Response monitoring repetitive behaviourand anterior cingulate abnormalities in ASDKatharine NThakkar1 Frida E Polli12 Robert M Joseph3 David S Tuch456 Nouchine Hadjikhani456

Jason JS Barton78 and Dara S Manoach156

1Department of Psychiatry Massachusetts General Hospital Harvard Medical School Boston MA 022152Department of Psychology Suffolk University Boston MA 02114 3Department of Anatomy and Neurobiology BostonUniversity Medical School Boston MA 4Department of Radiology Massachusetts General Hospital 5Athinoula A MartinosCenter for Biomedical Imaging Charlestown MA 02129 6Harvard Medical School Boston MA 02215 7Department ofNeurology and 8Department of Ophthalmology and Visual Sciences University of British Columbia Vancouver BC Canada

Correspondence to Dara S Manoach 149 13th St Room 2608 Charlestown MA 02129 USAE-mail daranmrmghharvardedu

Autism spectrum disorders (ASD) are characterized by inflexible and repetitive behaviour Response moni-toring involves evaluating the consequences of behaviour and making adjustments to optimize outcomesDeficiencies in this function and abnormalities in the anterior cingulate cortex (ACC) on which it relies havebeen reported as contributing factors to autistic disordersWe investigated whether ACC structure and func-tion during response monitoring were associated with repetitive behaviour in ASD We compared ACC activa-tion to correct and erroneous antisaccades using rapid presentation event-related functional MRI in 14 controland ten ASD participants Because response monitoring is the product of coordinated activity in ACCnetworks we also examined the microstructural integrity of the white matter (WM) underlying this brainregion using diffusion tensor imaging (DTI) measures of fractional anisotropy (FA) in 12 control and 12 adultASD participants ACC activation and FAwere examined in relation to Autism Diagnostic Interview-Revisedratings of restricted and repetitive behaviour Relative to controls ASD participants (i) made more antisac-cade errors and responded more quickly on correct trials (ii) showed reduced discrimination between errorand correct responses in rostral ACC (rACC) which was primarily due to (iii) abnormally increased activationon correct trials and (iv) showed reduced FA in WM underlying ACC Finally in ASD (v) increased activationon correct trials and reduced FA in rACC WM were related to higher ratings of repetitive behaviour Thesefindings demonstrate functional and structural abnormalities of the ACC in ASD that may contribute torepetitive behaviour rACC activity following errors is thought to reflect affective appraisal of the error Thusthe hyperactive rACC response to correct trials can be interpreted as a misleading affective signal thatsomething is awry which may trigger repetitive attempts at correction Another possible consequence ofreduced affective discrimination between error and correct responses is that it might interfere with thereinforcement of responses that optimize outcomes Furthermore dysconnection of the ACC as suggestedby reduced FA to regions involved in behavioural control might impair on-line modulations of responsespeed to optimize performance (ie speed-accuracy trade-off) and increase error likelihood These findingssuggest that in ASD structural and functional abnormalities of the ACC compromise response monitor-ing and thereby contribute to behaviour that is rigid and repetitive rather than flexible and responsive tocontingencies Illuminating the mechanisms and clinical significance of abnormal response monitoring inASD represents a fruitful avenue for further research

Keywords autism anterior cingulate cortex response monitoring functional MRI diffusion tensor imaging

Abbreviations ACC=anterior cingulate cortex ADI-R=autism diagnostic interview-revised ASD=autism spectrumdisorders BOLD=blood oxygen level dependent CWP=cluster-wise probability dACC=dorsal ACC DTI=diffusiontensor imaging ERN=error-related negativity HC=healthy control IQ= intelligence quotient MPRAGE=magnetizationprepared rapid gradient echo OCD=obsessive^ compulsive disorder rACC=rostral ACC TE=echo time TR=repetitiontime WM=white matter

Received December 19 2007 Revised March 14 2008 Accepted April 30 2008

doi101093brainawn099 Brain (2008) Page 1 of 15

2008 The Author(s)This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (httpcreativecommonsorglicensesby-nc20uk) whichpermits unrestricted non-commercial use distribution and reproduction in any medium provided the original work is properly cited

Brain Advance Access published June 11 2008

IntroductionAutism spectrum disorders (ASD) are neurodevelopmentaldisorders characterized by the early onset of markedimpairments in socialization and communication alongwith restricted and repetitive interests and activities Thereis growing evidence that executive function deficitscontribute to these core symptoms (Hill 2004 Lopezet al 2005 South et al 2007) Executive function refers todiverse cognitive abilities involved in behavioural controlincluding response monitoringmdashevaluating whether theconsequences of behaviour are consistent with its intentand instituting adjustments to optimize outcomesResponse monitoring is thought to rely on anteriorcingulate cortex (ACC) on the basis of extensive evidencefrom functional neuroimaging electrophysiological lesionand intracranial recording studies (see Taylor et al 2007for a recent review) There is evidence of abnormalresponse monitoring in autism including increased latency(and amplitude in a subset) of the error-related negativity(ERN) (Henderson et al 2006) an event-related potentialthat follows errors and is thought to be generated by theACC (Gehring et al 1993 Dehaene et al 1994 van Veenand Carter 2002) reduced error self-correction (Russelland Jarrold 1998) and reduced post-error slowing acompensatory mechanism to improve performance on thesubsequent trial (Bogte et al 2007) Since the evaluationof behaviour and its consequences is necessary to deter-mine whether or not current strategies should be main-tained abnormal response monitoring may contributeto behavioural rigidity in ASD The aims of the presentstudy were (i) to examine ACC function during responsemonitoring using functional MRI (fMRI) (ii) to assessthe microstructural integrity of the white matter (WM)underlying the ACC using diffusion tensor imaging (DTI)and (iii) to investigate the relations of ACC functionand WM microstructure to restricted repetitive behaviourin ASDThere is growing evidence of both functional and structural

abnormalities of the ACC in ASD Abnormal ACC activationhas been observed during a range of cognitive tasks (Haznedaret al 1997 2000 Hall et al 2003 Gomot et al 2006 Kanaet al 2007 Kennedy et al 2006 Silk et al 2006 Ashwin et al2007 Dichter and Belger 2007) There is also evidence ofaltered metabolism (Levitt et al 2003) reduced 5-HT2Areceptor binding (Murphy et al 2006) and reduced volume ofthe ACC (Haznedar et al 1997 2000) A DTI study foundreduced fractional anisotropy (FAmdashan index of WM micro-structure Beaulieu 2002) of the WM adjacent to ACC(Barnea-Goraly et al 2004) Aberrant connectivity is alsosuggested by findings of decreased functional connectivity(synchronization of activation) between the ACC andother structures during a task requiring response inhibition(Kana et al 2007) These findings are consistent with thetheory of abnormal inter-regional coordination in ASD(Courchesne and Pierce 2005)

Abnormal ACC connectivity and function may contri-bute to the defining symptoms of autism ACC dysfunctionhas been linked to indices of social impairment (Haznedaret al 2000 Ohnishi et al 2000 Kennedy et al 2006) butwe are aware of only one study linking it to restricted andrepetitive behaviours in ASD (Shafritz et al 2007) In thisstudy restricted repetitive behaviours were negativelycorrelated with fMRI activation in ACC during trials thatrequired a switch in response Such a link has face validitysince the ACC contributes to evaluating and modifyingbehaviour based on its consequences and flexible behaviourdepends on these functions A link between abnormalACC function and repetitive thoughts and behaviours hasbeen more extensively documented in obsessivendashcompulsivedisorder (OCD) which is often comorbid with ASD (Leyferet al 2006) Relative to controls individuals with OCDshow increased fMRI activation of the ACC and increasedERN amplitude not only to error trials (Gehring et al2000 Johannes et al 2001 Fitzgerald et al 2005 but seeNieuwenhuis et al 2005 for a negative finding) but alsoto correct trials with high response-conflict (ie those thatrequire suppression of a competing response) in some(Ursu et al 2003 Maltby et al 2005) but not allstudies (Gehring et al 2000 Fitzgerald et al 2005) Theseexaggerated responses suggest hyperactive monitoring ofresponses Moreover both increased ERN to errors(Gehring et al 2000) and increased ACC activation toerror (Fitzgerald et al 2005) and correct trials (Ursu et al2003) have been associated with the severity of obses-sions and compulsions in OCD These relations havebeen interpreted to reflect excessive or inappropriate errorsignalling that triggers a compulsion to repeat behaviourseven if they have already been successfully executed (Maltbyet al 2005) The findings that in OCD the ACC andassociated regions show increased activation during symp-tom provocation (Breiter et al 1996) and that cingulot-omy relieves obsessions and compulsions (Dougherty et al2002) further support the link between ACC function andrigid repetitive behaviours Measurements of obsessivendashcompulsive behaviour have also been related to responsemonitoring in non-clinical samples Obsessive character-istics are related to the amplitudes of ERN and errorpositivity (another event-related potential that followserrors and is thought to be generated by the ACCFalkenstein et al 1995 van Veen and Carter 2002) inchildren (Santesso et al 2006) and to the amplitude ofERN in college undergraduates (Hajcak and Simons 2002)These findings suggest that abnormally increased activity ofthe ACC following response execution contributes to rigidrepetitive behaviour Based on this literature we expectedto find increased ACC activation on both error and correcttrials that is correlated with repetitive behaviour in ASD

We investigated the role of the ACC in response mon-itoring using an antisaccade paradigm Antisaccades requiresuppression of the prepotent response of looking towards asuddenly appearing visual stimulus and substitution of the

Page 2 of 15 Brain (2008) K NThakkar et al

novel behaviour of looking in the opposite direction(Hallett 1978) Antisaccades are well-suited for examiningresponse monitoring because they have a relatively well-delineated neuroanatomy and neurophysiology (for reviewsee Munoz and Everling 2004) generate robust electro-phyisological error markers ERN error positivity(Nieuwenhuis et al 2001) and ACC hemodynamicresponses (Polli et al 2005) and involve a high degree ofresponse conflict Moreover individuals with developmen-tal disorders affecting social processing including ASDconsistently show an increased rate of antisaccade errors(ie a failure to suppress the prepotent prosaccade)(Manoach et al 1997 2004 Minshew et al 1999Goldberg et al 2002 Luna et al 2007) which may reflectin part aberrant evaluation of performance and modifica-tion of prepotent stimulus-response mappingsWe examined group differences in ACC activation during

correct and erroneous antisaccade trials using rapidpresentation event-related fMRI Comparisons wererestricted to antisaccade trials since healthy and ASDparticipants produce too few prosaccade errors with thisparadigm (Manoach et al 2004) We examined activity inrostral and dorsal ACC (rACC dACC) separately giventheir differences in cytoarchitecture function and connec-tivity (Devinsky et al 1995 Bush et al 1998 Whalen et al1998) and distinct patterns of haemodynamic activityduring antisaccade error commission (Polli et al 2005)Since response monitoring is the product of coordinatedactivity between the ACC and functionally related regions(Holroyd and Coles 2002) and the ACC shows evidence ofdecreased structural and functional connectivity in autism(Barnea-Goraly et al 2004 Kana et al 2007) we alsoexamined the microstructural integrity of the WM under-lying the ACC using DTI We expected ASD participants toshow (i) an increased antisaccade error rate (ii) increasedACC activity during both correct and erroneous antisaccadetrials and (iii) reduced FA in the WM underlying ACCFinally we hypothesized that (iv) structural and functionalACC abnormalities would be associated with rigidrepetitive behaviour in ASD

MethodsParticipantsTwelve adults with ASD and 14 healthy control (HC) participantswere recruited by poster and website advertisements (see Table 1for demographic data) Participants with ASD were diagnosedwith autism (n= 8) Aspergerrsquos disorder (n= 2) or pervasivedevelopmental disorder not otherwise specified (n= 2) by anexperienced clinician (RMJ) on the basis of current presentationand developmental history as determined by medical recordreview and clinical interview Potential participants meetingDSM-IV criteria for co-morbid psychiatric conditions or substanceabuse were excluded ASD diagnoses were confirmed using theAutism Diagnostic Interview-Revised (ADI-R Rutter et al2003) and the Autism Diagnostic Observation Schedule Module4 (Lord et al 1999) administered by research reliable personnel

Individuals with known autism-related medical conditions (egFragile-X syndrome tuberous sclerosis) were not included Fourof the 12 ASD participants were taking the following medicationsfluoxetine and lithium buproprion and clonazepam citalopramand sertraline and methylphenidateHC participants were screened to exclude a history of autism or

any other neurological or psychiatric condition All participantswere screened to exclude substance abuse or dependence within thepreceding 6 months and any independent condition that mightaffect brain function ASD and control groups were matched forage sex handedness as measured by a laterality score on themodified Edinburgh Handedness Inventory (scores of 100 and+100 denote exclusive use of left or right hands respectively)(Oldfield 1971 White and Ashton 1976) parental socioeconomicstatus on the Hollingshead Index (Hollingshead 1965) years ofeducation and estimated verbal intelligence quotient (IQ) based ona test of single word reading (American National Adult ReadingTest Blair and Spreen 1989) All ASD participants had average orabove estimates of verbal (124 12 range 106ndash141) and non-verbal(120 10 range 100ndash138) IQ as measured by the WechslerAbbreviated Scale of Intelligence (Wechsler 1999) The studywas approved by the Partners Human Research CommitteeAll participants gave written informed consent after the experi-mental procedures had been fully explainedfMRI data were unusable for one ASD participant due to eye

tracking malfunction For a second ASD participant data were notacquired because scanner-safe glasses of the correct prescriptionwere not available fMRI analyses were conducted on theremaining 10 ASD and 14 control participants Saccadic latencydata for one control participant was omitted from analysis due toan acquisition problem DTI data were acquired on all 12 ASDparticipants and 12 of the 14 controls

Saccadic paradigmPrior to scanning the task was explained and participants practicedin a mock scanner until their performance indicated that theyunderstood the directions and were comfortable with the taskParticipants were instructed to respond as quickly and accurately aspossible and told that they would receive a bonus of 5 cents for eachcorrect response in addition to a base rate of pay Each run of the

Table 1 Means standard deviations and groupcomparisons of demographic data

Subjectcharacteristics

HC(n=14)

ASD(n=12)

T P

Age 278 3011 069 050Sex 8M6F 10M2F f=028 022Lateralityscore(Handedness)

75 45 6138 086 040

Parental SESa 131048 117039 z=060 042Years ofeducation

162 163 051 062

Estimatedverbal IQb

114 9 116 8 062 054

aA lower score denotes higher status The Phi value is theresult of a Fisherrsquos exact test The z-value is the result of anon-parametric Mann^Whitney U comparison bBased on theAmerican National Adult ReadingTest

Abnormal response monitoring in autism Brain (2008) Page 3 of 15

task consisted of a pseudorandom sequence of prosaccade andantisaccade trials that were balanced for right and left movementsFigure 1 provides a graphic depiction of the task and a descriptionof task parameters Randomly interleaved with the saccadic trialswere intervals of fixation lasting 2 4 or 6 s The fixation intervalsprovided a baseline and their variable length introduced lsquotemporaljitterrsquo which optimizes the analysis of rapid presentation event-related fMRI designs (Buckner et al 1998 Burock and Dale 2000Miezin et al 2000) The schedule of events was determined using atechnique to optimize the statistical efficiency of event-relateddesigns (Dale 1999) Participants performed six runs of the taskeach lasting 5min 22 s with short rests between runs The totalexperiment lasted about 40min and generated a total of 211prosaccade and 211 antisaccade trials and 80 fixation intervals

Stimulus display and eye trackingDisplays of the eye movement task were generated usingthe Vision Shell programming platform (wwwvisionshellcomWatertown MA USA) and back-projected with a Sharp XG-2000color LCD projector (Osaka Japan) onto a screen at the rear of thebore that was viewed by the participant via a mirror on the headcoil The ISCAN fMRI Remote Eye Tracking Laboratory (ISCANBurlington MA) recorded saccades during scanning This systemused a video camera mounted at the rear of the MRI bore

The camera imaged the eye of the participant via an opticalcombiner a 45 cold transmissive mirror that reflects an infraredimage of the eye with the infrared illumination being provided by anLED mounted on the head coil The system used passive opticalcomponents with no ferrous content within the bore to minimizeartefacts in the MRI images Eye position was sampled at a rate of60Hz Eye images were processed by ISCANrsquos RK-726PCI highresolution PupilCorneal reflection tracker located outside of theshielded MRI room Stimuli presented by Vision Shell were digitallyencoded and relayed to ISCAN as triggers that were inserted into theeye movement recordings

Scoring and analysis of eye movement dataEye movement data were scored in MATLAB (Mathworks NatickMA) using a partially automated programme that determined thedirectional accuracy of each saccade with respect to the requiredresponse and the latency from target onset Saccades wereidentified as horizontal eye movements with velocities exceeding47s The onset of a saccade was defined as the point at which thevelocity of the eye movement first exceeded 31s Only trials withsaccades in the desired direction and latencies over 130ms wereconsidered correct and only correct saccades were included inthe latency analyses The cutoff of 130ms excluded anticipatorysaccades which are executed too quickly to be a valid response to

latency

latency

right

left

A Prosaccade

B Antisaccade

(see above)

Cue(300ms)

Fixation(1700ms)

Stimulus(1000ms)

Fixation(1000ms)

0 300 2000 3000 4000Time (ms)

C Timeline

target positioneye position

target positioneye position

10deg

Fig 1 Saccadic paradigm with idealized eye position traces Saccadic trials lasted 4000ms and began with an instructional cue at thecentre of the screen For half of the participants orange concentric rings were the cue for a prosaccade trial (A) and a blue cross was thecue for an antisaccade trial (B) These cues were reversed for the rest of the participants The cue was flanked horizontally by two smallgreen squares of 02 width that marked the potential locations of stimulus appearance10 left and right of centreThese squares remainedon the screen for the duration of each run (C) At 300ms the instructional cue was replaced by a green fixation ring at the centre of thescreen of 04 diameter and luminance of 20 cdm2 After 1700ms the ring shifted to one of the two target locations right or left withequal probabilityThis was the stimulus to which the participant responded by either making a saccade to it (prosaccade) or to the squareon the opposite side (antisaccade)The green ring remained in the peripheral location for 1000ms and then returned to the centre whereparticipants were also to return their gaze for 1000ms before the start of the next trial Fixation intervals were simply a continuation ofthe fixation display that constituted the final second of the previous saccadic trial

Page 4 of 15 Brain (2008) K NThakkar et al

the appearance of the target (Fischer and Breitmeyer 1987)We classified corrective saccades that followed antisaccade errorsinto short and long latency self-corrections (cf Polli et al 2006)those with latencies (from error initiation) 4130ms suggestingthat the correction was a conscious (aware) response tovisual feedback regarding an error and those with latencies5130ms suggesting that the correction represented a concurrentlyprogrammed correct saccade and that the participant wasunaware of having made an error (Mokler and Fischer 1999McPeek et al 2000) Post-error slowing was calculated for eachparticipant as the difference in latency between correct trialspreceded by an error and correct trials preceded by a correctresponse Error rate and latency of correct trials were analysedusing randomized block ANOVAs with Group (ASD versuscontrols) as the between-subjects factor Task (antisaccade versusprosaccade) as the within-subjects factor and subjects nestedwithin group as the random factor Post-error slowing and self-correction rates for antisaccades were compared by group usingt-tests

Image acquisitionImages were acquired with a 30T Siemens Trio whole body high-speed imaging device equipped for echo planar imaging (SiemensMedical Systems Erlangen Germany) Head stabilization wasachieved with cushioning and all participants wore earplugs(29 dB rating) to attenuate noise Automated shimming proce-dures were performed and scout images were obtained Two high-resolution structural images were acquired in the sagittal plane forslice prescription spatial normalization (spherical and Talairach)and cortical surface reconstruction using a high resolution 3Dmagnetization prepared rapid gradient echo (MPRAGE) sequence(repetition time (TR) 2530ms echo spacing 725ms echo time(TE) 3ms flip angle 7) with an in-plane resolution of 1 and13mm slice thickness T1- and T2-weighted structural imageswith the same slice specifications as the Blood Oxygen LevelDependent (BOLD) scans were obtained to assist in registeringfunctional and structural images

fMRIFunctional images were collected using a gradient echo T2$

weighted sequence (TRTEFlip = 2000 ms30 ms90) Twentycontiguous horizontal slices parallel to the intercommissural plane(voxel size 313 313 5mm) were acquired interleaved Thefunctional sequences included prospective acquisition correctionfor head motion (Thesen et al 2000) Prospective acquisitioncorrection adjusts slice position and orientation in real timeduring data acquisition This reduces motion-induced effects onmagnetization history

DTISingle-shot echo planar imaging DTI was acquired using a twicerefocused spin echo sequence (Reese et al 2003) with the fol-lowing sequence parameters TRTE= 840082ms b= 700 smm2NEX= 1 10 T2 images acquired with b= 0 72 diffusion directions128 128 matrix 2 2mm in-plane resolution 64 axial oblique(AC-PC) slices 2mm (0mm gap) slice thickness scan duration12044 The n= 72 diffusion directions were obtained using theelectrostatic shell algorithm (Jones 2004)

