.g. Dove et al ., 2000; Monchiet al ., 2001; Nagahama et al ., 2001; Cools et al ., 2002; Dreherand Berman, 2002; Nakahara et al ., 2002; Brass et al ., 2003)(for review see Aron et al ., 2004). The IFG/POp is one of themost heavily connected regions of the PFC (Miller andCohen, 2001), is one of the last to develop in both ontogenyand phylogeny (Pandya and Barnes, 1987) and is frequentlyimplicated in neuropsychiatric syndromes such as attentiondecit/hyperactivity disorder (Rubia et al ., 1999). Thepresent characterization of the POp as being critical forsuppression of inappropriate responses and/or task sets underconditions of frequent switches between tasks goes some wayto clarifying its function (which may extend across multipletask domains). Future research should probe this possibility,as well as the interaction between POp, ACC and other PFCregions (Gehring and Knight, 2000; Garavan et al ., 2002). Bycontrast, the left frontal cortex appears to be required not forthe reactive suppression of inappropriate responses, butinstead for the implementation of the endogenous control of the task-set (MacDonald et al ., 2000; Garavan et al ., 2002).While such top-down control has been modelled extensivelyfor the task-switching paradigm in formal/computationalterms (Meiran, 2000; Gilbert and Shallice, 2002; Yeung andMonsell, 2003), a neurobiological account is lacking. Thecurrent nding that the extent of damage to the left MFGcorrelates reliably with measures of such top-down controlprovides relevant data for such an account.

Task-switching requires multiple cognitive components.This study indicates the localization of two suchcomponentsthe reactive suppression of task-sets and/orresponses and top-down control of task-setto the right IFG/ POp and left MFG respectively.



12/13

AcknowledgementsThe authors wish to thank the managers and subjects of theCCNRP, Cambridge, and C. Rorden for helpful technicaladvice. The task-switching paradigm was developed under aStroke Association Grant to S. M. and I. Robertson. This

work was supported by an MRC studentship to A. R. A. and aProgram Grant from the Wellcome Trust (019408) and wascarried out at the MRC (Cambridge) Centre for Behaviouraland Clinical Neuroscience.

ReferencesAllport A, Wylie G. Task-switching: positive and negative priming of task-

set. In: Humphreys GW, Duncan J, Treisman A, editors. Attention, space,and action: studies in cognitive neuroscience. Oxford: Oxford UniversityPress; 1999. p. 27396

Allport DA, Styles EA, Hsieh S. Shifting intentional set: exploring thedynamic control of tasks. In: Umilta C, Moscovitch M, editors. Attentionand performance XV. Cambridge (MA): MIT Press; 1994. p. 42152.

Arbuthnott K, Frank J. Executive control in set switching: residual switchcost and task-set inhibition. Can J Exp Psychol 2000; 54: 3341.

Aron AR, Fletcher PC, Bullmore ET, Sahakian BJ, Robbins TW. Stop-signalinhibition disrupted by damage to right inferior frontal gyrus in humans.Nat Neurosci 2003; 6: 1156.

Aron AR, Robbins TW, Poldrack RA. Inhibition and the right inferior frontalcortex. Trends Cogn Sci 2004; 8: 1707

Bedard S, Richer F. Is there really a task-shifting decit after frontal lesion?Brain Cogn 1999; 40: 414.

Brass M, Ruge H, Meivan N, Rubin O, Koch I, Zysset S, et al. When thesame response has different meanings: recoding the response meaning inthe lateral prefrontal cortex. Neuroimage 2003; 20: 102631.

Brett M, Leff AP, Rorden C, Ashburner J. Spatial normalization of brainimages with focal lesions using cost function masking. Neuroimage 2001;14: 486500.

Bunge SA, Dudukovic NM, Thomason ME, Vaidya CJ, Gabrieli JDE.

Immature frontal lobe contributions to cognitive control in children:evidence from fMRI. Neuron 2002; 33: 30111.

Cools R, Clark L, Owen AM, Robbins TW. Dening the neural mechanismsof probabilistic reversal learning using event-related functional magneticresonance imaging. J Neurosci 2002; 22: 45637.