Surface-based analyses for fMRI and DTIFunctional and FA volumes were aligned to the 3D structuralimage for each participant that was created by averaging the twoMPRAGE scans after correcting for motion The averagedMPRAGE scans were used to construct inflated (2D) models ofindividual cortical surfaces using previously described segmenta-tion surface reconstruction and inflation algorithms (Dale et al1999 Fischl et al 1999a) To register data across participantsanatomical functional and DTI scans were spatially normalizedusing a surface-based spherical coordinate system that employs anon-rigid alignment algorithm that explicitly aligns corticalfolding patterns and is relatively robust to inter-individualdifferences in the gyral and sulcal anatomy of cingulate cortex(Dale et al 1999 Fischl et al 1999a b) For both fMRI and DTIthe ACC was localized using an automated surface-basedparcellation system that includes a label for the ACC (Fischlet al 2004) The ACC label was divided into dorsal and rostralsegments by drawing a line perpendicular to the intercommissuralplane at the anterior boundary of the genu of the corpus callosum(Devinsky et al 1995)fMRI and DTI results were displayed on a template brain

consisting of the averaged cortical surface of an independentsample of 40 adults from the Buckner laboratory at WashingtonUniversity To correct for multiple comparisons we ran 10 000Monte Carlo simulations of synthesized white Gaussian noiseusing a P-value of 4005 and the smoothing resampling andaveraging parameters of the functional analyses This determinesthe likelihood that a cluster of a certain size would be found bychance for a given threshold To test our a priori hypothesesconcerning ACC function and structure we restricted thesimulations to our ACC ROIs To determine whether otherregions also showed significant group differences we also ransimulations to correct for multiple comparisons on the entirecortical surface To facilitate comparison with other studiesapproximate Talairach coordinates were derived by mapping thesurface-based coordinates of activation and FA maxima back tothe original structural volume for each participant registering thevolumes to the Montreal Neurological Institute (MNI305) atlas(Collins et al 1994) and averaging the MNI305 coordinates thatcorresponded to the surface maxima across participants Theresulting coordinates were transformed to standard Talairachspace using an algorithm developed by Matthew Brett (httpimagingmrc-cbucamacukimagingMniTalairach)

fMRI data analysesAnalyses were conducted using FreeSurfer (Fischl et al 1999a)and FreeSurfer Functional Analysis Stream (FS-FAST) software(Burock and Dale 2000) In addition to on-line motion correction(prospective acquisition correction) functional scans were cor-rected retrospectively for motion using the AFNI algorithm (Coxand Jesmanowicz 1999) To characterize average motion for eachparticipant total motion in millimetres for all six directions(x y z and three rotational directions) as determined by AFNIwas averaged across the six runs of the task and comparedbetween groups Following motion correction scans were intensitynormalized and smoothed using a 3D 8mm FWHM Gaussiankernel Finite impulse response estimates (Burock and Dale 2000Miezin et al 2000) of the event-related haemodynamic responseswere calculated for each of the four trial types (correctprosaccades error prosaccades correct antisaccades and error

Abnormal response monitoring in autism Brain (2008) Page 5 of 15

novel behaviour of looking in the opposite direction(Hallett 1978) Antisaccades are well-suited for examiningresponse monitoring because they have a relatively well-delineated neuroanatomy and neurophysiology (for reviewsee Munoz and Everling 2004) generate robust electro-phyisological error markers ERN error positivity(Nieuwenhuis et al 2001) and ACC hemodynamicresponses (Polli et al 2005) and involve a high degree ofresponse conflict Moreover individuals with developmen-tal disorders affecting social processing including ASDconsistently show an increased rate of antisaccade errors(ie a failure to suppress the prepotent prosaccade)(Manoach et al 1997 2004 Minshew et al 1999Goldberg et al 2002 Luna et al 2007) which may reflectin part aberrant evaluation of performance and modifica-tion of prepotent stimulus-response mappingsWe examined group differences in ACC activation during

correct and erroneous antisaccade trials using rapidpresentation event-related fMRI Comparisons wererestricted to antisaccade trials since healthy and ASDparticipants produce too few prosaccade errors with thisparadigm (Manoach et al 2004) We examined activity inrostral and dorsal ACC (rACC dACC) separately giventheir differences in cytoarchitecture function and connec-tivity (Devinsky et al 1995 Bush et al 1998 Whalen et al1998) and distinct patterns of haemodynamic activityduring antisaccade error commission (Polli et al 2005)Since response monitoring is the product of coordinatedactivity between the ACC and functionally related regions(Holroyd and Coles 2002) and the ACC shows evidence ofdecreased structural and functional connectivity in autism(Barnea-Goraly et al 2004 Kana et al 2007) we alsoexamined the microstructural integrity of the WM under-lying the ACC using DTI We expected ASD participants toshow (i) an increased antisaccade error rate (ii) increasedACC activity during both correct and erroneous antisaccadetrials and (iii) reduced FA in the WM underlying ACCFinally we hypothesized that (iv) structural and functionalACC abnormalities would be associated with rigidrepetitive behaviour in ASD

MethodsParticipantsTwelve adults with ASD and 14 healthy control (HC) participantswere recruited by poster and website advertisements (see Table 1for demographic data) Participants with ASD were diagnosedwith autism (n= 8) Aspergerrsquos disorder (n= 2) or pervasivedevelopmental disorder not otherwise specified (n= 2) by anexperienced clinician (RMJ) on the basis of current presentationand developmental history as determined by medical recordreview and clinical interview Potential participants meetingDSM-IV criteria for co-morbid psychiatric conditions or substanceabuse were excluded ASD diagnoses were confirmed using theAutism Diagnostic Interview-Revised (ADI-R Rutter et al2003) and the Autism Diagnostic Observation Schedule Module4 (Lord et al 1999) administered by research reliable personnel

Individuals with known autism-related medical conditions (egFragile-X syndrome tuberous sclerosis) were not included Fourof the 12 ASD participants were taking the following medicationsfluoxetine and lithium buproprion and clonazepam citalopramand sertraline and methylphenidateHC participants were screened to exclude a history of autism or

any other neurological or psychiatric condition All participantswere screened to exclude substance abuse or dependence within thepreceding 6 months and any independent condition that mightaffect brain function ASD and control groups were matched forage sex handedness as measured by a laterality score on themodified Edinburgh Handedness Inventory (scores of 100 and+100 denote exclusive use of left or right hands respectively)(Oldfield 1971 White and Ashton 1976) parental socioeconomicstatus on the Hollingshead Index (Hollingshead 1965) years ofeducation and estimated verbal intelligence quotient (IQ) based ona test of single word reading (American National Adult ReadingTest Blair and Spreen 1989) All ASD participants had average orabove estimates of verbal (124 12 range 106ndash141) and non-verbal(120 10 range 100ndash138) IQ as measured by the WechslerAbbreviated Scale of Intelligence (Wechsler 1999) The studywas approved by the Partners Human Research CommitteeAll participants gave written informed consent after the experi-mental procedures had been fully explainedfMRI data were unusable for one ASD participant due to eye

tracking malfunction For a second ASD participant data were notacquired because scanner-safe glasses of the correct prescriptionwere not available fMRI analyses were conducted on theremaining 10 ASD and 14 control participants Saccadic latencydata for one control participant was omitted from analysis due toan acquisition problem DTI data were acquired on all 12 ASDparticipants and 12 of the 14 controls

Saccadic paradigmPrior to scanning the task was explained and participants practicedin a mock scanner until their performance indicated that theyunderstood the directions and were comfortable with the taskParticipants were instructed to respond as quickly and accurately aspossible and told that they would receive a bonus of 5 cents for eachcorrect response in addition to a base rate of pay Each run of the

Table 1 Means standard deviations and groupcomparisons of demographic data

Subjectcharacteristics

HC(n=14)

ASD(n=12)

T P

Age 278 3011 069 050Sex 8M6F 10M2F f=028 022Lateralityscore(Handedness)

75 45 6138 086 040

Parental SESa 131048 117039 z=060 042Years ofeducation

162 163 051 062

Estimatedverbal IQb

114 9 116 8 062 054

aA lower score denotes higher status The Phi value is theresult of a Fisherrsquos exact test The z-value is the result of anon-parametric Mann^Whitney U comparison bBased on theAmerican National Adult ReadingTest

Abnormal response monitoring in autism Brain (2008) Page 3 of 15

task consisted of a pseudorandom sequence of prosaccade andantisaccade trials that were balanced for right and left movementsFigure 1 provides a graphic depiction of the task and a descriptionof task parameters Randomly interleaved with the saccadic trialswere intervals of fixation lasting 2 4 or 6 s The fixation intervalsprovided a baseline and their variable length introduced lsquotemporaljitterrsquo which optimizes the analysis of rapid presentation event-related fMRI designs (Buckner et al 1998 Burock and Dale 2000Miezin et al 2000) The schedule of events was determined using atechnique to optimize the statistical efficiency of event-relateddesigns (Dale 1999) Participants performed six runs of the taskeach lasting 5min 22 s with short rests between runs The totalexperiment lasted about 40min and generated a total of 211prosaccade and 211 antisaccade trials and 80 fixation intervals

Stimulus display and eye trackingDisplays of the eye movement task were generated usingthe Vision Shell programming platform (wwwvisionshellcomWatertown MA USA) and back-projected with a Sharp XG-2000color LCD projector (Osaka Japan) onto a screen at the rear of thebore that was viewed by the participant via a mirror on the headcoil The ISCAN fMRI Remote Eye Tracking Laboratory (ISCANBurlington MA) recorded saccades during scanning This systemused a video camera mounted at the rear of the MRI bore

The camera imaged the eye of the participant via an opticalcombiner a 45 cold transmissive mirror that reflects an infraredimage of the eye with the infrared illumination being provided by anLED mounted on the head coil The system used passive opticalcomponents with no ferrous content within the bore to minimizeartefacts in the MRI images Eye position was sampled at a rate of60Hz Eye images were processed by ISCANrsquos RK-726PCI highresolution PupilCorneal reflection tracker located outside of theshielded MRI room Stimuli presented by Vision Shell were digitallyencoded and relayed to ISCAN as triggers that were inserted into theeye movement recordings

Scoring and analysis of eye movement dataEye movement data were scored in MATLAB (Mathworks NatickMA) using a partially automated programme that determined thedirectional accuracy of each saccade with respect to the requiredresponse and the latency from target onset Saccades wereidentified as horizontal eye movements with velocities exceeding47s The onset of a saccade was defined as the point at which thevelocity of the eye movement first exceeded 31s Only trials withsaccades in the desired direction and latencies over 130ms wereconsidered correct and only correct saccades were included inthe latency analyses The cutoff of 130ms excluded anticipatorysaccades which are executed too quickly to be a valid response to

latency

latency

right

left

A Prosaccade

B Antisaccade

(see above)

Cue(300ms)

Fixation(1700ms)

Stimulus(1000ms)

Fixation(1000ms)

0 300 2000 3000 4000Time (ms)

C Timeline

target positioneye position

target positioneye position

10deg

Fig 1 Saccadic paradigm with idealized eye position traces Saccadic trials lasted 4000ms and began with an instructional cue at thecentre of the screen For half of the participants orange concentric rings were the cue for a prosaccade trial (A) and a blue cross was thecue for an antisaccade trial (B) These cues were reversed for the rest of the participants The cue was flanked horizontally by two smallgreen squares of 02 width that marked the potential locations of stimulus appearance10 left and right of centreThese squares remainedon the screen for the duration of each run (C) At 300ms the instructional cue was replaced by a green fixation ring at the centre of thescreen of 04 diameter and luminance of 20 cdm2 After 1700ms the ring shifted to one of the two target locations right or left withequal probabilityThis was the stimulus to which the participant responded by either making a saccade to it (prosaccade) or to the squareon the opposite side (antisaccade)The green ring remained in the peripheral location for 1000ms and then returned to the centre whereparticipants were also to return their gaze for 1000ms before the start of the next trial Fixation intervals were simply a continuation ofthe fixation display that constituted the final second of the previous saccadic trial

Page 4 of 15 Brain (2008) K NThakkar et al

the appearance of the target (Fischer and Breitmeyer 1987)We classified corrective saccades that followed antisaccade errorsinto short and long latency self-corrections (cf Polli et al 2006)those with latencies (from error initiation) 4130ms suggestingthat the correction was a conscious (aware) response tovisual feedback regarding an error and those with latencies5130ms suggesting that the correction represented a concurrentlyprogrammed correct saccade and that the participant wasunaware of having made an error (Mokler and Fischer 1999McPeek et al 2000) Post-error slowing was calculated for eachparticipant as the difference in latency between correct trialspreceded by an error and correct trials preceded by a correctresponse Error rate and latency of correct trials were analysedusing randomized block ANOVAs with Group (ASD versuscontrols) as the between-subjects factor Task (antisaccade versusprosaccade) as the within-subjects factor and subjects nestedwithin group as the random factor Post-error slowing and self-correction rates for antisaccades were compared by group usingt-tests

Image acquisitionImages were acquired with a 30T Siemens Trio whole body high-speed imaging device equipped for echo planar imaging (SiemensMedical Systems Erlangen Germany) Head stabilization wasachieved with cushioning and all participants wore earplugs(29 dB rating) to attenuate noise Automated shimming proce-dures were performed and scout images were obtained Two high-resolution structural images were acquired in the sagittal plane forslice prescription spatial normalization (spherical and Talairach)and cortical surface reconstruction using a high resolution 3Dmagnetization prepared rapid gradient echo (MPRAGE) sequence(repetition time (TR) 2530ms echo spacing 725ms echo time(TE) 3ms flip angle 7) with an in-plane resolution of 1 and13mm slice thickness T1- and T2-weighted structural imageswith the same slice specifications as the Blood Oxygen LevelDependent (BOLD) scans were obtained to assist in registeringfunctional and structural images

fMRIFunctional images were collected using a gradient echo T2$

weighted sequence (TRTEFlip = 2000 ms30 ms90) Twentycontiguous horizontal slices parallel to the intercommissural plane(voxel size 313 313 5mm) were acquired interleaved Thefunctional sequences included prospective acquisition correctionfor head motion (Thesen et al 2000) Prospective acquisitioncorrection adjusts slice position and orientation in real timeduring data acquisition This reduces motion-induced effects onmagnetization history

DTISingle-shot echo planar imaging DTI was acquired using a twicerefocused spin echo sequence (Reese et al 2003) with the fol-lowing sequence parameters TRTE= 840082ms b= 700 smm2NEX= 1 10 T2 images acquired with b= 0 72 diffusion directions128 128 matrix 2 2mm in-plane resolution 64 axial oblique(AC-PC) slices 2mm (0mm gap) slice thickness scan duration12044 The n= 72 diffusion directions were obtained using theelectrostatic shell algorithm (Jones 2004)

Surface-based analyses for fMRI and DTIFunctional and FA volumes were aligned to the 3D structuralimage for each participant that was created by averaging the twoMPRAGE scans after correcting for motion The averagedMPRAGE scans were used to construct inflated (2D) models ofindividual cortical surfaces using previously described segmenta-tion surface reconstruction and inflation algorithms (Dale et al1999 Fischl et al 1999a) To register data across participantsanatomical functional and DTI scans were spatially normalizedusing a surface-based spherical coordinate system that employs anon-rigid alignment algorithm that explicitly aligns corticalfolding patterns and is relatively robust to inter-individualdifferences in the gyral and sulcal anatomy of cingulate cortex(Dale et al 1999 Fischl et al 1999a b) For both fMRI and DTIthe ACC was localized using an automated surface-basedparcellation system that includes a label for the ACC (Fischlet al 2004) The ACC label was divided into dorsal and rostralsegments by drawing a line perpendicular to the intercommissuralplane at the anterior boundary of the genu of the corpus callosum(Devinsky et al 1995)fMRI and DTI results were displayed on a template brain

consisting of the averaged cortical surface of an independentsample of 40 adults from the Buckner laboratory at WashingtonUniversity To correct for multiple comparisons we ran 10 000Monte Carlo simulations of synthesized white Gaussian noiseusing a P-value of 4005 and the smoothing resampling andaveraging parameters of the functional analyses This determinesthe likelihood that a cluster of a certain size would be found bychance for a given threshold To test our a priori hypothesesconcerning ACC function and structure we restricted thesimulations to our ACC ROIs To determine whether otherregions also showed significant group differences we also ransimulations to correct for multiple comparisons on the entirecortical surface To facilitate comparison with other studiesapproximate Talairach coordinates were derived by mapping thesurface-based coordinates of activation and FA maxima back tothe original structural volume for each participant registering thevolumes to the Montreal Neurological Institute (MNI305) atlas(Collins et al 1994) and averaging the MNI305 coordinates thatcorresponded to the surface maxima across participants Theresulting coordinates were transformed to standard Talairachspace using an algorithm developed by Matthew Brett (httpimagingmrc-cbucamacukimagingMniTalairach)

fMRI data analysesAnalyses were conducted using FreeSurfer (Fischl et al 1999a)and FreeSurfer Functional Analysis Stream (FS-FAST) software(Burock and Dale 2000) In addition to on-line motion correction(prospective acquisition correction) functional scans were cor-rected retrospectively for motion using the AFNI algorithm (Coxand Jesmanowicz 1999) To characterize average motion for eachparticipant total motion in millimetres for all six directions(x y z and three rotational directions) as determined by AFNIwas averaged across the six runs of the task and comparedbetween groups Following motion correction scans were intensitynormalized and smoothed using a 3D 8mm FWHM Gaussiankernel Finite impulse response estimates (Burock and Dale 2000Miezin et al 2000) of the event-related haemodynamic responseswere calculated for each of the four trial types (correctprosaccades error prosaccades correct antisaccades and error

Abnormal response monitoring in autism Brain (2008) Page 5 of 15

task consisted of a pseudorandom sequence of prosaccade andantisaccade trials that were balanced for right and left movementsFigure 1 provides a graphic depiction of the task and a descriptionof task parameters Randomly interleaved with the saccadic trialswere intervals of fixation lasting 2 4 or 6 s The fixation intervalsprovided a baseline and their variable length introduced lsquotemporaljitterrsquo which optimizes the analysis of rapid presentation event-related fMRI designs (Buckner et al 1998 Burock and Dale 2000Miezin et al 2000) The schedule of events was determined using atechnique to optimize the statistical efficiency of event-relateddesigns (Dale 1999) Participants performed six runs of the taskeach lasting 5min 22 s with short rests between runs The totalexperiment lasted about 40min and generated a total of 211prosaccade and 211 antisaccade trials and 80 fixation intervals

Stimulus display and eye trackingDisplays of the eye movement task were generated usingthe Vision Shell programming platform (wwwvisionshellcomWatertown MA USA) and back-projected with a Sharp XG-2000color LCD projector (Osaka Japan) onto a screen at the rear of thebore that was viewed by the participant via a mirror on the headcoil The ISCAN fMRI Remote Eye Tracking Laboratory (ISCANBurlington MA) recorded saccades during scanning This systemused a video camera mounted at the rear of the MRI bore

The camera imaged the eye of the participant via an opticalcombiner a 45 cold transmissive mirror that reflects an infraredimage of the eye with the infrared illumination being provided by anLED mounted on the head coil The system used passive opticalcomponents with no ferrous content within the bore to minimizeartefacts in the MRI images Eye position was sampled at a rate of60Hz Eye images were processed by ISCANrsquos RK-726PCI highresolution PupilCorneal reflection tracker located outside of theshielded MRI room Stimuli presented by Vision Shell were digitallyencoded and relayed to ISCAN as triggers that were inserted into theeye movement recordings

Scoring and analysis of eye movement dataEye movement data were scored in MATLAB (Mathworks NatickMA) using a partially automated programme that determined thedirectional accuracy of each saccade with respect to the requiredresponse and the latency from target onset Saccades wereidentified as horizontal eye movements with velocities exceeding47s The onset of a saccade was defined as the point at which thevelocity of the eye movement first exceeded 31s Only trials withsaccades in the desired direction and latencies over 130ms wereconsidered correct and only correct saccades were included inthe latency analyses The cutoff of 130ms excluded anticipatorysaccades which are executed too quickly to be a valid response to

latency

latency

right

left

A Prosaccade

B Antisaccade

(see above)

Cue(300ms)

Fixation(1700ms)

Stimulus(1000ms)

Fixation(1000ms)

0 300 2000 3000 4000Time (ms)

C Timeline

target positioneye position

target positioneye position

10deg

Fig 1 Saccadic paradigm with idealized eye position traces Saccadic trials lasted 4000ms and began with an instructional cue at thecentre of the screen For half of the participants orange concentric rings were the cue for a prosaccade trial (A) and a blue cross was thecue for an antisaccade trial (B) These cues were reversed for the rest of the participants The cue was flanked horizontally by two smallgreen squares of 02 width that marked the potential locations of stimulus appearance10 left and right of centreThese squares remainedon the screen for the duration of each run (C) At 300ms the instructional cue was replaced by a green fixation ring at the centre of thescreen of 04 diameter and luminance of 20 cdm2 After 1700ms the ring shifted to one of the two target locations right or left withequal probabilityThis was the stimulus to which the participant responded by either making a saccade to it (prosaccade) or to the squareon the opposite side (antisaccade)The green ring remained in the peripheral location for 1000ms and then returned to the centre whereparticipants were also to return their gaze for 1000ms before the start of the next trial Fixation intervals were simply a continuation ofthe fixation display that constituted the final second of the previous saccadic trial

Page 4 of 15 Brain (2008) K NThakkar et al

the appearance of the target (Fischer and Breitmeyer 1987)We classified corrective saccades that followed antisaccade errorsinto short and long latency self-corrections (cf Polli et al 2006)those with latencies (from error initiation) 4130ms suggestingthat the correction was a conscious (aware) response tovisual feedback regarding an error and those with latencies5130ms suggesting that the correction represented a concurrentlyprogrammed correct saccade and that the participant wasunaware of having made an error (Mokler and Fischer 1999McPeek et al 2000) Post-error slowing was calculated for eachparticipant as the difference in latency between correct trialspreceded by an error and correct trials preceded by a correctresponse Error rate and latency of correct trials were analysedusing randomized block ANOVAs with Group (ASD versuscontrols) as the between-subjects factor Task (antisaccade versusprosaccade) as the within-subjects factor and subjects nestedwithin group as the random factor Post-error slowing and self-correction rates for antisaccades were compared by group usingt-tests

Image acquisitionImages were acquired with a 30T Siemens Trio whole body high-speed imaging device equipped for echo planar imaging (SiemensMedical Systems Erlangen Germany) Head stabilization wasachieved with cushioning and all participants wore earplugs(29 dB rating) to attenuate noise Automated shimming proce-dures were performed and scout images were obtained Two high-resolution structural images were acquired in the sagittal plane forslice prescription spatial normalization (spherical and Talairach)and cortical surface reconstruction using a high resolution 3Dmagnetization prepared rapid gradient echo (MPRAGE) sequence(repetition time (TR) 2530ms echo spacing 725ms echo time(TE) 3ms flip angle 7) with an in-plane resolution of 1 and13mm slice thickness T1- and T2-weighted structural imageswith the same slice specifications as the Blood Oxygen LevelDependent (BOLD) scans were obtained to assist in registeringfunctional and structural images

fMRIFunctional images were collected using a gradient echo T2$

weighted sequence (TRTEFlip = 2000 ms30 ms90) Twentycontiguous horizontal slices parallel to the intercommissural plane(voxel size 313 313 5mm) were acquired interleaved Thefunctional sequences included prospective acquisition correctionfor head motion (Thesen et al 2000) Prospective acquisitioncorrection adjusts slice position and orientation in real timeduring data acquisition This reduces motion-induced effects onmagnetization history

DTISingle-shot echo planar imaging DTI was acquired using a twicerefocused spin echo sequence (Reese et al 2003) with the fol-lowing sequence parameters TRTE= 840082ms b= 700 smm2NEX= 1 10 T2 images acquired with b= 0 72 diffusion directions128 128 matrix 2 2mm in-plane resolution 64 axial oblique(AC-PC) slices 2mm (0mm gap) slice thickness scan duration12044 The n= 72 diffusion directions were obtained using theelectrostatic shell algorithm (Jones 2004)

Surface-based analyses for fMRI and DTIFunctional and FA volumes were aligned to the 3D structuralimage for each participant that was created by averaging the twoMPRAGE scans after correcting for motion The averagedMPRAGE scans were used to construct inflated (2D) models ofindividual cortical surfaces using previously described segmenta-tion surface reconstruction and inflation algorithms (Dale et al1999 Fischl et al 1999a) To register data across participantsanatomical functional and DTI scans were spatially normalizedusing a surface-based spherical coordinate system that employs anon-rigid alignment algorithm that explicitly aligns corticalfolding patterns and is relatively robust to inter-individualdifferences in the gyral and sulcal anatomy of cingulate cortex(Dale et al 1999 Fischl et al 1999a b) For both fMRI and DTIthe ACC was localized using an automated surface-basedparcellation system that includes a label for the ACC (Fischlet al 2004) The ACC label was divided into dorsal and rostralsegments by drawing a line perpendicular to the intercommissuralplane at the anterior boundary of the genu of the corpus callosum(Devinsky et al 1995)fMRI and DTI results were displayed on a template brain

consisting of the averaged cortical surface of an independentsample of 40 adults from the Buckner laboratory at WashingtonUniversity To correct for multiple comparisons we ran 10 000Monte Carlo simulations of synthesized white Gaussian noiseusing a P-value of 4005 and the smoothing resampling andaveraging parameters of the functional analyses This determinesthe likelihood that a cluster of a certain size would be found bychance for a given threshold To test our a priori hypothesesconcerning ACC function and structure we restricted thesimulations to our ACC ROIs To determine whether otherregions also showed significant group differences we also ransimulations to correct for multiple comparisons on the entirecortical surface To facilitate comparison with other studiesapproximate Talairach coordinates were derived by mapping thesurface-based coordinates of activation and FA maxima back tothe original structural volume for each participant registering thevolumes to the Montreal Neurological Institute (MNI305) atlas(Collins et al 1994) and averaging the MNI305 coordinates thatcorresponded to the surface maxima across participants Theresulting coordinates were transformed to standard Talairachspace using an algorithm developed by Matthew Brett (httpimagingmrc-cbucamacukimagingMniTalairach)

fMRI data analysesAnalyses were conducted using FreeSurfer (Fischl et al 1999a)and FreeSurfer Functional Analysis Stream (FS-FAST) software(Burock and Dale 2000) In addition to on-line motion correction(prospective acquisition correction) functional scans were cor-rected retrospectively for motion using the AFNI algorithm (Coxand Jesmanowicz 1999) To characterize average motion for eachparticipant total motion in millimetres for all six directions(x y z and three rotational directions) as determined by AFNIwas averaged across the six runs of the task and comparedbetween groups Following motion correction scans were intensitynormalized and smoothed using a 3D 8mm FWHM Gaussiankernel Finite impulse response estimates (Burock and Dale 2000Miezin et al 2000) of the event-related haemodynamic responseswere calculated for each of the four trial types (correctprosaccades error prosaccades correct antisaccades and error

Abnormal response monitoring in autism Brain (2008) Page 5 of 15

the appearance of the target (Fischer and Breitmeyer 1987)We classified corrective saccades that followed antisaccade errorsinto short and long latency self-corrections (cf Polli et al 2006)those with latencies (from error initiation) 4130ms suggestingthat the correction was a conscious (aware) response tovisual feedback regarding an error and those with latencies5130ms suggesting that the correction represented a concurrentlyprogrammed correct saccade and that the participant wasunaware of having made an error (Mokler and Fischer 1999McPeek et al 2000) Post-error slowing was calculated for eachparticipant as the difference in latency between correct trialspreceded by an error and correct trials preceded by a correctresponse Error rate and latency of correct trials were analysedusing randomized block ANOVAs with Group (ASD versuscontrols) as the between-subjects factor Task (antisaccade versusprosaccade) as the within-subjects factor and subjects nestedwithin group as the random factor Post-error slowing and self-correction rates for antisaccades were compared by group usingt-tests

Image acquisitionImages were acquired with a 30T Siemens Trio whole body high-speed imaging device equipped for echo planar imaging (SiemensMedical Systems Erlangen Germany) Head stabilization wasachieved with cushioning and all participants wore earplugs(29 dB rating) to attenuate noise Automated shimming proce-dures were performed and scout images were obtained Two high-resolution structural images were acquired in the sagittal plane forslice prescription spatial normalization (spherical and Talairach)and cortical surface reconstruction using a high resolution 3Dmagnetization prepared rapid gradient echo (MPRAGE) sequence(repetition time (TR) 2530ms echo spacing 725ms echo time(TE) 3ms flip angle 7) with an in-plane resolution of 1 and13mm slice thickness T1- and T2-weighted structural imageswith the same slice specifications as the Blood Oxygen LevelDependent (BOLD) scans were obtained to assist in registeringfunctional and structural images

fMRIFunctional images were collected using a gradient echo T2$

weighted sequence (TRTEFlip = 2000 ms30 ms90) Twentycontiguous horizontal slices parallel to the intercommissural plane(voxel size 313 313 5mm) were acquired interleaved Thefunctional sequences included prospective acquisition correctionfor head motion (Thesen et al 2000) Prospective acquisitioncorrection adjusts slice position and orientation in real timeduring data acquisition This reduces motion-induced effects onmagnetization history

DTISingle-shot echo planar imaging DTI was acquired using a twicerefocused spin echo sequence (Reese et al 2003) with the fol-lowing sequence parameters TRTE= 840082ms b= 700 smm2NEX= 1 10 T2 images acquired with b= 0 72 diffusion directions128 128 matrix 2 2mm in-plane resolution 64 axial oblique(AC-PC) slices 2mm (0mm gap) slice thickness scan duration12044 The n= 72 diffusion directions were obtained using theelectrostatic shell algorithm (Jones 2004)

Surface-based analyses for fMRI and DTIFunctional and FA volumes were aligned to the 3D structuralimage for each participant that was created by averaging the twoMPRAGE scans after correcting for motion The averagedMPRAGE scans were used to construct inflated (2D) models ofindividual cortical surfaces using previously described segmenta-tion surface reconstruction and inflation algorithms (Dale et al1999 Fischl et al 1999a) To register data across participantsanatomical functional and DTI scans were spatially normalizedusing a surface-based spherical coordinate system that employs anon-rigid alignment algorithm that explicitly aligns corticalfolding patterns and is relatively robust to inter-individualdifferences in the gyral and sulcal anatomy of cingulate cortex(Dale et al 1999 Fischl et al 1999a b) For both fMRI and DTIthe ACC was localized using an automated surface-basedparcellation system that includes a label for the ACC (Fischlet al 2004) The ACC label was divided into dorsal and rostralsegments by drawing a line perpendicular to the intercommissuralplane at the anterior boundary of the genu of the corpus callosum(Devinsky et al 1995)fMRI and DTI results were displayed on a template brain

consisting of the averaged cortical surface of an independentsample of 40 adults from the Buckner laboratory at WashingtonUniversity To correct for multiple comparisons we ran 10 000Monte Carlo simulations of synthesized white Gaussian noiseusing a P-value of 4005 and the smoothing resampling andaveraging parameters of the functional analyses This determinesthe likelihood that a cluster of a certain size would be found bychance for a given threshold To test our a priori hypothesesconcerning ACC function and structure we restricted thesimulations to our ACC ROIs To determine whether otherregions also showed significant group differences we also ransimulations to correct for multiple comparisons on the entirecortical surface To facilitate comparison with other studiesapproximate Talairach coordinates were derived by mapping thesurface-based coordinates of activation and FA maxima back tothe original structural volume for each participant registering thevolumes to the Montreal Neurological Institute (MNI305) atlas(Collins et al 1994) and averaging the MNI305 coordinates thatcorresponded to the surface maxima across participants Theresulting coordinates were transformed to standard Talairachspace using an algorithm developed by Matthew Brett (httpimagingmrc-cbucamacukimagingMniTalairach)

fMRI data analysesAnalyses were conducted using FreeSurfer (Fischl et al 1999a)and FreeSurfer Functional Analysis Stream (FS-FAST) software(Burock and Dale 2000) In addition to on-line motion correction(prospective acquisition correction) functional scans were cor-rected retrospectively for motion using the AFNI algorithm (Coxand Jesmanowicz 1999) To characterize average motion for eachparticipant total motion in millimetres for all six directions(x y z and three rotational directions) as determined by AFNIwas averaged across the six runs of the task and comparedbetween groups Following motion correction scans were intensitynormalized and smoothed using a 3D 8mm FWHM Gaussiankernel Finite impulse response estimates (Burock and Dale 2000Miezin et al 2000) of the event-related haemodynamic responseswere calculated for each of the four trial types (correctprosaccades error prosaccades correct antisaccades and error

Abnormal response monitoring in autism Brain (2008) Page 5 of 15

antisaccades) for each participant This involved using a linearmodel to provide unbiased estimates of the average signal intensityat each time point for each trial type without making a prioriassumptions about the shape of the haemodynamic responseHaemodynamic response estimates were computed at 12 timepoints with an interval of 2 s (corresponding to the TR) rangingfrom 4 s prior to the start of a trial to 18 s after the startTemporal correlations in the noise were accounted for by pre-whitening using a global estimate of the residual error auto-correlation function truncated at 30 s (Burock and Dale 2000)Registered group data were smoothed with a 2D 46mm

FWHM Gaussian kernel We compared activation in error versuscorrect antisaccades and in both correct and error antisaccadesversus fixation both within and between groups using a randomeffects model We examined the 6 and 8 s time points that showedpeak error-related activation in previous studies (Polli et al 20052008)

DTI analysisThe objective of the DTI analysis was to measure FA in the WMunderlying specific regions using a vertex-wise analysis of theregistered group data FA was measured in the undeformed WM2mm below the WMgray matter boundary This strategyminimizes partial volume contributions from cortical graymatter and preserves regional specificity (the rationale for thesurface-based approach is elaborated in Manoach et al 2007a)DTI data were analysed using the following multi-step procedure

(1) Motioneddy current distortion correction Raw diffusiondata were corrected for head motion and residual eddycurrent distortion by registering all images to the first T2image which was acquired with b= 0 The registration wasperformed using the FLIRT tool (Jenkinson and Smith 2001)from the FSL software library (httpwwwfmriboxacukfsl)The registration used a global affine (12 df) transformation amutual information cost function and sinc resampling

(2) Tensor reconstruction The diffusion tensor and FA volumeswere reconstructed using the standard least-squares fit to thelog diffusion signal (Basser et al 1994)

(3) Registration of FA and T1 volumes FA volumes wereregistered to the high-resolution structural (T1) volumes foreach participant using the T2 volume as an intermediaryThis avoids the potential bias that might arise from using theFA volume in the registration since FA is the variable ofinterest In addition the graywhite boundary in the T1volume has better correspondence with the T2 volume thanthe FA volume allowing for a more accurate registration TheT2 volume was taken from the b= 0 image in the DTIacquisition and was therefore in register with the FAvolume Each participantrsquos averaged T2 volume was manuallyregistered to the T1 volume with Tkregister2 (httpsurfernmrmghharvardedu) using 9 df including translationrotation and scaling Manual registration was used tomaximize anatomic agreement along the medial wall andbecause echo planar imaging susceptibility distortions in DTIcan confound global registration procedures The resultingspatial transformation was applied to the FA volume thusbringing the FA and T1 volumes into register

(4) Projection of subcortical FA values onto the WMgray matterinterface For each participant a surface representation of theWMgray matter boundary was derived from the T1 volume

using the previously described segmentation surface recon-struction and inflation algorithm (Dale et al 1999 Fischlet al 1999a) FA was sampled in the WM 2mm below theWMgray matter boundary for each vertex on the surface andthen projected onto the WMgray interface FA values weresmoothed using n= 50 iterations of replacement by nearestneighbours This corresponds to smoothing by amp10mmFWHM on the surface

(5) Registration of surface FA maps across participants This wasachieved by applying the surface transformations from thespatial normalization of the T1 maps to the FA maps whichwere in register with T1 maps (step 3)

(6) To test for significant group differences in FA in theregistered group data a t-test was performed in the WMunderlying each vertex

Relations of ACC function and structure torestricted repetitive behaviour in ASDfMRI activation in each contrast at 6 and 8 s and FA values wereregressed onto ADI-R diagnostic algorithm scores of restricted andrepetitive behaviours for every vertex on the cortical surface Inthe regression of FA values on ADI-R scores age was used as acovariate given the documented relation of increasing age withdecreasing FA (Pfefferbaum et al 2000)

ResultsSaccadic performanceASD participants made significantly more errors than HC[F(122) = 782 P= 0008] Although the group by taskinteraction was not significant [F(122) = 099 P= 032)ASD participants had a significantly higher antisaccadeerror rate than controls [Fig 2A t(22) = 268 P= 001 HC655 494 range 143ndash1659 ASD 1241 902range 237ndash2667] but did not differ significantly in theerror rate for prosaccades [t(22) = 162 P= 021 HC204 150 ASD 482 406] (Fig 2A) ASD partici-pants responded more quickly on correct trials[F(121) = 974 P= 0005] and there was a trend to agroup by task interaction [F(121) = 333 P= 008] reflect-ing a greater group latency difference for antisaccade thanprosaccade trials [antisaccade t(21) = 347 P= 0002 HC309 40ms ASD 253 20ms prosaccade t(21) = 256P= 002 HC 254 50ms ASD 212 29ms] (Fig 2B)ASD participants were slightly more likely to correctantisaccade errors [trend t(22) = 177 P= 009 HC85 13 ASD 93 5] (Fig 2C) and did not differfrom controls in the proportion of long versus short self-corrections [t(22) = 150 P= 015 HC 67 29 ASD51 20] Controls did not show significant post-errorslowing for either antisaccades [0 26ms t(12) = 006P= 095] or prosaccades [14 29ms t(12) = 171P= 011] ASD participants showed post-error speeding forantisaccades [11 15ms t(9) = 248 P= 004] and nopost-error change in latency for prosaccades [ASD3 13ms t(9) = 073 P= 048]

Page 6 of 15 Brain (2008) K NThakkar et al

fMRI analyses of response monitoringin the ACCPatients and controls did not differ in mean motion duringfunctional scans [controls 171 081mm patients178 047mm t(22) = 062 P= 083] As group differencesin activation were more pronounced at the 8 s than the 6 stime point we report findings from 8 s onlyControls showed a more differentiated response to error

versus correct trials than ASD participants in the right rACCextending into the dACC (Fig 3A Table 2) The basis of thisgroup difference was that ASD participants showed signifi-cantly greater activation to correct responses in bilateralrACC (Fig 3B Table 2) In this region the group difference inthe response to errors was not significant (Fig 3C) andwithin group analyses revealed that while both groups showeda significant response to errors bilaterally (Control leftmaximum 9 37 12 cluster size = 1204 mm2 cluster-wiseprobability (CWP) = 00001 right maximum 7 26 18 clustersize = 1561 mm2 CWP=00001 ASD left maximum 5 345 cluster size = 1112 mm2 CWP= 00001 right maximum 938 12 cluster size = 850 mm2 CWP= 00001) only the ASDgroup showed a significant response to correct trials (leftmaximum 7 40 1 cluster size = 1107mm2 CWP=00001right maximum 7 39 11 cluster size = 1498mm2CWP=00001) Compared with controls ASD participantsalso showed significantly greater activation for correct trials inanother more anterior perigenual rACC region bilaterally(Fig 3B) and in the left hemisphere this region also showed asignificantly greater response to errors (Fig 3C) Controls didnot activate this perigenual rACC region for either correct orerror responses When multiple comparisons correction wasapplied to the entire cortical surface the finding that the ASDgroup showed an exaggerated response to correct trials inright rACC extending into dACC remained significant(CWP=00001) The only other significant group differencein cortical activation was a less differentiated response forASD participants to error versus correct responses in themedial aspect of the right superior frontal gyrus

Four of our ASD participants were taking medicationsthat could affect ACC function A re-analysis of the datawith only unmedicated ASD participants confirmed themain findings that in the right rACC ASD participantsshowed significantly less discrimination between correctand error trials and greater activation to correct trials(Table 2 Supplementary Figure)

DTI analysesASD participants showed significantly reduced FA in theWM underlying rACC and dACC bilaterally (Fig 4Table 2) The maxima in both hemispheres were in rACCWM near the border with dACC and FA reductionsextended into the corpus callosum These reductions metmultiple comparisons correction for the entire corticalsurface Significantly reduced activation FA in ASD was alsoseen in the WM underlying bilateral frontopolar cortexwhich extended into dorsolateral and ventrolateral pre-frontal cortex in the left hemisphere bilateral intraparietaland parietal transverse sulci and left precentral gyrusThe only region showing significantly greater FA in ASDparticipants versus controls was in the WM underlyingright insula

Relations of ACC function and structure toADI-R scores of restricted and repetitivebehaviour in ASDRegression analyses showed that in ASD higher ADI-Rrestricted and repetitive behaviour scores were associatedwith greater activation during correct trials versus fixationat 8 s in the right rACC (Fig 5A Table 2) This right rACCregion overlapped with the regions showing both signifi-cantly increased activation to correct trials and reduced FARepetitive behaviour scores were not significantly associatedwith ACC activation in other contrasts at either 6 or 8 sThe only other cortical region showing a significant relationat 8 s was the right paracentral gyrus extending into thepostcentral gyrus in the error versus correct contrast

0

10

20

30

40

50

60

70

80

90

100

S

elf-

corr

ecte

d er

rors

p=09

0

2

4

6

8

10

12

14

16

E

rror

p=01

ASD HC ASD HC ASD HC0

50

100

150

200

250

300

350

Late

ncy

(ms)

0

50

100

150

200

250

300

350

p=005A B C

Fig 2 Bar graphs of performance for each group as measured by mean and standard errors of (A) antisaccade error rate (B) latency ofcorrect antisaccades and (C) percentage of antisaccade errors that were self-corrected

Abnormal response monitoring in autism Brain (2008) Page 7 of 15

Lower FA in the WM underlying ACC was significantlyrelated to higher repetitive behaviour scores in leftsubgenual rACC with a maximum vertex in subcallosalgyrus (Fig 5B Table 2) No other cortical regions showedsignificant relations in underlying WMThe literature on repetitive behaviour in autism supports

a division into lsquolow-levelrsquo repetitive motor behaviour andmore complex behaviours that appear to be driven byideation (for review see Turner 1999) To assess therelative importance of lsquosensorimotorrsquo versus lsquoideationalrsquoforms of repetition in accounting for the observed relationswith ACC activation and structure we first divided eachparticipantsrsquo ADI-R repetitive behaviour score into lsquosenso-rimotorrsquo (ie D3D4 sterotypies mannerisms repetitiveobject use unusual sensory interests) and lsquoideationalrsquo (ieD1D2 narrow restricted interests and rituals) subscores(Hollander et al 2003) On average the ideational subscoreaccounted for 65 14 of the total repetitive behaviourscore We then performed multiple regression analysiswith BOLD or FA value at the ACC maxima of theregression with total score as the dependent variable andthe two subscores as covariates Sensorimotor subscoreswere more strongly related to activation in the ACC(P= 004) than ideational subscores (P= 017) The oppositewas true for FA which showed a stronger correlationswith ideational subscores (P= 005) than sensorimotor

subscores (P= 018) The subscores did not differ signifi-cantly in their contribution to activation or FA and neitherimproved on the relations seen with the total repetitivebehaviour score (activation P= 00009 FA P= 004)suggesting that both subscores contributed to the relationsobserved

DiscussionThe present study demonstrates functional and structuralabnormalities of the ACC in ASD that may contribute torestricted repetitive and stereotyped patterns of behavioura defining feature of ASD Compared with controls ASDparticipants showed increased rACC activation to bothcorrect and error responses and reduced microstructuralintegrity of the WM underlying ACC including in regionsshowing abnormally increased response monitoring activa-tion Moreover in rACC increased activation to correcttrials and reduced FA were related to ratings of rigidrepetitive behaviour These findings support the hypothesisof hyperactive response monitoring in ASD and link it torepetitive behaviour

Relative to controls ASD participants showed reduceddiscrimination between error and correct trials in a rACCregion that responds to errors (Polli et al 2005) This wasprimarily due to an exaggerated response to correct trials

Fig 3 Statistical maps of group differences in fMRI activation at 8 s for contrasts of (A) error versus correct antisaccades (B) correctantisaccades versus fixation (C) error antisaccades versus fixation Statistical maps are displayed on the inflated medial cortical surfacesof the template brain at P4 005 Regions of greater activation in controls are depicted in warm colours greater activation in ASD patientsis depicted in blue The rACC is outlined in red and the dACC is outlined in blue The gray masks cover non-surface regions in whichactivity is displaced Haemodynamic response time course graphs with standard error bars are displayed for the indicated vertices withpeak activation in the group comparisons

Page 8 of 15 Brain (2008) K NThakkar et al

Table 2 Maxima and locations of clusters showing significant group differences and significant relations to ADI-R repetitionscores in ASD

ApproximateTalairach coordinates

Cluster size (mm) x y z t-value (max) CWP

fMRI group comparisonsError versus correctHC4ASDRight rostral anterior cingulate gyrus 488 7 38 12 331 00001

Unmedicated ASD participants 424 7 38 12 218 0008Right medial superior frontal gyrusa 2071 9 50 38 353 0001Correct versus fixationASD4HCLeft pericollosal sulcus (rACC) 413 6 30 6 222 00001

Unmedicated ASD participants 102 6 31 6 159 041Right rostral anterior cingulate gyrusa 1746 7 39 11 347 00001

Unmedicated ASD participants 912 9 39 10 361 00002Pericollosal anterior cingulate gyrus 6 36 6

Error versus fixationASD4HCLeft rostral anterior cingulate gyrus 255 2 26 4 364 002

Unmedicated ASD participants 80 1 26 3 238 078DTI-FA group comparisonbHC4ASDLeft rostral anterior cingulate gyrusa 1598 7 39 6 605 00001Right rostral anterior cingulate gyrusa 6620 7 38 3 564 00001Frontopolar transverse sulcus 20 61 14