De Jong R. An intentionactivation account of residual switch costs. In:Monsell S, Driver J, editors. Control of cognitive processes: Attention andperformance XVIII. Cambridge (MA): MIT Press; 2000. p. 35776.

de Zubicaray GI, Andrew C, Zelaya FO, Williams SC, Dumanoir C. Motorresponse suppression and the prepotent tendency to respond: a parametricfMRI study. Neuropsychologia 2000; 38: 128091.

Dove A, Pollmann S, Schubert T, Wiggins CJ, von Cramon DY. Prefrontalcortex activation in task switching: an event-related fMRI study. BrainRes Cognit Brain Res 2000; 9: 1039.

Dreher JC, Berman KF. Fractionating the neural substrate of cognitive

control processes. Proc Natl Acad Sci USA 2002; 99: 14595600.Duncan J, Owen AM. Common regions of the human frontal lobe recruitedby diverse cognitive demands. Trends Neurosci 2000; 23: 47583.

Garavan H, Ross TJ, Stein EA. Right hemispheric dominance of inhibitorycontrol: an event-related functional MRI study. Proc Natl Acad Sci USA1999; 96: 83016.

Garavan H, Ross TJ, Murphy K, Roche RA, Stein EA. Dissociable executivefunctions in the dynamic control of behavior: inhibition, error detection,and correction. Neuroimage 2002; 17: 18209.

Gehring WJ, Knight RT. Prefrontal-cingulate interactions in actionmonitoring. Nat Neurosci 2000; 3: 51620.

Gilbert SJ, Shallice T. Task switching: a PDP model. Cognit Psychol 2002;44: 297337.

Goschke T. Intentional reconguration and involuntary persistence in task set switching. In: Monsell S, Driver J, editors. Control of cognitive

processes: Attention and performance XVIII. Cambridge (MA): MITPress; 2000. p. 33155

Hazeltine E, Poldrack R, Gabrieli JDE. Neural activation during responsecompetition. J Cogn Neurosci 2000; 12 Suppl 2: 11829.

Keele SW, Rafal R. Decits in task set in patients with left prefrontal cortexlesions. In: Monsell S, Driver J, editors. Control of cognitive processes:Attention and performance XVIII. Cambridge (MA): MIT Press; 2000.p. 62751.

Konishi S, Nakajima K, Uchida I, Kameyama M, Nakahara K, Sekihara K,et al. Transient activation of inferior prefrontal cortex during cognitive setshifting. Nat Neurosci 1998; 1: 804.

Konishi S, Nakajima K, Uchida I, Kikyo H, Kameyama M, Miyashita Y.Common inhibitory mechanism in human inferior prefrontal cortexrevealed by event-related fMRI. Brain 1999; 122: 98191.

Lhermitte F. `Utilization behaviour' and its relation to lesions of the frontallobes. Brain 1983; 106: 23755.

Logan GD, Cowan WB. On the ability to inhibit thought and action: a theoryof an act of control. Psychol Rev 1984; 91: 295327.

MacDonald AW, Cohen JD, Stenger VA, Carter CS. Dissociating the role of the dorsolateral prefrontal and anterior cingulate cortex in cognitivecontrol. Science 2000; 288: 18358.

Mayr U. Inhibition of action rules. Psychon Bull Rev 2002; 9: 939.Mayr U, Keele S. Changing internal constraints on action: the role of

backward inhibition. J Exp Psychol Gen 2000; 129: 426.Mecklinger AD, von Cramon DY, Springer A, Matthes-von Cramon G.

Executive control functions in task switching: evidence from brain injuredpatients. J Clin Exp Neuropsychol 1999; 21: 60619.

Meiran N. Reconguration of processing mode prior to task performance. JExp Psychol Learn Mem Cognit 1996; 22: 142342.

Meiran N. Modeling cognitive control in task-switching. Psychol Res 2000;63: 23449.

Meiran N, Chorev Z, Sapir A. Component processes in task switching.Cognit Psychol 2000; 41: 21153.

Menon V, Adleman NE, White CD, Glover GH, Reiss AL. Error-relatedbrain activation during a Go/NoGo response inhibition task. Hum BrainMapp 2001; 12: 13143.

Meuter RFI, Allport DA. Bilingual language switching in naming:

asymmetrical costs of language selection. J Mem Lang 1999; 40: 2540.Miller EK, Cohen JD. An integrative theory of prefrontal cortex function.