Left transverse gyrus frontopolara 6152 20 59 6 72 00001Left precentral gyrusa 1022 47 8 41 511 0009Left intraparietal and parietal transverse sa 4878 40 50 42 505 00001Right intraparietal and parietal transverse sa 1465 21 56 52 406 00008ASD4HCRight anterior circular sulcus insula$ 838 35 31 2 291 005fMRI regressions with ADI-RCorrect versus fixationRight rostral anterior cingulate gyrus 684 6 37 7 323 00001Error versus correctRight paracentral gryrusa 1642 6 25 51 259 0005DTI-FA regressions with ADI-RLeft subcollosal gyrus (rACC) 148 5 18 8 204 005

aMeets correction for entire cortical surface bCWP levels are based on correction for entire cortical surface Important local maximaare indented Results for fMRI group comparisons restricted to unmedicated ASD participants (n=6) are indented and italicized

Fig 4 Statistical map of group differences in FA displayed on the inflated medial cortical surfaces of the template brain at athreshold of P4 005 Regions of greater FA in controls are depicted in warm colours The rACC is outlined in red and the dACC isoutlined in blue

Abnormal response monitoring in autism Brain (2008) Page 9 of 15

rather than a blunted response to errors Reduceddiscrimination between response outcomes may interferewith adjusting responses to obtain the most favourableoutcome both with regard to subsequent trials and longerterm modification of the strength of stimulus-responsemappings This could contribute to behaviour that isstimulus-bound and repetitive rather than flexible andresponsive to contingencies This provides a plausibleexplanation for the relation between increased rACCactivity to correct trials and repetitive behaviourA complementary explanation has been offered for the

relation of increased error signals to the severity ofobsessions and compulsions in OCD (Gehring et al2000 Ursu et al 2003) A longstanding theory of OCDposits that inappropriate and exaggerated error signals leadto a pervasive sense of incompleteness and self-doubt andbehavioural repetition represents a vain attempt to reducethese error signals (Pitman 1987) According to this theoryexaggerated ACC activation to correct trials reflectsinappropriate error signalling and triggers a compulsionto repeat behaviours that were already successfully com-pleted (Maltby et al 2005) Similar to individuals withOCD ASD participants showed an exaggerated ACCresponse to correct trials (Ursu et al 2003 Maltby et al2005) While in OCD fMRI activation to correct trials wasmaximal in dACC (Maltby et al 2005) in ASD it was

maximal in rACC and was related to repetitive behaviourWhile dACC activity following errors is thought to reflecteither conflict or error detection (for review see Ullspergerand von Cramon 2004) rACC activity following errors hasbeen interpreted as appraisal of the affective or motivationalsalience of errors (van Veen and Carter 2002 Luu et al2003 Taylor et al 2006) Thus one interpretation of thisrelation is that it reflects an inappropriate affective responseto correct performance that triggers behavioural repetitionThis exaggerated affective response may occur even withawareness that one has performed correctly Regardless of itsbasis the relation between increased rACC responses tocorrect trials and repetitive behaviour suggests that aberrantresponse monitoring contributes to a defining feature of ASD

Both ASD and OCD are characterized by behaviour thatis rigid and not optimally responsive to contingency andthere is evidence of both phenomenological overlap anddistinctions between the obsessions and compulsions ofOCD and the restricted and repetitive behaviours of ASDboth in terms of subjective experience and type ofbehaviour (Baron-Cohen 1989 McDougle et al 1995Russell et al 2005 Zandt et al 2007) Interestingly a studyof multiplex families of children with autism found thatparents were significantly more likely to have obsessivendashcompulsive traits or disorder if the children had high ADI-Rrepetitive behaviour scores and this was due to loading

Fig 5 Relations of ACC function and structure to restricted repetitive behaviour in ASD (A) Statistical map of regression of rawBOLD signal on ADI-R repetitive behaviour score BOLD signal is from the correct antisaccade versus fixation contrast at 8 s Results aredisplayed on the inflated right medial cortical surface Scatter plot shows BOLD signal from the maximum vertex in right rACC foreach ASD participant on the y-axis and ADI-R score on the x-axis (B) Statistical map of regression of FA on ADI-R repetitive behaviourscore Results are displayed on the inflated left medial surface Scatter plot shows FA from the maximum vertex in left rACC for eachASD participant on the y-axis and ADI-R score on the x-axis The rACC is outlined in red and the dACC is outlined in blue

Page 10 of 15 Brain (2008) K NThakkar et al

on ideational but not sensorimotor items (Hollander et al2003) This finding supports the distinction betweenrepetition of movements and repetition of more complexideational behaviours (for review see Turner 1999) andsuggests that the latter may share underlying mechanisms inOCD and ASD In the present study both sensorimotorand ideational subscores contributed to the relations seenwith response monitoring activation and FA At presentrepetitive behaviours in autism are incompletely under-stood and neurobiologically valid dimensions are yet to bedelineated The present findings suggest that ACC abnorm-alities may contribute to repetitive behaviour in ASDpossibly on the basis of their contribution to aberrantresponse monitoring but linking ACC abnormalities tospecific dimensions of repetitive behaviour and identifyingthe underlying mechanisms requires further study withlarge samplesAbnormal response monitoring may also contribute to

deficient cognitive control of behaviour in ASD which inthe present study was reflected in an increased antisaccadeerror rate consistent with prior work (Manoach et al1997 2004 Minshew et al 1999 Goldberg et al 2002Luna et al 2007) and possibly in reduced dynamicperformance adjustment In spite of making more errorsASD participants responded more quickly than controls oncorrect trials suggesting a disruption in the speed-accuracytrade-off This apparent difficulty in modulating responsespeed to improve accuracy is unlikely to reflect reducederror awareness since ASD participants were as or morelikely than controls to self-correct errors Nor is it likely toreflect a blunted neural response to errors given both theincreased response to errors in left perigenual rACC andprevious work showing normal or increased amplitude ofthe ERN in ASD (Henderson et al 2006) Deficient use ofoutcomes to adjust performance in a continuous dynamicfashion may stem from response monitoring dysfunction inthe ACC andor from dysconnection of the ACC fromocular motor regions involved in saccade generationAberrant connectivity of ACC regions subserving responsemonitoring in ASD is suggested by our finding of reducedFA in the underlying WM This finding is consistent with aprevious report of reduced FA in ACC white matter inautism (Barnea-Goraly et al 2004) and with evidence ofdecreased functional connectivity of the ACC to otherregions during a task involving response inhibition (Kanaet al 2007)How might information concerning the outcome of the

prior response be communicated to ocular motor regionsresponsible for adjusting behaviour in the subsequent trialMonkey electrophysiology studies suggest a basis for speed-accuracy trade-offs in the ocular motor system Comparedwith prosaccades antisaccades are associated with reducedpre-target activity of saccade-related neurons in the frontaleye field (Everling et al 1999 Everling and Munoz 2000)and superior colliculus (Everling et al 1999) key regions insaccade generation Moreover lower pre-target frontal eye

field activity is correlated with increased saccadic latencyand fewer antisaccade errors Reduced preparatory neuralactivity in response to the instruction to perform anantisaccade presumably makes it more difficult for activityrelated to the incorrect but prepotent prosaccade to reachthe threshold for triggering an eye movement However theprice for this improved accuracy is that the activitygenerating the correct antisaccade also takes longer toreach threshold resulting in prolonged saccadic latenciesThus we hypothesize that the slowing of responses thatfollow an antisaccade error (ie post-error slowingNieuwenhuis et al 2001) stems from an ACC errorresponse that signals the frontal eye field to reduce itspreparatory activity during the subsequent trial Thishypothesis is consistent with previous work demonstratingthat other aspects of trial history affect preparatory neuralactivity in the frontal eye field during antisaccades(Manoach et al 2007b) How might error signals beconveyed to frontal eye field The use of outcomes todynamically adjust responses has been attributed to ACCacting with lateral prefrontal cortex (Gehring and Knight2000 Paus 2001 Garavan et al 2002) and lateralprefrontal cortex has been proposed to provide inhibitoryinput to frontal eye field during antisaccades (DeSouzaet al 2003) Thus in ASD aberrant communicationbetween response evaluation in the ACC and regionsinitiating compensatory adjustments possibly lateral pre-frontal cortex could result in a failure to optimally reducepreparatory firing rates in frontal eye field in response toa prior error This reduced dynamic performance adjust-ment could lead to faster responding and more errorsWe are presently testing this hypothesis using an anti-saccade paradigm that gives rise to post-error slowing(Rabbitt 1966) and will allow a more direct examinationof compensatory behaviour and its supporting neuralcircuitry Unfortunately although post-error slowing isseen when antisaccades are the only task (Polli et al 2006)the intermixed antisaccade and prosaccade paradigmemployed here gives rise to other inter-trial factors thatalso affect latency (eg task-switching and prior antisac-cades Cherkasova et al 2002 Fecteau et al 2004 Bartonet al 2006 Manoach et al 2007b) and may obscure theeffects of a prior error There is evidence of reduced post-error slowing in adults with high functioning autismhowever from a previous study that employed a memorysearch task (Bogte et al 2007) Generalizing beyond theocular motor system failures of dynamic performanceadjustment due to ACC dysfunction and dysconnectivitymay contribute to the rigid and perseverative behavioursthat characterize ASD

Repetitive behaviour scores in ASD were also associatedwith reduced FA in the WM underlying left subgenualrACC Subgenual rACC is part of a network that is thoughtto play an important role in generating autonomicresponses to motivationally salient stimuli evaluatingtheir significance in predicting reward and in using this

Abnormal response monitoring in autism Brain (2008) Page 11 of 15

information to guide behaviour (Damasio et al 1990Drevets 2000) An important consideration in interpretingthis relation of FA to repetition is that subgenual rACCregion was not functionally or structurally abnormal in thisASD sample This suggests that even normal variation in FAof the subgenual rACC may contribute to a tendency torepeat unreinforced stimulus-response mappings Moregenerally the relations we observed between behaviouralrepetition and ACC function and structure may not bespecific to ASD Accumulating evidence suggests that thethree symptom clusters that define autismmdashsocial impair-ment communication deficits and restricted repetitivebehaviourmdasharise from distinct genetic and cognitivemechanisms and may represent continuous traits in thepopulation (for reviews see Happe et al 2006 London2007) Previous studies have reported relations betweenindices of response monitoring and repetition in OCD(Gehring et al 2000 Ursu et al 2003) and in non-clinically ascertained samples of children and adults withobsessivendashcompulsive characteristics (Hajcak and Simons2002 Santesso et al 2006)Our findings also raise the question of whether the

response monitoring abnormalities that we observed arespecific to ASD Similar abnormalities have been reportedin OCD (Maltby et al 2005) and attention deficithyperactivity disorder (van Meel et al 2007) Whilemethodological differences preclude a direct comparisonof these findings we previously studied patients withschizophrenia using the identical paradigm (Polli et al2008) Schizophrenia is also a neurodevelopmental disorderthat is characterized by rigid and perseverative behaviourresponse monitoring abnormalities (Carter et al 2001)reduced FA in ACC WM (Manoach et al 2007a) and aconsistently elevated antisaccade error rate (Radant et al2007 Manoach et al 2002) Although both schizophreniaand ASD samples made more antisaccade errors thancontrols ASD participants performed correct trials fasterand schizophrenia patients performed more slowly Whileboth groups showed reduced discrimination between errorand correct trials in the ACC this occurred for differentreasons In ASD this reflected a hyperactive response tocorrect trials while in schizophrenia it reflected a bluntedACC response to errors consistent with numerous previousreports (Laurens et al 2003) These differences argue fordistinct behavioural and neural signatures of responsemonitoring deficits in schizophrenia and ASD Thesefindings along with those of recent neuroimaging studiessuggesting genetic mediation of response monitoring(Fallgatter et al 2004 Klein et al 2007 Kramer et al2007) support the candidacy of neuroimaging-basedindices of response monitoring as neurocognitive endophe-notypes Whether response monitoring abnormalities canbe distinguished in ASD OCD and other neuropsychiatricdisorders whether they are related to diagnosis or specifictraits and whether they are also present in unaffectedfamily members remains to be seen

Several additional limitations to and alternative inter-pretations of our findings merit consideration First mightthe hyperactive rACC response to correct antisaccades inASD reflect a less efficient response to conflict ratherthan response monitoring The ACC is thought to play arole in detecting conflict and engaging cognitive controlto resolve it (for review see Carter and van Veen 2007)Antisaccades involve a high degree of conflict as one mustchoose between two competing responses and inhibit theprepotent one Errors give rise to another type of conflictwhen the actual response is compared with the intendedone Thus both correct and erroneous antisaccadesengender conflict While it is not possible to entirely ruleout the conflict explanation several observations argueagainst it If this rACC region were detecting conflict itshould respond to both correct and error trials Insteadcontrols showed a significant response only to errors Inaddition the timing of the haemodynamic response tocorrect antisaccades in ASD which started amp4 s and peakedat amp8 s (Fig 3) argues for a role in response monitoringrather than successful conflict resolution In this paradigmdifferential fMRI activation for correct antisaccade versusprosaccade trials including in dACC presumably reflectingconflict detection response inhibition and saccade genera-tion peaks at amp4 s and is mostly gone by 8 s (Manoachet al 2007b see also Polli et al 2005 for a discussionof ACC regions involved in saccade generation versusresponse monitoring) Finally it is dorsal rather than rACCactivation that is associated with engaging cognitive controlon correct high-conflict and error trials (Carter and vanVeen 2007)

An important limitation is that the small sample size ofthe present study may have compromised our power todetect additional group differences and relations of interestparticularly in regions outside our a priori ACC regions ofinterest for which more stringent multiple comparisonscorrections were applied In spite of the relatively smallsample size our a priori hypotheses were confirmedIn addition four of our ASD participants were takingmedications that could affect ACC function A re-analysisof the fMRI data with only unmedicated ASD participantshowever supported our main findings of significantly lessdiscrimination between response outcomes in the rACCand greater activation to correct trials It is also important tonote that our sample was limited to high functioning adultswith ASD so it is not clear that our findings would generalizeto lower functioning or younger samples Although increasedantisaccade error rates are seen in autism as early as ages 8ndash12(Luna et al 2007) we limited our study to adults sincesaccadic inhibition may not fully develop until late adoles-cence (Klein and Foerster 2001) and larger samples would benecessary to discriminate between the effects of diagnosis andthose due to normal development Another limitationconcerns our interpretation of FA as reflecting connectivityFA reflects the degree of directional coherence of waterdiffusion in tissue and is therefore an indirect measure of

Page 12 of 15 Brain (2008) K NThakkar et al

WMmicrostructure While FA correlates with axon myelinat-ion (Harsan et al 2006) not all of its biophysicaldeterminants are fully understood (Beaulieu 2002) Thus itis not possible to attribute FA differences or the relations weobserved to a particular WM propertyIn summary the present study reports hyperactive

response monitoring in the rACC in ASD that is relatedto restricted repetitive behaviour and is accompanied byreduced FA in the underlying white matter This suggests astructural correlate to abnormal function that may com-promise connectivity in the circuitry subserving responsemonitoring These findings suggest that structural and fun-ctional abnormalities of the ACC compromise responsemonitoring and contribute to behavioural repetition inASD Impairments in evaluating and learning from errorsmay lead to behaviour that is rigid and repetitive ratherthan optimally guided by outcomes and may compromiseperformance across a wide range of tasks Illuminating theneural basis and clinical significance of response monitoringdeficits in ASD represents a fruitful avenue for furtherresearch

Supplementary materialSupplementary material is available at Brain online

AcknowledgementsThe authors wish to thank Mark Vangel for statisticalconsultation Robert Levy Kim Ono Christopher Shermanand Jeremy Young for help with manuscript preparationand Matt Cain and Jay Edelman for technical assistanceSome of these data were presented at the annual meetingof the Society for Neuroscience in Washington DC inNovember 2005 This study was supported by NationalInstitute for Mental Health [R01 MH67720 (DSM)MH72120 (FEP)] Mental Illness Neuroscience Discovery(MIND) Institute (DOE DE-FG02-99ER62764) TheNational Center for Research Resources (P41RR14075)Funding to pay the Open Access publication charges forthis article was provided by the National Institute forMental Health NRSA MH72120

ReferencesAshwin C Baron-Cohen S Wheelwright S OrsquoRiordan M Bullmore ETDifferential activation of the amygdala and the lsquosocial brainrsquo duringfearful face-processing in Asperger syndrome Neuropsychologia 200745 2ndash14

Barnea-Goraly N Kwon H Menon V Eliez S Lotspeich L Reiss ALWhite matter structure in autism preliminary evidence from diffusiontensor imaging Biol Psychiatry 2004 55 323ndash6

Baron-Cohen S Do autistic children have obsessions and compulsionsBr J Clin Psychol 1989 28 (Pt 3) 193ndash200

Barton JJ Greenzang C Hefter R Edelman J Manoach DS Switchingplasticity and prediction in a saccadic task-switch paradigm Exp BrainRes 2006 168 76ndash87

Basser PJ Mattiello J LeBihan D MR diffusion tensor spectroscopy andimaging Biophys J 1994 66 259ndash67

Beaulieu C The basis of anisotropic water diffusion in the nervous system- a technical review NMR Biomed 2002 15 435ndash55

Blair JR Spreen O Predicting premorbid IQ a revision of the NationalAdult Reading Test Clin Neuropsychol 1989 3 129ndash136

Bogte H Flamma B van der Meere J van Engeland H Post-erroradaptation in adults with high functioning autism Neuropsychologia2007 45 1707ndash14

Breiter HC Rauch SL Kwong KK Baker JR Weisskoff RM Kennedy DNet al Functional magnetic resonance imaging of symptom provocationin obsessive-compulsive disorder Arch Gen Psychiatry 1996 53595ndash606

Buckner RL Goodman J Burock M Rotte M Koutstaal W Schacter Det al Functional-anatomic correlates of object priming in humansrevealed by rapid presentation event-related fMRI Neuron 1998 20285ndash96

Burock MA Dale AM Estimation and detection of event-related fMRIsignals with temporally correlated noise a statistically efficient andunbiased approach Hum Brain Mapp 2000 11 249ndash60

Bush G Whalen PJ Rosen BR Jenike MA McInerney SC Rauch SL Thecounting stroop an interference task specialized for functionalneuroimaging- -validation study with functional MRI Hum BrainMapp 1998 6 270ndash82

Carter CS MacDonald AW 3rd Ross LL Stenger VA Anterior cingulatecortex activity and impaired self-monitoring of performance in patientswith schizophrenia an event-related fMRI study Am J Psychiatry 2001158 1423ndash8

Carter CS van Veen V Anterior cingulate cortex and conflict detection anupdate of theory and data Cogn Affect Behav Neurosci 2007 7 367ndash79

Cherkasova MV Manoach DS Intriligator JM Barton JJS Antisaccadesand task-switching interactions in controlled processing Exp Brain Res2002 144 528ndash37

Collins DL Neelin P Peters TM Evans AC Automatic 3D intersubjectregistration of MR volumetric data in standardized Talairach spaceJ Comput Assist Tomogr 1994 18 192ndash205

Courchesne E Pierce K Why the frontal cortex in autism might be talkingonly to itself local over-connectivity but long-distance disconnectionCurr Opin Neurobiol 2005 15 225ndash30

Cox RW Jesmanowicz A Real-time 3D image registration for functionalMRI Magn Reson Med 1999 42 1014ndash8

Dale AM Optimal experimental design for event-related fMRI Hum BrainMapp 1999 8 109ndash40

Dale AM Fischl B Sereno MI Cortical surface-based analysisI Segmentation and surface reconstruction Neuroimage 1999 9 179ndash94

Damasio AR Tranel D Damasio H Individuals with sociopathic behaviorcaused by frontal damage fail to respond autonomically to social stimuliBehav Brain Res 1990 41 81ndash94

Dehaene S Posner MI Tucker DM Localization of a neural system forerror detection and compensation Psychol Sci 1994 5 303ndash5

DeSouza JF Menon RS Everling S Preparatory set associated with pro-saccades and anti-saccades in humans investigated with event-relatedFMRI J Neurophysiol 2003 89 1016ndash23

Devinsky O Morrell MJ Vogt BA Contributions of anterior cingulatecortex to behaviour Brain 1995 118 (Pt 1) 279ndash306

Dichter GS Belger A Social stimuli interfere with cognitive control inautism Neuroimage 2007 35 1219ndash30

Dougherty DD Baer L Cosgrove GR Cassem EH Price BHNierenberg AA et al Prospective long-term follow-up of 44 patientswho received cingulotomy for treatment-refractory obsessive-compulsivedisorder Am J Psychiatry 2002 159 269ndash75

Drevets WC Neuroimaging studies of mood disorders Biol Psychiatry2000 48 813ndash29

Everling S Dorris MC Klein RM Munoz DP Role of primate superiorcolliculus in preparation and execution of anti-saccades and pro-saccades J Neurosci 1999 19 2740ndash54

Everling S Munoz DP Neuronal correlates for preparatory set associatedwith pro-saccades and anti-saccades in the primate frontal eye fieldJ Neurosci 2000 20 387ndash400

Abnormal response monitoring in autism Brain (2008) Page 13 of 15

Falkenstein M Hohnsbein J Hoormann J Event-related correlates of errorsin reaction tasks In Karmos G Molnar M Csepe I Desmedt JE editorsPerspectives of event-related potentials research AmsterdamThe Netherlands Elsevier Science 1995 p 287ndash96

Fallgatter AJ Herrmann MJ Roemmler J Ehlis AC Wagener AHeidrich A et al Allelic variation of serotonin transporter functionmodulates the brain electrical response for error processingNeuropsychopharmacology 2004 29 1506ndash11

Fecteau JH Au C Armstrong IT Munoz DP Sensory biases producealternation advantage found in sequential saccadic eye movement tasksExp Brain Res 2004 159 84ndash91

Fischer B Breitmeyer B Mechanisms of visual attention revealed bysaccadic eye movements Neuropsychologia 1987 25 73ndash83

Fischl B Sereno MI Dale AM Cortical surface-based analysis II Inflationflattening and a surface-based coordinate system Neuroimage 1999a 9195ndash207

Fischl B Sereno MI Tootell RBH Dale AM High-resolution intersubjectaveraging and a coordinate system for the cortical surface Hum BrainMapp 1999b 8 272ndash84

Fischl B van der Kouwe A Destrieux C Halgren E Segonne F Salat DHet al Automatically parcellating the human cerebral cortex CerebCortex 2004 14 11ndash22

Fitzgerald KD Welsh RC Gehring WJ Abelson JL Himle JA Liberzon Iet al Error-related hyperactivity of the anterior cingulate cortex inobsessive-compulsive disorder Biol Psychiatry 2005 57 287ndash94

Garavan H Ross TJ Murphy K Roche RA Stein EA Dissociable executivefunctions in the dynamic control of behavior inhibition errordetection and correction NeuroImage 2002 17 1820ndash9

Gehring WJ Goss B Coles MG Meyer DE Donchin E A neuralsystem for error detection and compensation Psychol Sci 1993 4385ndash90

Gehring WJ Himle J Nisenson LG Action-monitoring dysfunction inobsessive-compulsive disorder Psychol Sci 2000 11 1ndash6

Gehring WJ Knight RT Prefrontal-cingulate interactions in actionmonitoring Nat Neurosci 2000 3 516ndash20

Goldberg MC Lasker AG Zee DS Garth E Tien A Landa RJ Deficits inthe initiation of eye movements in the absence of a visual target inadolescents with high functioning autism Neuropsychologia 2002 402039ndash49

Gomot M Bernard FA Davis MH Belmonte MK Ashwin CBullmore ET et al Change detection in children with autism anauditory event-related fMRI study Neuroimage 2006 29 475ndash84

Hajcak G Simons RF Error-related brain activity in obsessive-compulsiveundergraduates Psychiatry Res 2002 110 63ndash72

Hall GB Szechtman H Nahmias C Enhanced salience and emotionrecognition in autism a PET study Am J Psychiatry 2003 160 1439ndash41

Hallett PE Primary and secondary saccades to goals defined byinstructions Vision Res 1978 18 1279ndash96

Happe F Ronald A Plomin R Time to give up on a single explanation forautism Nat Neurosci 2006 9 1218ndash20

Harsan LA Poulet P Guignard B Steibel J Parizel N de Sousa PL et alBrain dysmyelination and recovery assessment by noninvasive in vivodiffusion tensor magnetic resonance imaging J Neurosci Res 2006 83392ndash402

Haznedar MM Buchsbaum MS Metzger M Solimando A Spiegel-Cohen J Hollander E Anterior cingulate gyrus volume and glucosemetabolism in autistic disorder Am J Psychiatry 1997 154 1047ndash50

Haznedar MM Buchsbaum MS Wei TC Hof PR Cartwright CBienstock CA et al Limbic circuitry in patients with autism spectrumdisorders studied with positron emission tomography and magneticresonance imaging Am J Psychiatry 2000 157 1994ndash2001

Henderson H Schwartz C Mundy P Burnette C Sutton S Zahka N et alResponse monitoring the error-related negativity and differences insocial behavior in autism Brain Cogn 2006 61 96ndash109

Hill EL Executive dysfunction in autism Trends Cogn Sci 2004 8 26ndash32

Hollander E King A Delaney K Smith CJ Silverman JM Obsessive-compulsive behaviors in parents of multiplex autism families PsychiatryRes 2003 117 11ndash6

Hollingshead AB Two factor index of social position New Haven CTYale University Press 1965

Holroyd CB Coles MG The neural basis of human error processingreinforcement learning dopamine and the error-related negativityPsychol Rev 2002 109 679ndash709

Jenkinson M Smith S A global optimisation method for robust affineregistration of brain images Med Image Anal 2001 5 143ndash56

Johannes S Wieringa BM Nager W Rada D Dengler R Emrich HMet al Discrepant target detection and action monitoring in obsessive-compulsive disorder Psychiatry Res 2001 108 101ndash10

Jones DK The effect of gradient sampling schemes on measures derivedfrom diffusion tensor MRI A Monte Carlo study Magn Reson Med2004 51 807ndash15

Kana RK Keller TA Minshew NJ Just MA Inhibitory control in high-functioning autism decreased activation and underconnectivity ininhibition networks Biol Psychiatry 2007 62 198ndash206

Kennedy DP Redcay E Courchesne E Failing to deactivate restingfunctional abnormalities in autism Proc Natl Acad Sci USA 2006 1038275ndash80

Klein C Foerster F Development of prosaccade and antisaccade taskperformance in participants aged 6 to 26 years Psychophysiology 200138 179ndash89

Klein TA Neumann J Reuter M Hennig J von Cramon DYUllsperger M Genetically determined differences in learning fromerrors Science 2007 318 1642ndash5

Kramer UM Cunillera T Camara E Marco-Pallares J Cucurell DNager W et al The impact of catechol-O-methyltransferase anddopamine D4 receptor genotypes on neurophysiological markers ofperformance monitoring J Neurosci 2007 27 14190ndash8

Laurens KR Ngan ET Bates AT Kiehl KA Liddle PF Rostral anteriorcingulate cortex dysfunction during error processing in schizophreniaBrain 2003 126 610ndash22

Levitt JG OrsquoNeill J Blanton RE Smalley S Fadale D McCracken JT et alProton magnetic resonance spectroscopic imaging of the brain inchildhood autism Biol Psychiatry 2003 54 1355ndash66

Leyfer OT Folstein SE Bacalman S Davis NO Dinh E Morgan J et alComorbid psychiatric disorders in children with autism interview devel-opment and rates of disorders J Autism Dev Disord 2006 36 849ndash61

London E The role of the neurobiologist in redefining the diagnosis ofautism Brain Pathol 2007 17 408ndash11

Lopez BR Lincoln AJ Ozonoff S Lai Z Examining the relationshipbetween executive functions and restricted repetitive symptoms ofAutistic Disorder J Autism Dev Disord 2005 35 445ndash60

Lord C Rutter M DiLavore PC Risi S Autism diagnostic observationschedule - WPS (ADOS-WPS) Los Angeles CA Western PsychologicalServices 1999

Luna B Doll SK Hegedus SJ Minshew NJ Sweeney JA Maturation ofexecutive function in autism Biol Psychiatry 2007 61 474ndash81

Luu P Tucker DM Derryberry D Reed M Poulsen C Electrophysiolog-ical responses to errors and feedback in the process of action regulationPsychol Sci 2003 14 47ndash53

Maltby N Tolin DF Worhunsky P OrsquoKeefe TM Kiehl KA Dysfunctionalaction monitoring hyperactivates frontal-striatal circuits in obsessive-compulsive disorder an event-related fMRI study Neuroimage 2005 24495ndash503

Manoach DS Ketwaroo GA Polli FE Thakkar KN Barton JJ Goff DCet al Reduced microstructural integrity of the white matter underlyinganterior cingulate cortex is associated with increased saccadic latency inschizophrenia Neuroimage 2007a 37 599ndash610

Manoach DS Lindgren KA Barton JJ Deficient saccadic inhibition inAspergerrsquos disorder and the social-emotional processing disorderJ Neurol Neurosurg Psychiatry 2004 75 1719ndash26

Page 14 of 15 Brain (2008) K NThakkar et al

Manoach DS Lindgren KA Cherkasova MV Goff DC Halpern EFIntriligator J et al Schizophrenic subjects show deficient inhibition butintact task-switching on saccadic tasks Biol Psychiatry 2002 51 816ndash26

Manoach DS Thakkar KN Cain MS Polli FE Edelman JA Fischl B et alNeural activity is modulated by trial history a functional magneticresonance imaging study of the effects of a previous antisaccadeJ Neurosci 2007b 27 1791ndash8

Manoach DS Weintraub S Daffner KR Scinto LF Deficient antisaccadesin the social-emotional processing disorder Neuroreport 1997 8 901ndash5

McDougle CJ Kresch LE Goodman WK Naylor ST Volkmar FRCohen DJ et al A case-controlled study of repetitive thoughts andbehavior in adults with autistic disorder and obsessive-compulsivedisorder Am J Psychiatry 1995 152 772ndash7

McPeek RM Skavenski AA Nakayama K Concurrent processing ofsaccades in visual search Vision Res 2000 40 2499ndash516

Miezin FM Maccotta L Ollinger JM Petersen SE Buckner RLCharacterizing the hemodynamic response effects of presentation ratesampling procedure and the possibility of ordering brain activity basedon relative timing Neuroimage 2000 11 735ndash59

Minshew NJ Luna B Sweeney JA Oculomotor evidence for neocorticalsystems but not cerebellar dysfunction in autism Neurology 1999 52917ndash22

Mokler A Fischer B The recognition and correction of involuntaryprosaccades in an antisaccade task Exp Brain Res 1999 125 511ndash6

Munoz DP Everling S Look away the anti-saccade task and the voluntarycontrol of eye movement Nat Rev Neurosci 2004 5 218ndash28

Murphy DG Daly E Schmitz N Toal F Murphy K Curran S et alCortical serotonin 5-HT2A receptor binding and social communicationin adults with Aspergerrsquos syndrome an in vivo SPECT study Am JPsychiatry 2006 163 934ndash6

Nieuwenhuis S Nielen MM Mol N Hajcak G Veltman DJ Performancemonitoring in obsessive-compulsive disorder Psychiatry Res 2005 134111ndash22

Nieuwenhuis S Ridderinkhof KR Blom J Band GP Kok A Error-relatedbrain potentials are differentially related to awareness of response errorsevidence from an antisaccade task Psychophysiology 2001 38 752ndash60

Ohnishi T Matsuda H Hashimoto T Kunihiro T Nishikawa M Uema Tet al Abnormal regional cerebral blood flow in childhood autism Brain2000 123 (Pt 9) 1838ndash44

Oldfield RC The assessment and analysis of handedness The EdinburghInventory Neuropsychologia 1971 9 97ndash113

Paus T Primate anterior cingulate cortex where motor control drive andcognition interface Nat Rev Neurosci 2001 2 417ndash24

Pfefferbaum A Sullivan EV Hedehus M Lim KO Adalsteinsson EMoseley M Age-related decline in brain white matter anisotropymeasured with spatially corrected echo-planar diffusion tensor imagingMagn Reson Med 2000 44 259ndash68

Pitman RK A cybernetic model of obsessive-compulsive psychopathologyCompr Psychiatry 1987 28 334ndash43

Polli FE Barton JJ Cain MS Thakkar KN Rauch SL Manoach DSRostral and dorsal anterior cingulate cortex make dissociable contribu-tions during antisaccade error commission Proc Natl Acad Sci USA2005 102 15700ndash5

Polli FE Barton JJ Thakkar KN Greve DN Goff DC Rauch SL et alReduced error-related activation in two anterior cingulate circuits is relatedto impaired performance in schizophrenia Brain 2008 131 971ndash86

Polli FE Barton JJ Vangel M Goff DC Iguchi L Manoach DSSchizophrenia patients show intact immediate error-related performanceadjustments on an antisaccade task Schizophr Res 2006 82 191ndash201

Rabbitt PM Errors and error correction in choice-response tasks J ExpPsychol 1966 71 264ndash72

Radant AD Dobie DJ Calkins ME Olincy A Braff DL Cadenhead KSet al Successful multi-site measurement of antisaccade performancedeficits in schizophrenia Schizophr Res 2007 89 320ndash9

Reese TG Heid O Weisskoff RM Wedeen VJ Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echoMagn Reson Med 2003 49 177ndash82

Russell J Jarrold C Error-correction problems in autism evidence for amonitoring impairment J Autism Dev Disord 1998 28 177ndash88

Russell AJ Mataix-Cols D Anson M Murphy DG Obsessions andcompulsions in Asperger syndrome and high-functioning autism Br JPsychiatry 2005 186 525ndash8

Rutter M Le Couteur A Lord C Autism diagnostic interview-revised LosAngeles CA Western Psychological Services 2003

Santesso DL Segalowitz SJ Schmidt LA Error-related electrocorticalresponses are enhanced in children with obsessive-compulsive behaviorsDev Neuropsychol 2006 29 431ndash45

Shafritz KM Dichter GS Baranek GT Belger A The neural circuitrymediating shifts in behavioral response and cognitive set in autism BiolPsychiatry 2007 63 974ndash80

Silk TJ Rinehart N Bradshaw JL Tonge B Egan G OrsquoBoyle MW et alVisuospatial processing and the function of prefrontal-parietal networksin autism spectrum disorders a functional MRI study Am J Psychiatry2006 163 1440ndash3

South M Ozonoff S McMahon WM The relationship between executivefunctioning central coherence and repetitive behaviors in the high-functioning autism spectrum Autism 2007 11 437ndash51

Taylor SF Martis B Fitzgerald KD Welsh RC Abelson JL Liberzon Iet al Medial frontal cortex activity and loss-related responses to errorsJ Neurosci 2006 26 4063ndash70

Taylor SF Stern ER Gehring WJ Neural systems for error monitoringrecent findings and theoretical perspectives Neuroscientist 2007 13160ndash72

Thesen S Heid O Mueller E Schad LR Prospective acquisition correctionfor head motion with image-based tracking for real-time fMRI MagnReson Med 2000 44 457ndash65

Turner M Annotation Repetitive behaviour in autism areview of psychological research J Child Psychol Psychiatry 1999 40839ndash49

Ullsperger M von Cramon DY Neuroimaging of performance monitoringerror detection and beyond Cortex 2004 40 593ndash604

Ursu S Stenger VA Shear MK Jones MR Carter CS Overactiveaction monitoring in obsessive-compulsive disorder evidencefrom functional magnetic resonance imaging Psychol Sci 2003 14347ndash53

Meel CS Heslenfeld DJ Oosterlaan J Sergeant JA Adaptive controldeficits in attention-deficithyperactivity disorder (ADHD) the role oferror processing Psychiatry Res 2007 151 211ndash20

van Veen V Carter CS The timing of action-monitoring processes in theanterior cingulate cortex J Cogn Neurosci 2002 14 593ndash602

Wechsler D Wechsler Abbreviated Scale of intelligence San Antonio TXThe Psychological Corporation 1999

Whalen PJ Bush G McNally RJ Wilhelm S McInerney SC Jenike MAet al The emotional counting Stroop paradigm a functional magneticresonance imaging probe of the anterior cingulate affective division BiolPsychiatry 1998 44 1219ndash28

White K Ashton R Handedness assessment inventory Neuropsychologia1976 14 261ndash4

Zandt F Prior M Kyrios M Repetitive behaviour in children with highfunctioning autism and obsessive compulsive disorder J Autism DevDisord 2007 37 251ndash9

Abnormal response monitoring in autism Brain (2008) Page 15 of 15

Lower FA in the WM underlying ACC was significantlyrelated to higher repetitive behaviour scores in leftsubgenual rACC with a maximum vertex in subcallosalgyrus (Fig 5B Table 2) No other cortical regions showedsignificant relations in underlying WMThe literature on repetitive behaviour in autism supports

a division into lsquolow-levelrsquo repetitive motor behaviour andmore complex behaviours that appear to be driven byideation (for review see Turner 1999) To assess therelative importance of lsquosensorimotorrsquo versus lsquoideationalrsquoforms of repetition in accounting for the observed relationswith ACC activation and structure we first divided eachparticipantsrsquo ADI-R repetitive behaviour score into lsquosenso-rimotorrsquo (ie D3D4 sterotypies mannerisms repetitiveobject use unusual sensory interests) and lsquoideationalrsquo (ieD1D2 narrow restricted interests and rituals) subscores(Hollander et al 2003) On average the ideational subscoreaccounted for 65 14 of the total repetitive behaviourscore We then performed multiple regression analysiswith BOLD or FA value at the ACC maxima of theregression with total score as the dependent variable andthe two subscores as covariates Sensorimotor subscoreswere more strongly related to activation in the ACC(P= 004) than ideational subscores (P= 017) The oppositewas true for FA which showed a stronger correlationswith ideational subscores (P= 005) than sensorimotor

subscores (P= 018) The subscores did not differ signifi-cantly in their contribution to activation or FA and neitherimproved on the relations seen with the total repetitivebehaviour score (activation P= 00009 FA P= 004)suggesting that both subscores contributed to the relationsobserved

DiscussionThe present study demonstrates functional and structuralabnormalities of the ACC in ASD that may contribute torestricted repetitive and stereotyped patterns of behavioura defining feature of ASD Compared with controls ASDparticipants showed increased rACC activation to bothcorrect and error responses and reduced microstructuralintegrity of the WM underlying ACC including in regionsshowing abnormally increased response monitoring activa-tion Moreover in rACC increased activation to correcttrials and reduced FA were related to ratings of rigidrepetitive behaviour These findings support the hypothesisof hyperactive response monitoring in ASD and link it torepetitive behaviour

Relative to controls ASD participants showed reduceddiscrimination between error and correct trials in a rACCregion that responds to errors (Polli et al 2005) This wasprimarily due to an exaggerated response to correct trials

Fig 3 Statistical maps of group differences in fMRI activation at 8 s for contrasts of (A) error versus correct antisaccades (B) correctantisaccades versus fixation (C) error antisaccades versus fixation Statistical maps are displayed on the inflated medial cortical surfacesof the template brain at P4 005 Regions of greater activation in controls are depicted in warm colours greater activation in ASD patientsis depicted in blue The rACC is outlined in red and the dACC is outlined in blue The gray masks cover non-surface regions in whichactivity is displaced Haemodynamic response time course graphs with standard error bars are displayed for the indicated vertices withpeak activation in the group comparisons

Page 8 of 15 Brain (2008) K NThakkar et al

Table 2 Maxima and locations of clusters showing significant group differences and significant relations to ADI-R repetitionscores in ASD

ApproximateTalairach coordinates

Cluster size (mm) x y z t-value (max) CWP

fMRI group comparisonsError versus correctHC4ASDRight rostral anterior cingulate gyrus 488 7 38 12 331 00001

Unmedicated ASD participants 424 7 38 12 218 0008Right medial superior frontal gyrusa 2071 9 50 38 353 0001Correct versus fixationASD4HCLeft pericollosal sulcus (rACC) 413 6 30 6 222 00001

Unmedicated ASD participants 102 6 31 6 159 041Right rostral anterior cingulate gyrusa 1746 7 39 11 347 00001

Unmedicated ASD participants 912 9 39 10 361 00002Pericollosal anterior cingulate gyrus 6 36 6

Error versus fixationASD4HCLeft rostral anterior cingulate gyrus 255 2 26 4 364 002

Unmedicated ASD participants 80 1 26 3 238 078DTI-FA group comparisonbHC4ASDLeft rostral anterior cingulate gyrusa 1598 7 39 6 605 00001Right rostral anterior cingulate gyrusa 6620 7 38 3 564 00001Frontopolar transverse sulcus 20 61 14

Left transverse gyrus frontopolara 6152 20 59 6 72 00001Left precentral gyrusa 1022 47 8 41 511 0009Left intraparietal and parietal transverse sa 4878 40 50 42 505 00001Right intraparietal and parietal transverse sa 1465 21 56 52 406 00008ASD4HCRight anterior circular sulcus insula$ 838 35 31 2 291 005fMRI regressions with ADI-RCorrect versus fixationRight rostral anterior cingulate gyrus 684 6 37 7 323 00001Error versus correctRight paracentral gryrusa 1642 6 25 51 259 0005DTI-FA regressions with ADI-RLeft subcollosal gyrus (rACC) 148 5 18 8 204 005

aMeets correction for entire cortical surface bCWP levels are based on correction for entire cortical surface Important local maximaare indented Results for fMRI group comparisons restricted to unmedicated ASD participants (n=6) are indented and italicized

Fig 4 Statistical map of group differences in FA displayed on the inflated medial cortical surfaces of the template brain at athreshold of P4 005 Regions of greater FA in controls are depicted in warm colours The rACC is outlined in red and the dACC isoutlined in blue

Abnormal response monitoring in autism Brain (2008) Page 9 of 15

rather than a blunted response to errors Reduceddiscrimination between response outcomes may interferewith adjusting responses to obtain the most favourableoutcome both with regard to subsequent trials and longerterm modification of the strength of stimulus-responsemappings This could contribute to behaviour that isstimulus-bound and repetitive rather than flexible andresponsive to contingencies This provides a plausibleexplanation for the relation between increased rACCactivity to correct trials and repetitive behaviourA complementary explanation has been offered for the

relation of increased error signals to the severity ofobsessions and compulsions in OCD (Gehring et al2000 Ursu et al 2003) A longstanding theory of OCDposits that inappropriate and exaggerated error signals leadto a pervasive sense of incompleteness and self-doubt andbehavioural repetition represents a vain attempt to reducethese error signals (Pitman 1987) According to this theoryexaggerated ACC activation to correct trials reflectsinappropriate error signalling and triggers a compulsionto repeat behaviours that were already successfully com-pleted (Maltby et al 2005) Similar to individuals withOCD ASD participants showed an exaggerated ACCresponse to correct trials (Ursu et al 2003 Maltby et al2005) While in OCD fMRI activation to correct trials wasmaximal in dACC (Maltby et al 2005) in ASD it was

maximal in rACC and was related to repetitive behaviourWhile dACC activity following errors is thought to reflecteither conflict or error detection (for review see Ullspergerand von Cramon 2004) rACC activity following errors hasbeen interpreted as appraisal of the affective or motivationalsalience of errors (van Veen and Carter 2002 Luu et al2003 Taylor et al 2006) Thus one interpretation of thisrelation is that it reflects an inappropriate affective responseto correct performance that triggers behavioural repetitionThis exaggerated affective response may occur even withawareness that one has performed correctly Regardless of itsbasis the relation between increased rACC responses tocorrect trials and repetitive behaviour suggests that aberrantresponse monitoring contributes to a defining feature of ASD

Both ASD and OCD are characterized by behaviour thatis rigid and not optimally responsive to contingency andthere is evidence of both phenomenological overlap anddistinctions between the obsessions and compulsions ofOCD and the restricted and repetitive behaviours of ASDboth in terms of subjective experience and type ofbehaviour (Baron-Cohen 1989 McDougle et al 1995Russell et al 2005 Zandt et al 2007) Interestingly a studyof multiplex families of children with autism found thatparents were significantly more likely to have obsessivendashcompulsive traits or disorder if the children had high ADI-Rrepetitive behaviour scores and this was due to loading

Fig 5 Relations of ACC function and structure to restricted repetitive behaviour in ASD (A) Statistical map of regression of rawBOLD signal on ADI-R repetitive behaviour score BOLD signal is from the correct antisaccade versus fixation contrast at 8 s Results aredisplayed on the inflated right medial cortical surface Scatter plot shows BOLD signal from the maximum vertex in right rACC foreach ASD participant on the y-axis and ADI-R score on the x-axis (B) Statistical map of regression of FA on ADI-R repetitive behaviourscore Results are displayed on the inflated left medial surface Scatter plot shows FA from the maximum vertex in left rACC for eachASD participant on the y-axis and ADI-R score on the x-axis The rACC is outlined in red and the dACC is outlined in blue