Annu Rev Neurosci 2001; 24: 167202.Monchi O, Petrides M, Petre V, Worsley K, Dagher A. Wisconsin Card

Sorting revisited: distinct neural circuits participating in different stages of the task identied by event-related functional magnetic resonanceimaging. J Neurosci 2001; 21: 773341.

Monsell S. Task switching. Trends Cogn Sci 2003; 7: 13440.Monsell S, Taylor TJ, Murphy K. Naming the color of a word: is it responses

or task sets that compete? Mem Cognit 2001; 29: 13751.Monsell S, Sumner P, Waters H. Task-set reconguration with predictable

and unpredictable task switches. Mem Cognit 2003; 31: 32742.Nagahama Y, Okada T, Katsumi Y, Hayashi T, Yamauchi H, Oyanagi C,

et al. Dissociable mechanisms of attentional control within the humanprefrontal cortex. Cereb Cortex 2001; 11: 8592.

Nakahara K, Hayashi T, Konishi S, Miyashita Y. Functional MRI of macaque monkeys performing a cognitive set-shifting task. Science 2002;295: 15326.

Norman D, Shallice T. Attention to action: willed and automatic control of behavior. In: Davidson R, Schwartz G, Shapiro D, editors. Consciousnessand self regulation: Advances in research and theory, Vol. 4. New York:Plenum; 1986. p. 118.

Pandya DN, Barnes CL. Architecture and connections of the frontal lobe. In:Perecman E, editor. The frontal lobes revisited. Hillsdale (NJ): LawrenceErlbaum; 1987. p. 4168.

Rogers R, Monsell S. Costs of a predictable switch between simple cognitivetasks. J Exp Psychol Gen 1995; 124: 20731.

Rogers RD, Sahakian BJ, Hodges JR, Polkey CE, Kennard C, Robbins TW.Dissociating executive mechanisms of task control following frontal lobedamage and Parkinson's disease. Brain 1998; 121: 81542.



13/13

Rubia K, Overmeyer S, Taylor E, Brammer M, Williams SC, Simmons A,et al. Hypofrontality in attention decit hyperactivity disorder duringhigher-order motor control: a study with functional MRI. Am J Psychiatry1999; 156: 8916.

Rubia K, Smith AB, Brammer MJ, Taylor E. Right inferior prefrontal cortexmediates response inhibition while mesial prefrontal cortex is responsiblefor error detection. Neuroimage 2003; 20: 3518.

Rubinstein J, Meyer DE, Evans JE. Executive control of cognitive processesin task switching. J Exp Psychol Hum Percept Perform 2001; 27: 76397.

Rushworth MFS, Hadland KA, Paus T, Sipila PK. Role of the human medialfrontal cortex in task switching: a combined fMRI and TMS study. JNeurophysiol 2002; 87: 257792.

Schuch S, Koch I. The role of response selection for inhibition of task sets intask shifting. J Exp Psychol Hum Percept Perform 2003; 28: 192201.

Sohn MH, Ursu S, Anderson JR, Stenger VA, Carter CS. Inaugural article:the role of prefrontal cortex and posterior parietal cortex in task switching.Proc Natl Acad Sci USA 2000; 97: 1344853.

Stablum F, Leonardi G, Mazzoldi M, Umilta C, Morra S. Attention andcontrol decits following closed-head injury. Cortex 1994; 30: 60318.

Swainson R, Cunnington R, Jackson GM, Rorden C, Peters A, Morris PG,et al. Cognitive control mechanisms revealed by ERP and fMRI: evidencefrom repeated task-set switching. J Cogn Neurosci 2003; 15: 78599.

Tzourio-Mazoyer N, Landeau B, Papathanassiou D, Crivello F, Etard O,

Delcroix N, et al. Automated anatomical labeling of activations in SPMusing a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 2002; 15: 27389.

Williams EJ. The comparison of regression variables. J Roy Stat Soc (SeriesB) 1959; 21: 3969.

Wylie G, Allport A. Task switching and the measurement of `switch costs'.Psychol Res 2000; 63: 21233.

Yeung N, Monsell S. Switching between tasks of unequal familiarity: therole of stimulus-attribute and response-set selection. J Exp Psychol HumPercept Perform 2003; 29: 45569.


a componential analysis of task‐switching deficits associated with lesions of left and right...

Documents