Page 10 of 15 Brain (2008) K NThakkar et al

on ideational but not sensorimotor items (Hollander et al2003) This finding supports the distinction betweenrepetition of movements and repetition of more complexideational behaviours (for review see Turner 1999) andsuggests that the latter may share underlying mechanisms inOCD and ASD In the present study both sensorimotorand ideational subscores contributed to the relations seenwith response monitoring activation and FA At presentrepetitive behaviours in autism are incompletely under-stood and neurobiologically valid dimensions are yet to bedelineated The present findings suggest that ACC abnorm-alities may contribute to repetitive behaviour in ASDpossibly on the basis of their contribution to aberrantresponse monitoring but linking ACC abnormalities tospecific dimensions of repetitive behaviour and identifyingthe underlying mechanisms requires further study withlarge samplesAbnormal response monitoring may also contribute to

deficient cognitive control of behaviour in ASD which inthe present study was reflected in an increased antisaccadeerror rate consistent with prior work (Manoach et al1997 2004 Minshew et al 1999 Goldberg et al 2002Luna et al 2007) and possibly in reduced dynamicperformance adjustment In spite of making more errorsASD participants responded more quickly than controls oncorrect trials suggesting a disruption in the speed-accuracytrade-off This apparent difficulty in modulating responsespeed to improve accuracy is unlikely to reflect reducederror awareness since ASD participants were as or morelikely than controls to self-correct errors Nor is it likely toreflect a blunted neural response to errors given both theincreased response to errors in left perigenual rACC andprevious work showing normal or increased amplitude ofthe ERN in ASD (Henderson et al 2006) Deficient use ofoutcomes to adjust performance in a continuous dynamicfashion may stem from response monitoring dysfunction inthe ACC andor from dysconnection of the ACC fromocular motor regions involved in saccade generationAberrant connectivity of ACC regions subserving responsemonitoring in ASD is suggested by our finding of reducedFA in the underlying WM This finding is consistent with aprevious report of reduced FA in ACC white matter inautism (Barnea-Goraly et al 2004) and with evidence ofdecreased functional connectivity of the ACC to otherregions during a task involving response inhibition (Kanaet al 2007)How might information concerning the outcome of the

prior response be communicated to ocular motor regionsresponsible for adjusting behaviour in the subsequent trialMonkey electrophysiology studies suggest a basis for speed-accuracy trade-offs in the ocular motor system Comparedwith prosaccades antisaccades are associated with reducedpre-target activity of saccade-related neurons in the frontaleye field (Everling et al 1999 Everling and Munoz 2000)and superior colliculus (Everling et al 1999) key regions insaccade generation Moreover lower pre-target frontal eye

field activity is correlated with increased saccadic latencyand fewer antisaccade errors Reduced preparatory neuralactivity in response to the instruction to perform anantisaccade presumably makes it more difficult for activityrelated to the incorrect but prepotent prosaccade to reachthe threshold for triggering an eye movement However theprice for this improved accuracy is that the activitygenerating the correct antisaccade also takes longer toreach threshold resulting in prolonged saccadic latenciesThus we hypothesize that the slowing of responses thatfollow an antisaccade error (ie post-error slowingNieuwenhuis et al 2001) stems from an ACC errorresponse that signals the frontal eye field to reduce itspreparatory activity during the subsequent trial Thishypothesis is consistent with previous work demonstratingthat other aspects of trial history affect preparatory neuralactivity in the frontal eye field during antisaccades(Manoach et al 2007b) How might error signals beconveyed to frontal eye field The use of outcomes todynamically adjust responses has been attributed to ACCacting with lateral prefrontal cortex (Gehring and Knight2000 Paus 2001 Garavan et al 2002) and lateralprefrontal cortex has been proposed to provide inhibitoryinput to frontal eye field during antisaccades (DeSouzaet al 2003) Thus in ASD aberrant communicationbetween response evaluation in the ACC and regionsinitiating compensatory adjustments possibly lateral pre-frontal cortex could result in a failure to optimally reducepreparatory firing rates in frontal eye field in response toa prior error This reduced dynamic performance adjust-ment could lead to faster responding and more errorsWe are presently testing this hypothesis using an anti-saccade paradigm that gives rise to post-error slowing(Rabbitt 1966) and will allow a more direct examinationof compensatory behaviour and its supporting neuralcircuitry Unfortunately although post-error slowing isseen when antisaccades are the only task (Polli et al 2006)the intermixed antisaccade and prosaccade paradigmemployed here gives rise to other inter-trial factors thatalso affect latency (eg task-switching and prior antisac-cades Cherkasova et al 2002 Fecteau et al 2004 Bartonet al 2006 Manoach et al 2007b) and may obscure theeffects of a prior error There is evidence of reduced post-error slowing in adults with high functioning autismhowever from a previous study that employed a memorysearch task (Bogte et al 2007) Generalizing beyond theocular motor system failures of dynamic performanceadjustment due to ACC dysfunction and dysconnectivitymay contribute to the rigid and perseverative behavioursthat characterize ASD

Repetitive behaviour scores in ASD were also associatedwith reduced FA in the WM underlying left subgenualrACC Subgenual rACC is part of a network that is thoughtto play an important role in generating autonomicresponses to motivationally salient stimuli evaluatingtheir significance in predicting reward and in using this

Abnormal response monitoring in autism Brain (2008) Page 11 of 15

information to guide behaviour (Damasio et al 1990Drevets 2000) An important consideration in interpretingthis relation of FA to repetition is that subgenual rACCregion was not functionally or structurally abnormal in thisASD sample This suggests that even normal variation in FAof the subgenual rACC may contribute to a tendency torepeat unreinforced stimulus-response mappings Moregenerally the relations we observed between behaviouralrepetition and ACC function and structure may not bespecific to ASD Accumulating evidence suggests that thethree symptom clusters that define autismmdashsocial impair-ment communication deficits and restricted repetitivebehaviourmdasharise from distinct genetic and cognitivemechanisms and may represent continuous traits in thepopulation (for reviews see Happe et al 2006 London2007) Previous studies have reported relations betweenindices of response monitoring and repetition in OCD(Gehring et al 2000 Ursu et al 2003) and in non-clinically ascertained samples of children and adults withobsessivendashcompulsive characteristics (Hajcak and Simons2002 Santesso et al 2006)Our findings also raise the question of whether the

response monitoring abnormalities that we observed arespecific to ASD Similar abnormalities have been reportedin OCD (Maltby et al 2005) and attention deficithyperactivity disorder (van Meel et al 2007) Whilemethodological differences preclude a direct comparisonof these findings we previously studied patients withschizophrenia using the identical paradigm (Polli et al2008) Schizophrenia is also a neurodevelopmental disorderthat is characterized by rigid and perseverative behaviourresponse monitoring abnormalities (Carter et al 2001)reduced FA in ACC WM (Manoach et al 2007a) and aconsistently elevated antisaccade error rate (Radant et al2007 Manoach et al 2002) Although both schizophreniaand ASD samples made more antisaccade errors thancontrols ASD participants performed correct trials fasterand schizophrenia patients performed more slowly Whileboth groups showed reduced discrimination between errorand correct trials in the ACC this occurred for differentreasons In ASD this reflected a hyperactive response tocorrect trials while in schizophrenia it reflected a bluntedACC response to errors consistent with numerous previousreports (Laurens et al 2003) These differences argue fordistinct behavioural and neural signatures of responsemonitoring deficits in schizophrenia and ASD Thesefindings along with those of recent neuroimaging studiessuggesting genetic mediation of response monitoring(Fallgatter et al 2004 Klein et al 2007 Kramer et al2007) support the candidacy of neuroimaging-basedindices of response monitoring as neurocognitive endophe-notypes Whether response monitoring abnormalities canbe distinguished in ASD OCD and other neuropsychiatricdisorders whether they are related to diagnosis or specifictraits and whether they are also present in unaffectedfamily members remains to be seen

Several additional limitations to and alternative inter-pretations of our findings merit consideration First mightthe hyperactive rACC response to correct antisaccades inASD reflect a less efficient response to conflict ratherthan response monitoring The ACC is thought to play arole in detecting conflict and engaging cognitive controlto resolve it (for review see Carter and van Veen 2007)Antisaccades involve a high degree of conflict as one mustchoose between two competing responses and inhibit theprepotent one Errors give rise to another type of conflictwhen the actual response is compared with the intendedone Thus both correct and erroneous antisaccadesengender conflict While it is not possible to entirely ruleout the conflict explanation several observations argueagainst it If this rACC region were detecting conflict itshould respond to both correct and error trials Insteadcontrols showed a significant response only to errors Inaddition the timing of the haemodynamic response tocorrect antisaccades in ASD which started amp4 s and peakedat amp8 s (Fig 3) argues for a role in response monitoringrather than successful conflict resolution In this paradigmdifferential fMRI activation for correct antisaccade versusprosaccade trials including in dACC presumably reflectingconflict detection response inhibition and saccade genera-tion peaks at amp4 s and is mostly gone by 8 s (Manoachet al 2007b see also Polli et al 2005 for a discussionof ACC regions involved in saccade generation versusresponse monitoring) Finally it is dorsal rather than rACCactivation that is associated with engaging cognitive controlon correct high-conflict and error trials (Carter and vanVeen 2007)

An important limitation is that the small sample size ofthe present study may have compromised our power todetect additional group differences and relations of interestparticularly in regions outside our a priori ACC regions ofinterest for which more stringent multiple comparisonscorrections were applied In spite of the relatively smallsample size our a priori hypotheses were confirmedIn addition four of our ASD participants were takingmedications that could affect ACC function A re-analysisof the fMRI data with only unmedicated ASD participantshowever supported our main findings of significantly lessdiscrimination between response outcomes in the rACCand greater activation to correct trials It is also important tonote that our sample was limited to high functioning adultswith ASD so it is not clear that our findings would generalizeto lower functioning or younger samples Although increasedantisaccade error rates are seen in autism as early as ages 8ndash12(Luna et al 2007) we limited our study to adults sincesaccadic inhibition may not fully develop until late adoles-cence (Klein and Foerster 2001) and larger samples would benecessary to discriminate between the effects of diagnosis andthose due to normal development Another limitationconcerns our interpretation of FA as reflecting connectivityFA reflects the degree of directional coherence of waterdiffusion in tissue and is therefore an indirect measure of

Page 12 of 15 Brain (2008) K NThakkar et al

WMmicrostructure While FA correlates with axon myelinat-ion (Harsan et al 2006) not all of its biophysicaldeterminants are fully understood (Beaulieu 2002) Thus itis not possible to attribute FA differences or the relations weobserved to a particular WM propertyIn summary the present study reports hyperactive

response monitoring in the rACC in ASD that is relatedto restricted repetitive behaviour and is accompanied byreduced FA in the underlying white matter This suggests astructural correlate to abnormal function that may com-promise connectivity in the circuitry subserving responsemonitoring These findings suggest that structural and fun-ctional abnormalities of the ACC compromise responsemonitoring and contribute to behavioural repetition inASD Impairments in evaluating and learning from errorsmay lead to behaviour that is rigid and repetitive ratherthan optimally guided by outcomes and may compromiseperformance across a wide range of tasks Illuminating theneural basis and clinical significance of response monitoringdeficits in ASD represents a fruitful avenue for furtherresearch

Supplementary materialSupplementary material is available at Brain online

AcknowledgementsThe authors wish to thank Mark Vangel for statisticalconsultation Robert Levy Kim Ono Christopher Shermanand Jeremy Young for help with manuscript preparationand Matt Cain and Jay Edelman for technical assistanceSome of these data were presented at the annual meetingof the Society for Neuroscience in Washington DC inNovember 2005 This study was supported by NationalInstitute for Mental Health [R01 MH67720 (DSM)MH72120 (FEP)] Mental Illness Neuroscience Discovery(MIND) Institute (DOE DE-FG02-99ER62764) TheNational Center for Research Resources (P41RR14075)Funding to pay the Open Access publication charges forthis article was provided by the National Institute forMental Health NRSA MH72120

ReferencesAshwin C Baron-Cohen S Wheelwright S OrsquoRiordan M Bullmore ETDifferential activation of the amygdala and the lsquosocial brainrsquo duringfearful face-processing in Asperger syndrome Neuropsychologia 200745 2ndash14

Barnea-Goraly N Kwon H Menon V Eliez S Lotspeich L Reiss ALWhite matter structure in autism preliminary evidence from diffusiontensor imaging Biol Psychiatry 2004 55 323ndash6

Baron-Cohen S Do autistic children have obsessions and compulsionsBr J Clin Psychol 1989 28 (Pt 3) 193ndash200

Barton JJ Greenzang C Hefter R Edelman J Manoach DS Switchingplasticity and prediction in a saccadic task-switch paradigm Exp BrainRes 2006 168 76ndash87

Basser PJ Mattiello J LeBihan D MR diffusion tensor spectroscopy andimaging Biophys J 1994 66 259ndash67

Beaulieu C The basis of anisotropic water diffusion in the nervous system- a technical review NMR Biomed 2002 15 435ndash55

Blair JR Spreen O Predicting premorbid IQ a revision of the NationalAdult Reading Test Clin Neuropsychol 1989 3 129ndash136

Bogte H Flamma B van der Meere J van Engeland H Post-erroradaptation in adults with high functioning autism Neuropsychologia2007 45 1707ndash14

Breiter HC Rauch SL Kwong KK Baker JR Weisskoff RM Kennedy DNet al Functional magnetic resonance imaging of symptom provocationin obsessive-compulsive disorder Arch Gen Psychiatry 1996 53595ndash606

Buckner RL Goodman J Burock M Rotte M Koutstaal W Schacter Det al Functional-anatomic correlates of object priming in humansrevealed by rapid presentation event-related fMRI Neuron 1998 20285ndash96

Burock MA Dale AM Estimation and detection of event-related fMRIsignals with temporally correlated noise a statistically efficient andunbiased approach Hum Brain Mapp 2000 11 249ndash60

Bush G Whalen PJ Rosen BR Jenike MA McInerney SC Rauch SL Thecounting stroop an interference task specialized for functionalneuroimaging- -validation study with functional MRI Hum BrainMapp 1998 6 270ndash82

Carter CS MacDonald AW 3rd Ross LL Stenger VA Anterior cingulatecortex activity and impaired self-monitoring of performance in patientswith schizophrenia an event-related fMRI study Am J Psychiatry 2001158 1423ndash8

Carter CS van Veen V Anterior cingulate cortex and conflict detection anupdate of theory and data Cogn Affect Behav Neurosci 2007 7 367ndash79

Cherkasova MV Manoach DS Intriligator JM Barton JJS Antisaccadesand task-switching interactions in controlled processing Exp Brain Res2002 144 528ndash37

Collins DL Neelin P Peters TM Evans AC Automatic 3D intersubjectregistration of MR volumetric data in standardized Talairach spaceJ Comput Assist Tomogr 1994 18 192ndash205

Courchesne E Pierce K Why the frontal cortex in autism might be talkingonly to itself local over-connectivity but long-distance disconnectionCurr Opin Neurobiol 2005 15 225ndash30

Cox RW Jesmanowicz A Real-time 3D image registration for functionalMRI Magn Reson Med 1999 42 1014ndash8

Dale AM Optimal experimental design for event-related fMRI Hum BrainMapp 1999 8 109ndash40

Dale AM Fischl B Sereno MI Cortical surface-based analysisI Segmentation and surface reconstruction Neuroimage 1999 9 179ndash94

Damasio AR Tranel D Damasio H Individuals with sociopathic behaviorcaused by frontal damage fail to respond autonomically to social stimuliBehav Brain Res 1990 41 81ndash94

Dehaene S Posner MI Tucker DM Localization of a neural system forerror detection and compensation Psychol Sci 1994 5 303ndash5

DeSouza JF Menon RS Everling S Preparatory set associated with pro-saccades and anti-saccades in humans investigated with event-relatedFMRI J Neurophysiol 2003 89 1016ndash23

Devinsky O Morrell MJ Vogt BA Contributions of anterior cingulatecortex to behaviour Brain 1995 118 (Pt 1) 279ndash306

Dichter GS Belger A Social stimuli interfere with cognitive control inautism Neuroimage 2007 35 1219ndash30

Dougherty DD Baer L Cosgrove GR Cassem EH Price BHNierenberg AA et al Prospective long-term follow-up of 44 patientswho received cingulotomy for treatment-refractory obsessive-compulsivedisorder Am J Psychiatry 2002 159 269ndash75

Drevets WC Neuroimaging studies of mood disorders Biol Psychiatry2000 48 813ndash29

Everling S Dorris MC Klein RM Munoz DP Role of primate superiorcolliculus in preparation and execution of anti-saccades and pro-saccades J Neurosci 1999 19 2740ndash54

Everling S Munoz DP Neuronal correlates for preparatory set associatedwith pro-saccades and anti-saccades in the primate frontal eye fieldJ Neurosci 2000 20 387ndash400

Abnormal response monitoring in autism Brain (2008) Page 13 of 15

Falkenstein M Hohnsbein J Hoormann J Event-related correlates of errorsin reaction tasks In Karmos G Molnar M Csepe I Desmedt JE editorsPerspectives of event-related potentials research AmsterdamThe Netherlands Elsevier Science 1995 p 287ndash96

Fallgatter AJ Herrmann MJ Roemmler J Ehlis AC Wagener AHeidrich A et al Allelic variation of serotonin transporter functionmodulates the brain electrical response for error processingNeuropsychopharmacology 2004 29 1506ndash11

Fecteau JH Au C Armstrong IT Munoz DP Sensory biases producealternation advantage found in sequential saccadic eye movement tasksExp Brain Res 2004 159 84ndash91

Fischer B Breitmeyer B Mechanisms of visual attention revealed bysaccadic eye movements Neuropsychologia 1987 25 73ndash83

Fischl B Sereno MI Dale AM Cortical surface-based analysis II Inflationflattening and a surface-based coordinate system Neuroimage 1999a 9195ndash207

Fischl B Sereno MI Tootell RBH Dale AM High-resolution intersubjectaveraging and a coordinate system for the cortical surface Hum BrainMapp 1999b 8 272ndash84

Fischl B van der Kouwe A Destrieux C Halgren E Segonne F Salat DHet al Automatically parcellating the human cerebral cortex CerebCortex 2004 14 11ndash22

Fitzgerald KD Welsh RC Gehring WJ Abelson JL Himle JA Liberzon Iet al Error-related hyperactivity of the anterior cingulate cortex inobsessive-compulsive disorder Biol Psychiatry 2005 57 287ndash94

Garavan H Ross TJ Murphy K Roche RA Stein EA Dissociable executivefunctions in the dynamic control of behavior inhibition errordetection and correction NeuroImage 2002 17 1820ndash9

Gehring WJ Goss B Coles MG Meyer DE Donchin E A neuralsystem for error detection and compensation Psychol Sci 1993 4385ndash90

Gehring WJ Himle J Nisenson LG Action-monitoring dysfunction inobsessive-compulsive disorder Psychol Sci 2000 11 1ndash6

Gehring WJ Knight RT Prefrontal-cingulate interactions in actionmonitoring Nat Neurosci 2000 3 516ndash20

Goldberg MC Lasker AG Zee DS Garth E Tien A Landa RJ Deficits inthe initiation of eye movements in the absence of a visual target inadolescents with high functioning autism Neuropsychologia 2002 402039ndash49

Gomot M Bernard FA Davis MH Belmonte MK Ashwin CBullmore ET et al Change detection in children with autism anauditory event-related fMRI study Neuroimage 2006 29 475ndash84

Hajcak G Simons RF Error-related brain activity in obsessive-compulsiveundergraduates Psychiatry Res 2002 110 63ndash72

Hall GB Szechtman H Nahmias C Enhanced salience and emotionrecognition in autism a PET study Am J Psychiatry 2003 160 1439ndash41

Hallett PE Primary and secondary saccades to goals defined byinstructions Vision Res 1978 18 1279ndash96

Happe F Ronald A Plomin R Time to give up on a single explanation forautism Nat Neurosci 2006 9 1218ndash20

Harsan LA Poulet P Guignard B Steibel J Parizel N de Sousa PL et alBrain dysmyelination and recovery assessment by noninvasive in vivodiffusion tensor magnetic resonance imaging J Neurosci Res 2006 83392ndash402

Haznedar MM Buchsbaum MS Metzger M Solimando A Spiegel-Cohen J Hollander E Anterior cingulate gyrus volume and glucosemetabolism in autistic disorder Am J Psychiatry 1997 154 1047ndash50

Haznedar MM Buchsbaum MS Wei TC Hof PR Cartwright CBienstock CA et al Limbic circuitry in patients with autism spectrumdisorders studied with positron emission tomography and magneticresonance imaging Am J Psychiatry 2000 157 1994ndash2001

Henderson H Schwartz C Mundy P Burnette C Sutton S Zahka N et alResponse monitoring the error-related negativity and differences insocial behavior in autism Brain Cogn 2006 61 96ndash109

Hill EL Executive dysfunction in autism Trends Cogn Sci 2004 8 26ndash32

Hollander E King A Delaney K Smith CJ Silverman JM Obsessive-compulsive behaviors in parents of multiplex autism families PsychiatryRes 2003 117 11ndash6

Hollingshead AB Two factor index of social position New Haven CTYale University Press 1965

Holroyd CB Coles MG The neural basis of human error processingreinforcement learning dopamine and the error-related negativityPsychol Rev 2002 109 679ndash709

Jenkinson M Smith S A global optimisation method for robust affineregistration of brain images Med Image Anal 2001 5 143ndash56

Johannes S Wieringa BM Nager W Rada D Dengler R Emrich HMet al Discrepant target detection and action monitoring in obsessive-compulsive disorder Psychiatry Res 2001 108 101ndash10

Jones DK The effect of gradient sampling schemes on measures derivedfrom diffusion tensor MRI A Monte Carlo study Magn Reson Med2004 51 807ndash15

Kana RK Keller TA Minshew NJ Just MA Inhibitory control in high-functioning autism decreased activation and underconnectivity ininhibition networks Biol Psychiatry 2007 62 198ndash206

Kennedy DP Redcay E Courchesne E Failing to deactivate restingfunctional abnormalities in autism Proc Natl Acad Sci USA 2006 1038275ndash80

Klein C Foerster F Development of prosaccade and antisaccade taskperformance in participants aged 6 to 26 years Psychophysiology 200138 179ndash89

Klein TA Neumann J Reuter M Hennig J von Cramon DYUllsperger M Genetically determined differences in learning fromerrors Science 2007 318 1642ndash5

Kramer UM Cunillera T Camara E Marco-Pallares J Cucurell DNager W et al The impact of catechol-O-methyltransferase anddopamine D4 receptor genotypes on neurophysiological markers ofperformance monitoring J Neurosci 2007 27 14190ndash8

Laurens KR Ngan ET Bates AT Kiehl KA Liddle PF Rostral anteriorcingulate cortex dysfunction during error processing in schizophreniaBrain 2003 126 610ndash22

Levitt JG OrsquoNeill J Blanton RE Smalley S Fadale D McCracken JT et alProton magnetic resonance spectroscopic imaging of the brain inchildhood autism Biol Psychiatry 2003 54 1355ndash66

Leyfer OT Folstein SE Bacalman S Davis NO Dinh E Morgan J et alComorbid psychiatric disorders in children with autism interview devel-opment and rates of disorders J Autism Dev Disord 2006 36 849ndash61

London E The role of the neurobiologist in redefining the diagnosis ofautism Brain Pathol 2007 17 408ndash11

Lopez BR Lincoln AJ Ozonoff S Lai Z Examining the relationshipbetween executive functions and restricted repetitive symptoms ofAutistic Disorder J Autism Dev Disord 2005 35 445ndash60

Lord C Rutter M DiLavore PC Risi S Autism diagnostic observationschedule - WPS (ADOS-WPS) Los Angeles CA Western PsychologicalServices 1999

Luna B Doll SK Hegedus SJ Minshew NJ Sweeney JA Maturation ofexecutive function in autism Biol Psychiatry 2007 61 474ndash81

Luu P Tucker DM Derryberry D Reed M Poulsen C Electrophysiolog-ical responses to errors and feedback in the process of action regulationPsychol Sci 2003 14 47ndash53

Maltby N Tolin DF Worhunsky P OrsquoKeefe TM Kiehl KA Dysfunctionalaction monitoring hyperactivates frontal-striatal circuits in obsessive-compulsive disorder an event-related fMRI study Neuroimage 2005 24495ndash503

Manoach DS Ketwaroo GA Polli FE Thakkar KN Barton JJ Goff DCet al Reduced microstructural integrity of the white matter underlyinganterior cingulate cortex is associated with increased saccadic latency inschizophrenia Neuroimage 2007a 37 599ndash610

Manoach DS Lindgren KA Barton JJ Deficient saccadic inhibition inAspergerrsquos disorder and the social-emotional processing disorderJ Neurol Neurosurg Psychiatry 2004 75 1719ndash26

Page 14 of 15 Brain (2008) K NThakkar et al

Manoach DS Lindgren KA Cherkasova MV Goff DC Halpern EFIntriligator J et al Schizophrenic subjects show deficient inhibition butintact task-switching on saccadic tasks Biol Psychiatry 2002 51 816ndash26

Manoach DS Thakkar KN Cain MS Polli FE Edelman JA Fischl B et alNeural activity is modulated by trial history a functional magneticresonance imaging study of the effects of a previous antisaccadeJ Neurosci 2007b 27 1791ndash8

Manoach DS Weintraub S Daffner KR Scinto LF Deficient antisaccadesin the social-emotional processing disorder Neuroreport 1997 8 901ndash5

McDougle CJ Kresch LE Goodman WK Naylor ST Volkmar FRCohen DJ et al A case-controlled study of repetitive thoughts andbehavior in adults with autistic disorder and obsessive-compulsivedisorder Am J Psychiatry 1995 152 772ndash7

McPeek RM Skavenski AA Nakayama K Concurrent processing ofsaccades in visual search Vision Res 2000 40 2499ndash516

Miezin FM Maccotta L Ollinger JM Petersen SE Buckner RLCharacterizing the hemodynamic response effects of presentation ratesampling procedure and the possibility of ordering brain activity basedon relative timing Neuroimage 2000 11 735ndash59

Minshew NJ Luna B Sweeney JA Oculomotor evidence for neocorticalsystems but not cerebellar dysfunction in autism Neurology 1999 52917ndash22

Mokler A Fischer B The recognition and correction of involuntaryprosaccades in an antisaccade task Exp Brain Res 1999 125 511ndash6

Munoz DP Everling S Look away the anti-saccade task and the voluntarycontrol of eye movement Nat Rev Neurosci 2004 5 218ndash28

Murphy DG Daly E Schmitz N Toal F Murphy K Curran S et alCortical serotonin 5-HT2A receptor binding and social communicationin adults with Aspergerrsquos syndrome an in vivo SPECT study Am JPsychiatry 2006 163 934ndash6

Nieuwenhuis S Nielen MM Mol N Hajcak G Veltman DJ Performancemonitoring in obsessive-compulsive disorder Psychiatry Res 2005 134111ndash22

Nieuwenhuis S Ridderinkhof KR Blom J Band GP Kok A Error-relatedbrain potentials are differentially related to awareness of response errorsevidence from an antisaccade task Psychophysiology 2001 38 752ndash60

Ohnishi T Matsuda H Hashimoto T Kunihiro T Nishikawa M Uema Tet al Abnormal regional cerebral blood flow in childhood autism Brain2000 123 (Pt 9) 1838ndash44

Oldfield RC The assessment and analysis of handedness The EdinburghInventory Neuropsychologia 1971 9 97ndash113

Paus T Primate anterior cingulate cortex where motor control drive andcognition interface Nat Rev Neurosci 2001 2 417ndash24

Pfefferbaum A Sullivan EV Hedehus M Lim KO Adalsteinsson EMoseley M Age-related decline in brain white matter anisotropymeasured with spatially corrected echo-planar diffusion tensor imagingMagn Reson Med 2000 44 259ndash68

Pitman RK A cybernetic model of obsessive-compulsive psychopathologyCompr Psychiatry 1987 28 334ndash43

Polli FE Barton JJ Cain MS Thakkar KN Rauch SL Manoach DSRostral and dorsal anterior cingulate cortex make dissociable contribu-tions during antisaccade error commission Proc Natl Acad Sci USA2005 102 15700ndash5

Polli FE Barton JJ Thakkar KN Greve DN Goff DC Rauch SL et alReduced error-related activation in two anterior cingulate circuits is relatedto impaired performance in schizophrenia Brain 2008 131 971ndash86

Polli FE Barton JJ Vangel M Goff DC Iguchi L Manoach DSSchizophrenia patients show intact immediate error-related performanceadjustments on an antisaccade task Schizophr Res 2006 82 191ndash201

Rabbitt PM Errors and error correction in choice-response tasks J ExpPsychol 1966 71 264ndash72

Radant AD Dobie DJ Calkins ME Olincy A Braff DL Cadenhead KSet al Successful multi-site measurement of antisaccade performancedeficits in schizophrenia Schizophr Res 2007 89 320ndash9

Reese TG Heid O Weisskoff RM Wedeen VJ Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echoMagn Reson Med 2003 49 177ndash82

Russell J Jarrold C Error-correction problems in autism evidence for amonitoring impairment J Autism Dev Disord 1998 28 177ndash88

Russell AJ Mataix-Cols D Anson M Murphy DG Obsessions andcompulsions in Asperger syndrome and high-functioning autism Br JPsychiatry 2005 186 525ndash8

Rutter M Le Couteur A Lord C Autism diagnostic interview-revised LosAngeles CA Western Psychological Services 2003

Santesso DL Segalowitz SJ Schmidt LA Error-related electrocorticalresponses are enhanced in children with obsessive-compulsive behaviorsDev Neuropsychol 2006 29 431ndash45

Shafritz KM Dichter GS Baranek GT Belger A The neural circuitrymediating shifts in behavioral response and cognitive set in autism BiolPsychiatry 2007 63 974ndash80

Silk TJ Rinehart N Bradshaw JL Tonge B Egan G OrsquoBoyle MW et alVisuospatial processing and the function of prefrontal-parietal networksin autism spectrum disorders a functional MRI study Am J Psychiatry2006 163 1440ndash3

South M Ozonoff S McMahon WM The relationship between executivefunctioning central coherence and repetitive behaviors in the high-functioning autism spectrum Autism 2007 11 437ndash51

Taylor SF Martis B Fitzgerald KD Welsh RC Abelson JL Liberzon Iet al Medial frontal cortex activity and loss-related responses to errorsJ Neurosci 2006 26 4063ndash70

Taylor SF Stern ER Gehring WJ Neural systems for error monitoringrecent findings and theoretical perspectives Neuroscientist 2007 13160ndash72

Thesen S Heid O Mueller E Schad LR Prospective acquisition correctionfor head motion with image-based tracking for real-time fMRI MagnReson Med 2000 44 457ndash65

Turner M Annotation Repetitive behaviour in autism areview of psychological research J Child Psychol Psychiatry 1999 40839ndash49

Ullsperger M von Cramon DY Neuroimaging of performance monitoringerror detection and beyond Cortex 2004 40 593ndash604

Ursu S Stenger VA Shear MK Jones MR Carter CS Overactiveaction monitoring in obsessive-compulsive disorder evidencefrom functional magnetic resonance imaging Psychol Sci 2003 14347ndash53

Meel CS Heslenfeld DJ Oosterlaan J Sergeant JA Adaptive controldeficits in attention-deficithyperactivity disorder (ADHD) the role oferror processing Psychiatry Res 2007 151 211ndash20

van Veen V Carter CS The timing of action-monitoring processes in theanterior cingulate cortex J Cogn Neurosci 2002 14 593ndash602

Wechsler D Wechsler Abbreviated Scale of intelligence San Antonio TXThe Psychological Corporation 1999

Whalen PJ Bush G McNally RJ Wilhelm S McInerney SC Jenike MAet al The emotional counting Stroop paradigm a functional magneticresonance imaging probe of the anterior cingulate affective division BiolPsychiatry 1998 44 1219ndash28

White K Ashton R Handedness assessment inventory Neuropsychologia1976 14 261ndash4

Zandt F Prior M Kyrios M Repetitive behaviour in children with highfunctioning autism and obsessive compulsive disorder J Autism DevDisord 2007 37 251ndash9

Abnormal response monitoring in autism Brain (2008) Page 15 of 15

Table 2 Maxima and locations of clusters showing significant group differences and significant relations to ADI-R repetitionscores in ASD

ApproximateTalairach coordinates

Cluster size (mm) x y z t-value (max) CWP

fMRI group comparisonsError versus correctHC4ASDRight rostral anterior cingulate gyrus 488 7 38 12 331 00001

Unmedicated ASD participants 424 7 38 12 218 0008Right medial superior frontal gyrusa 2071 9 50 38 353 0001Correct versus fixationASD4HCLeft pericollosal sulcus (rACC) 413 6 30 6 222 00001

Unmedicated ASD participants 102 6 31 6 159 041Right rostral anterior cingulate gyrusa 1746 7 39 11 347 00001

Unmedicated ASD participants 912 9 39 10 361 00002Pericollosal anterior cingulate gyrus 6 36 6

Error versus fixationASD4HCLeft rostral anterior cingulate gyrus 255 2 26 4 364 002

Unmedicated ASD participants 80 1 26 3 238 078DTI-FA group comparisonbHC4ASDLeft rostral anterior cingulate gyrusa 1598 7 39 6 605 00001Right rostral anterior cingulate gyrusa 6620 7 38 3 564 00001Frontopolar transverse sulcus 20 61 14

Left transverse gyrus frontopolara 6152 20 59 6 72 00001Left precentral gyrusa 1022 47 8 41 511 0009Left intraparietal and parietal transverse sa 4878 40 50 42 505 00001Right intraparietal and parietal transverse sa 1465 21 56 52 406 00008ASD4HCRight anterior circular sulcus insula$ 838 35 31 2 291 005fMRI regressions with ADI-RCorrect versus fixationRight rostral anterior cingulate gyrus 684 6 37 7 323 00001Error versus correctRight paracentral gryrusa 1642 6 25 51 259 0005DTI-FA regressions with ADI-RLeft subcollosal gyrus (rACC) 148 5 18 8 204 005

aMeets correction for entire cortical surface bCWP levels are based on correction for entire cortical surface Important local maximaare indented Results for fMRI group comparisons restricted to unmedicated ASD participants (n=6) are indented and italicized

Fig 4 Statistical map of group differences in FA displayed on the inflated medial cortical surfaces of the template brain at athreshold of P4 005 Regions of greater FA in controls are depicted in warm colours The rACC is outlined in red and the dACC isoutlined in blue

Abnormal response monitoring in autism Brain (2008) Page 9 of 15

rather than a blunted response to errors Reduceddiscrimination between response outcomes may interferewith adjusting responses to obtain the most favourableoutcome both with regard to subsequent trials and longerterm modification of the strength of stimulus-responsemappings This could contribute to behaviour that isstimulus-bound and repetitive rather than flexible andresponsive to contingencies This provides a plausibleexplanation for the relation between increased rACCactivity to correct trials and repetitive behaviourA complementary explanation has been offered for the

relation of increased error signals to the severity ofobsessions and compulsions in OCD (Gehring et al2000 Ursu et al 2003) A longstanding theory of OCDposits that inappropriate and exaggerated error signals leadto a pervasive sense of incompleteness and self-doubt andbehavioural repetition represents a vain attempt to reducethese error signals (Pitman 1987) According to this theoryexaggerated ACC activation to correct trials reflectsinappropriate error signalling and triggers a compulsionto repeat behaviours that were already successfully com-pleted (Maltby et al 2005) Similar to individuals withOCD ASD participants showed an exaggerated ACCresponse to correct trials (Ursu et al 2003 Maltby et al2005) While in OCD fMRI activation to correct trials wasmaximal in dACC (Maltby et al 2005) in ASD it was

maximal in rACC and was related to repetitive behaviourWhile dACC activity following errors is thought to reflecteither conflict or error detection (for review see Ullspergerand von Cramon 2004) rACC activity following errors hasbeen interpreted as appraisal of the affective or motivationalsalience of errors (van Veen and Carter 2002 Luu et al2003 Taylor et al 2006) Thus one interpretation of thisrelation is that it reflects an inappropriate affective responseto correct performance that triggers behavioural repetitionThis exaggerated affective response may occur even withawareness that one has performed correctly Regardless of itsbasis the relation between increased rACC responses tocorrect trials and repetitive behaviour suggests that aberrantresponse monitoring contributes to a defining feature of ASD

Both ASD and OCD are characterized by behaviour thatis rigid and not optimally responsive to contingency andthere is evidence of both phenomenological overlap anddistinctions between the obsessions and compulsions ofOCD and the restricted and repetitive behaviours of ASDboth in terms of subjective experience and type ofbehaviour (Baron-Cohen 1989 McDougle et al 1995Russell et al 2005 Zandt et al 2007) Interestingly a studyof multiplex families of children with autism found thatparents were significantly more likely to have obsessivendashcompulsive traits or disorder if the children had high ADI-Rrepetitive behaviour scores and this was due to loading

Fig 5 Relations of ACC function and structure to restricted repetitive behaviour in ASD (A) Statistical map of regression of rawBOLD signal on ADI-R repetitive behaviour score BOLD signal is from the correct antisaccade versus fixation contrast at 8 s Results aredisplayed on the inflated right medial cortical surface Scatter plot shows BOLD signal from the maximum vertex in right rACC foreach ASD participant on the y-axis and ADI-R score on the x-axis (B) Statistical map of regression of FA on ADI-R repetitive behaviourscore Results are displayed on the inflated left medial surface Scatter plot shows FA from the maximum vertex in left rACC for eachASD participant on the y-axis and ADI-R score on the x-axis The rACC is outlined in red and the dACC is outlined in blue

Page 10 of 15 Brain (2008) K NThakkar et al

on ideational but not sensorimotor items (Hollander et al2003) This finding supports the distinction betweenrepetition of movements and repetition of more complexideational behaviours (for review see Turner 1999) andsuggests that the latter may share underlying mechanisms inOCD and ASD In the present study both sensorimotorand ideational subscores contributed to the relations seenwith response monitoring activation and FA At presentrepetitive behaviours in autism are incompletely under-stood and neurobiologically valid dimensions are yet to bedelineated The present findings suggest that ACC abnorm-alities may contribute to repetitive behaviour in ASDpossibly on the basis of their contribution to aberrantresponse monitoring but linking ACC abnormalities tospecific dimensions of repetitive behaviour and identifyingthe underlying mechanisms requires further study withlarge samplesAbnormal response monitoring may also contribute to

deficient cognitive control of behaviour in ASD which inthe present study was reflected in an increased antisaccadeerror rate consistent with prior work (Manoach et al1997 2004 Minshew et al 1999 Goldberg et al 2002Luna et al 2007) and possibly in reduced dynamicperformance adjustment In spite of making more errorsASD participants responded more quickly than controls oncorrect trials suggesting a disruption in the speed-accuracytrade-off This apparent difficulty in modulating responsespeed to improve accuracy is unlikely to reflect reducederror awareness since ASD participants were as or morelikely than controls to self-correct errors Nor is it likely toreflect a blunted neural response to errors given both theincreased response to errors in left perigenual rACC andprevious work showing normal or increased amplitude ofthe ERN in ASD (Henderson et al 2006) Deficient use ofoutcomes to adjust performance in a continuous dynamicfashion may stem from response monitoring dysfunction inthe ACC andor from dysconnection of the ACC fromocular motor regions involved in saccade generationAberrant connectivity of ACC regions subserving responsemonitoring in ASD is suggested by our finding of reducedFA in the underlying WM This finding is consistent with aprevious report of reduced FA in ACC white matter inautism (Barnea-Goraly et al 2004) and with evidence ofdecreased functional connectivity of the ACC to otherregions during a task involving response inhibition (Kanaet al 2007)How might information concerning the outcome of the

prior response be communicated to ocular motor regionsresponsible for adjusting behaviour in the subsequent trialMonkey electrophysiology studies suggest a basis for speed-accuracy trade-offs in the ocular motor system Comparedwith prosaccades antisaccades are associated with reducedpre-target activity of saccade-related neurons in the frontaleye field (Everling et al 1999 Everling and Munoz 2000)and superior colliculus (Everling et al 1999) key regions insaccade generation Moreover lower pre-target frontal eye

field activity is correlated with increased saccadic latencyand fewer antisaccade errors Reduced preparatory neuralactivity in response to the instruction to perform anantisaccade presumably makes it more difficult for activityrelated to the incorrect but prepotent prosaccade to reachthe threshold for triggering an eye movement However theprice for this improved accuracy is that the activitygenerating the correct antisaccade also takes longer toreach threshold resulting in prolonged saccadic latenciesThus we hypothesize that the slowing of responses thatfollow an antisaccade error (ie post-error slowingNieuwenhuis et al 2001) stems from an ACC errorresponse that signals the frontal eye field to reduce itspreparatory activity during the subsequent trial Thishypothesis is consistent with previous work demonstratingthat other aspects of trial history affect preparatory neuralactivity in the frontal eye field during antisaccades(Manoach et al 2007b) How might error signals beconveyed to frontal eye field The use of outcomes todynamically adjust responses has been attributed to ACCacting with lateral prefrontal cortex (Gehring and Knight2000 Paus 2001 Garavan et al 2002) and lateralprefrontal cortex has been proposed to provide inhibitoryinput to frontal eye field during antisaccades (DeSouzaet al 2003) Thus in ASD aberrant communicationbetween response evaluation in the ACC and regionsinitiating compensatory adjustments possibly lateral pre-frontal cortex could result in a failure to optimally reducepreparatory firing rates in frontal eye field in response toa prior error This reduced dynamic performance adjust-ment could lead to faster responding and more errorsWe are presently testing this hypothesis using an anti-saccade paradigm that gives rise to post-error slowing(Rabbitt 1966) and will allow a more direct examinationof compensatory behaviour and its supporting neuralcircuitry Unfortunately although post-error slowing isseen when antisaccades are the only task (Polli et al 2006)the intermixed antisaccade and prosaccade paradigmemployed here gives rise to other inter-trial factors thatalso affect latency (eg task-switching and prior antisac-cades Cherkasova et al 2002 Fecteau et al 2004 Bartonet al 2006 Manoach et al 2007b) and may obscure theeffects of a prior error There is evidence of reduced post-error slowing in adults with high functioning autismhowever from a previous study that employed a memorysearch task (Bogte et al 2007) Generalizing beyond theocular motor system failures of dynamic performanceadjustment due to ACC dysfunction and dysconnectivitymay contribute to the rigid and perseverative behavioursthat characterize ASD

Repetitive behaviour scores in ASD were also associatedwith reduced FA in the WM underlying left subgenualrACC Subgenual rACC is part of a network that is thoughtto play an important role in generating autonomicresponses to motivationally salient stimuli evaluatingtheir significance in predicting reward and in using this

Abnormal response monitoring in autism Brain (2008) Page 11 of 15

information to guide behaviour (Damasio et al 1990Drevets 2000) An important consideration in interpretingthis relation of FA to repetition is that subgenual rACCregion was not functionally or structurally abnormal in thisASD sample This suggests that even normal variation in FAof the subgenual rACC may contribute to a tendency torepeat unreinforced stimulus-response mappings Moregenerally the relations we observed between behaviouralrepetition and ACC function and structure may not bespecific to ASD Accumulating evidence suggests that thethree symptom clusters that define autismmdashsocial impair-ment communication deficits and restricted repetitivebehaviourmdasharise from distinct genetic and cognitivemechanisms and may represent continuous traits in thepopulation (for reviews see Happe et al 2006 London2007) Previous studies have reported relations betweenindices of response monitoring and repetition in OCD(Gehring et al 2000 Ursu et al 2003) and in non-clinically ascertained samples of children and adults withobsessivendashcompulsive characteristics (Hajcak and Simons2002 Santesso et al 2006)Our findings also raise the question of whether the

response monitoring abnormalities that we observed arespecific to ASD Similar abnormalities have been reportedin OCD (Maltby et al 2005) and attention deficithyperactivity disorder (van Meel et al 2007) Whilemethodological differences preclude a direct comparisonof these findings we previously studied patients withschizophrenia using the identical paradigm (Polli et al2008) Schizophrenia is also a neurodevelopmental disorderthat is characterized by rigid and perseverative behaviourresponse monitoring abnormalities (Carter et al 2001)reduced FA in ACC WM (Manoach et al 2007a) and aconsistently elevated antisaccade error rate (Radant et al2007 Manoach et al 2002) Although both schizophreniaand ASD samples made more antisaccade errors thancontrols ASD participants performed correct trials fasterand schizophrenia patients performed more slowly Whileboth groups showed reduced discrimination between errorand correct trials in the ACC this occurred for differentreasons In ASD this reflected a hyperactive response tocorrect trials while in schizophrenia it reflected a bluntedACC response to errors consistent with numerous previousreports (Laurens et al 2003) These differences argue fordistinct behavioural and neural signatures of responsemonitoring deficits in schizophrenia and ASD Thesefindings along with those of recent neuroimaging studiessuggesting genetic mediation of response monitoring(Fallgatter et al 2004 Klein et al 2007 Kramer et al2007) support the candidacy of neuroimaging-basedindices of response monitoring as neurocognitive endophe-notypes Whether response monitoring abnormalities canbe distinguished in ASD OCD and other neuropsychiatricdisorders whether they are related to diagnosis or specifictraits and whether they are also present in unaffectedfamily members remains to be seen

Several additional limitations to and alternative inter-pretations of our findings merit consideration First mightthe hyperactive rACC response to correct antisaccades inASD reflect a less efficient response to conflict ratherthan response monitoring The ACC is thought to play arole in detecting conflict and engaging cognitive controlto resolve it (for review see Carter and van Veen 2007)Antisaccades involve a high degree of conflict as one mustchoose between two competing responses and inhibit theprepotent one Errors give rise to another type of conflictwhen the actual response is compared with the intendedone Thus both correct and erroneous antisaccadesengender conflict While it is not possible to entirely ruleout the conflict explanation several observations argueagainst it If this rACC region were detecting conflict itshould respond to both correct and error trials Insteadcontrols showed a significant response only to errors Inaddition the timing of the haemodynamic response tocorrect antisaccades in ASD which started amp4 s and peakedat amp8 s (Fig 3) argues for a role in response monitoringrather than successful conflict resolution In this paradigmdifferential fMRI activation for correct antisaccade versusprosaccade trials including in dACC presumably reflectingconflict detection response inhibition and saccade genera-tion peaks at amp4 s and is mostly gone by 8 s (Manoachet al 2007b see also Polli et al 2005 for a discussionof ACC regions involved in saccade generation versusresponse monitoring) Finally it is dorsal rather than rACCactivation that is associated with engaging cognitive controlon correct high-conflict and error trials (Carter and vanVeen 2007)

An important limitation is that the small sample size ofthe present study may have compromised our power todetect additional group differences and relations of interestparticularly in regions outside our a priori ACC regions ofinterest for which more stringent multiple comparisonscorrections were applied In spite of the relatively smallsample size our a priori hypotheses were confirmedIn addition four of our ASD participants were takingmedications that could affect ACC function A re-analysisof the fMRI data with only unmedicated ASD participantshowever supported our main findings of significantly lessdiscrimination between response outcomes in the rACCand greater activation to correct trials It is also important tonote that our sample was limited to high functioning adultswith ASD so it is not clear that our findings would generalizeto lower functioning or younger samples Although increasedantisaccade error rates are seen in autism as early as ages 8ndash12(Luna et al 2007) we limited our study to adults sincesaccadic inhibition may not fully develop until late adoles-cence (Klein and Foerster 2001) and larger samples would benecessary to discriminate between the effects of diagnosis andthose due to normal development Another limitationconcerns our interpretation of FA as reflecting connectivityFA reflects the degree of directional coherence of waterdiffusion in tissue and is therefore an indirect measure of

Page 12 of 15 Brain (2008) K NThakkar et al

WMmicrostructure While FA correlates with axon myelinat-ion (Harsan et al 2006) not all of its biophysicaldeterminants are fully understood (Beaulieu 2002) Thus itis not possible to attribute FA differences or the relations weobserved to a particular WM propertyIn summary the present study reports hyperactive

response monitoring in the rACC in ASD that is relatedto restricted repetitive behaviour and is accompanied byreduced FA in the underlying white matter This suggests astructural correlate to abnormal function that may com-promise connectivity in the circuitry subserving responsemonitoring These findings suggest that structural and fun-ctional abnormalities of the ACC compromise responsemonitoring and contribute to behavioural repetition inASD Impairments in evaluating and learning from errorsmay lead to behaviour that is rigid and repetitive ratherthan optimally guided by outcomes and may compromiseperformance across a wide range of tasks Illuminating theneural basis and clinical significance of response monitoringdeficits in ASD represents a fruitful avenue for furtherresearch

Supplementary materialSupplementary material is available at Brain online

AcknowledgementsThe authors wish to thank Mark Vangel for statisticalconsultation Robert Levy Kim Ono Christopher Shermanand Jeremy Young for help with manuscript preparationand Matt Cain and Jay Edelman for technical assistanceSome of these data were presented at the annual meetingof the Society for Neuroscience in Washington DC inNovember 2005 This study was supported by NationalInstitute for Mental Health [R01 MH67720 (DSM)MH72120 (FEP)] Mental Illness Neuroscience Discovery(MIND) Institute (DOE DE-FG02-99ER62764) TheNational Center for Research Resources (P41RR14075)Funding to pay the Open Access publication charges forthis article was provided by the National Institute forMental Health NRSA MH72120

ReferencesAshwin C Baron-Cohen S Wheelwright S OrsquoRiordan M Bullmore ETDifferential activation of the amygdala and the lsquosocial brainrsquo duringfearful face-processing in Asperger syndrome Neuropsychologia 200745 2ndash14

Barnea-Goraly N Kwon H Menon V Eliez S Lotspeich L Reiss ALWhite matter structure in autism preliminary evidence from diffusiontensor imaging Biol Psychiatry 2004 55 323ndash6

Baron-Cohen S Do autistic children have obsessions and compulsionsBr J Clin Psychol 1989 28 (Pt 3) 193ndash200

Barton JJ Greenzang C Hefter R Edelman J Manoach DS Switchingplasticity and prediction in a saccadic task-switch paradigm Exp BrainRes 2006 168 76ndash87

Basser PJ Mattiello J LeBihan D MR diffusion tensor spectroscopy andimaging Biophys J 1994 66 259ndash67

Beaulieu C The basis of anisotropic water diffusion in the nervous system- a technical review NMR Biomed 2002 15 435ndash55

Blair JR Spreen O Predicting premorbid IQ a revision of the NationalAdult Reading Test Clin Neuropsychol 1989 3 129ndash136

Bogte H Flamma B van der Meere J van Engeland H Post-erroradaptation in adults with high functioning autism Neuropsychologia2007 45 1707ndash14

Breiter HC Rauch SL Kwong KK Baker JR Weisskoff RM Kennedy DNet al Functional magnetic resonance imaging of symptom provocationin obsessive-compulsive disorder Arch Gen Psychiatry 1996 53595ndash606

Buckner RL Goodman J Burock M Rotte M Koutstaal W Schacter Det al Functional-anatomic correlates of object priming in humansrevealed by rapid presentation event-related fMRI Neuron 1998 20285ndash96

Burock MA Dale AM Estimation and detection of event-related fMRIsignals with temporally correlated noise a statistically efficient andunbiased approach Hum Brain Mapp 2000 11 249ndash60

Bush G Whalen PJ Rosen BR Jenike MA McInerney SC Rauch SL Thecounting stroop an interference task specialized for functionalneuroimaging- -validation study with functional MRI Hum BrainMapp 1998 6 270ndash82

Carter CS MacDonald AW 3rd Ross LL Stenger VA Anterior cingulatecortex activity and impaired self-monitoring of performance in patientswith schizophrenia an event-related fMRI study Am J Psychiatry 2001158 1423ndash8

Carter CS van Veen V Anterior cingulate cortex and conflict detection anupdate of theory and data Cogn Affect Behav Neurosci 2007 7 367ndash79

Cherkasova MV Manoach DS Intriligator JM Barton JJS Antisaccadesand task-switching interactions in controlled processing Exp Brain Res2002 144 528ndash37

Collins DL Neelin P Peters TM Evans AC Automatic 3D intersubjectregistration of MR volumetric data in standardized Talairach spaceJ Comput Assist Tomogr 1994 18 192ndash205

Courchesne E Pierce K Why the frontal cortex in autism might be talkingonly to itself local over-connectivity but long-distance disconnectionCurr Opin Neurobiol 2005 15 225ndash30

Cox RW Jesmanowicz A Real-time 3D image registration for functionalMRI Magn Reson Med 1999 42 1014ndash8

Dale AM Optimal experimental design for event-related fMRI Hum BrainMapp 1999 8 109ndash40

Dale AM Fischl B Sereno MI Cortical surface-based analysisI Segmentation and surface reconstruction Neuroimage 1999 9 179ndash94

Damasio AR Tranel D Damasio H Individuals with sociopathic behaviorcaused by frontal damage fail to respond autonomically to social stimuliBehav Brain Res 1990 41 81ndash94

Dehaene S Posner MI Tucker DM Localization of a neural system forerror detection and compensation Psychol Sci 1994 5 303ndash5

DeSouza JF Menon RS Everling S Preparatory set associated with pro-saccades and anti-saccades in humans investigated with event-relatedFMRI J Neurophysiol 2003 89 1016ndash23

Devinsky O Morrell MJ Vogt BA Contributions of anterior cingulatecortex to behaviour Brain 1995 118 (Pt 1) 279ndash306

Dichter GS Belger A Social stimuli interfere with cognitive control inautism Neuroimage 2007 35 1219ndash30

Dougherty DD Baer L Cosgrove GR Cassem EH Price BHNierenberg AA et al Prospective long-term follow-up of 44 patientswho received cingulotomy for treatment-refractory obsessive-compulsivedisorder Am J Psychiatry 2002 159 269ndash75

Drevets WC Neuroimaging studies of mood disorders Biol Psychiatry2000 48 813ndash29

Everling S Dorris MC Klein RM Munoz DP Role of primate superiorcolliculus in preparation and execution of anti-saccades and pro-saccades J Neurosci 1999 19 2740ndash54

Everling S Munoz DP Neuronal correlates for preparatory set associatedwith pro-saccades and anti-saccades in the primate frontal eye fieldJ Neurosci 2000 20 387ndash400

Abnormal response monitoring in autism Brain (2008) Page 13 of 15

Falkenstein M Hohnsbein J Hoormann J Event-related correlates of errorsin reaction tasks In Karmos G Molnar M Csepe I Desmedt JE editorsPerspectives of event-related potentials research AmsterdamThe Netherlands Elsevier Science 1995 p 287ndash96

Fallgatter AJ Herrmann MJ Roemmler J Ehlis AC Wagener AHeidrich A et al Allelic variation of serotonin transporter functionmodulates the brain electrical response for error processingNeuropsychopharmacology 2004 29 1506ndash11

Fecteau JH Au C Armstrong IT Munoz DP Sensory biases producealternation advantage found in sequential saccadic eye movement tasksExp Brain Res 2004 159 84ndash91

Fischer B Breitmeyer B Mechanisms of visual attention revealed bysaccadic eye movements Neuropsychologia 1987 25 73ndash83

Fischl B Sereno MI Dale AM Cortical surface-based analysis II Inflationflattening and a surface-based coordinate system Neuroimage 1999a 9195ndash207

Fischl B Sereno MI Tootell RBH Dale AM High-resolution intersubjectaveraging and a coordinate system for the cortical surface Hum BrainMapp 1999b 8 272ndash84

Fischl B van der Kouwe A Destrieux C Halgren E Segonne F Salat DHet al Automatically parcellating the human cerebral cortex CerebCortex 2004 14 11ndash22

Fitzgerald KD Welsh RC Gehring WJ Abelson JL Himle JA Liberzon Iet al Error-related hyperactivity of the anterior cingulate cortex inobsessive-compulsive disorder Biol Psychiatry 2005 57 287ndash94

Garavan H Ross TJ Murphy K Roche RA Stein EA Dissociable executivefunctions in the dynamic control of behavior inhibition errordetection and correction NeuroImage 2002 17 1820ndash9

Gehring WJ Goss B Coles MG Meyer DE Donchin E A neuralsystem for error detection and compensation Psychol Sci 1993 4385ndash90

Gehring WJ Himle J Nisenson LG Action-monitoring dysfunction inobsessive-compulsive disorder Psychol Sci 2000 11 1ndash6

Gehring WJ Knight RT Prefrontal-cingulate interactions in actionmonitoring Nat Neurosci 2000 3 516ndash20

Goldberg MC Lasker AG Zee DS Garth E Tien A Landa RJ Deficits inthe initiation of eye movements in the absence of a visual target inadolescents with high functioning autism Neuropsychologia 2002 402039ndash49

Gomot M Bernard FA Davis MH Belmonte MK Ashwin CBullmore ET et al Change detection in children with autism anauditory event-related fMRI study Neuroimage 2006 29 475ndash84

Hajcak G Simons RF Error-related brain activity in obsessive-compulsiveundergraduates Psychiatry Res 2002 110 63ndash72

Hall GB Szechtman H Nahmias C Enhanced salience and emotionrecognition in autism a PET study Am J Psychiatry 2003 160 1439ndash41

Hallett PE Primary and secondary saccades to goals defined byinstructions Vision Res 1978 18 1279ndash96

Happe F Ronald A Plomin R Time to give up on a single explanation forautism Nat Neurosci 2006 9 1218ndash20

Harsan LA Poulet P Guignard B Steibel J Parizel N de Sousa PL et alBrain dysmyelination and recovery assessment by noninvasive in vivodiffusion tensor magnetic resonance imaging J Neurosci Res 2006 83392ndash402

Haznedar MM Buchsbaum MS Metzger M Solimando A Spiegel-Cohen J Hollander E Anterior cingulate gyrus volume and glucosemetabolism in autistic disorder Am J Psychiatry 1997 154 1047ndash50

Haznedar MM Buchsbaum MS Wei TC Hof PR Cartwright CBienstock CA et al Limbic circuitry in patients with autism spectrumdisorders studied with positron emission tomography and magneticresonance imaging Am J Psychiatry 2000 157 1994ndash2001

Henderson H Schwartz C Mundy P Burnette C Sutton S Zahka N et alResponse monitoring the error-related negativity and differences insocial behavior in autism Brain Cogn 2006 61 96ndash109

Hill EL Executive dysfunction in autism Trends Cogn Sci 2004 8 26ndash32

Hollander E King A Delaney K Smith CJ Silverman JM Obsessive-compulsive behaviors in parents of multiplex autism families PsychiatryRes 2003 117 11ndash6

Hollingshead AB Two factor index of social position New Haven CTYale University Press 1965

Holroyd CB Coles MG The neural basis of human error processingreinforcement learning dopamine and the error-related negativityPsychol Rev 2002 109 679ndash709

Jenkinson M Smith S A global optimisation method for robust affineregistration of brain images Med Image Anal 2001 5 143ndash56

Johannes S Wieringa BM Nager W Rada D Dengler R Emrich HMet al Discrepant target detection and action monitoring in obsessive-compulsive disorder Psychiatry Res 2001 108 101ndash10

Jones DK The effect of gradient sampling schemes on measures derivedfrom diffusion tensor MRI A Monte Carlo study Magn Reson Med2004 51 807ndash15

Kana RK Keller TA Minshew NJ Just MA Inhibitory control in high-functioning autism decreased activation and underconnectivity ininhibition networks Biol Psychiatry 2007 62 198ndash206

Kennedy DP Redcay E Courchesne E Failing to deactivate restingfunctional abnormalities in autism Proc Natl Acad Sci USA 2006 1038275ndash80

Klein C Foerster F Development of prosaccade and antisaccade taskperformance in participants aged 6 to 26 years Psychophysiology 200138 179ndash89

Klein TA Neumann J Reuter M Hennig J von Cramon DYUllsperger M Genetically determined differences in learning fromerrors Science 2007 318 1642ndash5

Kramer UM Cunillera T Camara E Marco-Pallares J Cucurell DNager W et al The impact of catechol-O-methyltransferase anddopamine D4 receptor genotypes on neurophysiological markers ofperformance monitoring J Neurosci 2007 27 14190ndash8

Laurens KR Ngan ET Bates AT Kiehl KA Liddle PF Rostral anteriorcingulate cortex dysfunction during error processing in schizophreniaBrain 2003 126 610ndash22

Levitt JG OrsquoNeill J Blanton RE Smalley S Fadale D McCracken JT et alProton magnetic resonance spectroscopic imaging of the brain inchildhood autism Biol Psychiatry 2003 54 1355ndash66

Leyfer OT Folstein SE Bacalman S Davis NO Dinh E Morgan J et alComorbid psychiatric disorders in children with autism interview devel-opment and rates of disorders J Autism Dev Disord 2006 36 849ndash61

London E The role of the neurobiologist in redefining the diagnosis ofautism Brain Pathol 2007 17 408ndash11

Lopez BR Lincoln AJ Ozonoff S Lai Z Examining the relationshipbetween executive functions and restricted repetitive symptoms ofAutistic Disorder J Autism Dev Disord 2005 35 445ndash60

Lord C Rutter M DiLavore PC Risi S Autism diagnostic observationschedule - WPS (ADOS-WPS) Los Angeles CA Western PsychologicalServices 1999

Luna B Doll SK Hegedus SJ Minshew NJ Sweeney JA Maturation ofexecutive function in autism Biol Psychiatry 2007 61 474ndash81

Luu P Tucker DM Derryberry D Reed M Poulsen C Electrophysiolog-ical responses to errors and feedback in the process of action regulationPsychol Sci 2003 14 47ndash53

Maltby N Tolin DF Worhunsky P OrsquoKeefe TM Kiehl KA Dysfunctionalaction monitoring hyperactivates frontal-striatal circuits in obsessive-compulsive disorder an event-related fMRI study Neuroimage 2005 24495ndash503

Manoach DS Ketwaroo GA Polli FE Thakkar KN Barton JJ Goff DCet al Reduced microstructural integrity of the white matter underlyinganterior cingulate cortex is associated with increased saccadic latency inschizophrenia Neuroimage 2007a 37 599ndash610

Manoach DS Lindgren KA Barton JJ Deficient saccadic inhibition inAspergerrsquos disorder and the social-emotional processing disorderJ Neurol Neurosurg Psychiatry 2004 75 1719ndash26

Page 14 of 15 Brain (2008) K NThakkar et al

Manoach DS Lindgren KA Cherkasova MV Goff DC Halpern EFIntriligator J et al Schizophrenic subjects show deficient inhibition butintact task-switching on saccadic tasks Biol Psychiatry 2002 51 816ndash26

Manoach DS Thakkar KN Cain MS Polli FE Edelman JA Fischl B et alNeural activity is modulated by trial history a functional magneticresonance imaging study of the effects of a previous antisaccadeJ Neurosci 2007b 27 1791ndash8

Manoach DS Weintraub S Daffner KR Scinto LF Deficient antisaccadesin the social-emotional processing disorder Neuroreport 1997 8 901ndash5

McDougle CJ Kresch LE Goodman WK Naylor ST Volkmar FRCohen DJ et al A case-controlled study of repetitive thoughts andbehavior in adults with autistic disorder and obsessive-compulsivedisorder Am J Psychiatry 1995 152 772ndash7

McPeek RM Skavenski AA Nakayama K Concurrent processing ofsaccades in visual search Vision Res 2000 40 2499ndash516

Miezin FM Maccotta L Ollinger JM Petersen SE Buckner RLCharacterizing the hemodynamic response effects of presentation ratesampling procedure and the possibility of ordering brain activity basedon relative timing Neuroimage 2000 11 735ndash59

Minshew NJ Luna B Sweeney JA Oculomotor evidence for neocorticalsystems but not cerebellar dysfunction in autism Neurology 1999 52917ndash22

Mokler A Fischer B The recognition and correction of involuntaryprosaccades in an antisaccade task Exp Brain Res 1999 125 511ndash6

Munoz DP Everling S Look away the anti-saccade task and the voluntarycontrol of eye movement Nat Rev Neurosci 2004 5 218ndash28

Murphy DG Daly E Schmitz N Toal F Murphy K Curran S et alCortical serotonin 5-HT2A receptor binding and social communicationin adults with Aspergerrsquos syndrome an in vivo SPECT study Am JPsychiatry 2006 163 934ndash6

Nieuwenhuis S Nielen MM Mol N Hajcak G Veltman DJ Performancemonitoring in obsessive-compulsive disorder Psychiatry Res 2005 134111ndash22

Nieuwenhuis S Ridderinkhof KR Blom J Band GP Kok A Error-relatedbrain potentials are differentially related to awareness of response errorsevidence from an antisaccade task Psychophysiology 2001 38 752ndash60

Ohnishi T Matsuda H Hashimoto T Kunihiro T Nishikawa M Uema Tet al Abnormal regional cerebral blood flow in childhood autism Brain2000 123 (Pt 9) 1838ndash44

Oldfield RC The assessment and analysis of handedness The EdinburghInventory Neuropsychologia 1971 9 97ndash113

Paus T Primate anterior cingulate cortex where motor control drive andcognition interface Nat Rev Neurosci 2001 2 417ndash24

Pfefferbaum A Sullivan EV Hedehus M Lim KO Adalsteinsson EMoseley M Age-related decline in brain white matter anisotropymeasured with spatially corrected echo-planar diffusion tensor imagingMagn Reson Med 2000 44 259ndash68

Pitman RK A cybernetic model of obsessive-compulsive psychopathologyCompr Psychiatry 1987 28 334ndash43

Polli FE Barton JJ Cain MS Thakkar KN Rauch SL Manoach DSRostral and dorsal anterior cingulate cortex make dissociable contribu-tions during antisaccade error commission Proc Natl Acad Sci USA2005 102 15700ndash5

Polli FE Barton JJ Thakkar KN Greve DN Goff DC Rauch SL et alReduced error-related activation in two anterior cingulate circuits is relatedto impaired performance in schizophrenia Brain 2008 131 971ndash86

Polli FE Barton JJ Vangel M Goff DC Iguchi L Manoach DSSchizophrenia patients show intact immediate error-related performanceadjustments on an antisaccade task Schizophr Res 2006 82 191ndash201

Rabbitt PM Errors and error correction in choice-response tasks J ExpPsychol 1966 71 264ndash72

Radant AD Dobie DJ Calkins ME Olincy A Braff DL Cadenhead KSet al Successful multi-site measurement of antisaccade performancedeficits in schizophrenia Schizophr Res 2007 89 320ndash9

Reese TG Heid O Weisskoff RM Wedeen VJ Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echoMagn Reson Med 2003 49 177ndash82

Russell J Jarrold C Error-correction problems in autism evidence for amonitoring impairment J Autism Dev Disord 1998 28 177ndash88

Russell AJ Mataix-Cols D Anson M Murphy DG Obsessions andcompulsions in Asperger syndrome and high-functioning autism Br JPsychiatry 2005 186 525ndash8

Rutter M Le Couteur A Lord C Autism diagnostic interview-revised LosAngeles CA Western Psychological Services 2003

Santesso DL Segalowitz SJ Schmidt LA Error-related electrocorticalresponses are enhanced in children with obsessive-compulsive behaviorsDev Neuropsychol 2006 29 431ndash45

Shafritz KM Dichter GS Baranek GT Belger A The neural circuitrymediating shifts in behavioral response and cognitive set in autism BiolPsychiatry 2007 63 974ndash80

Silk TJ Rinehart N Bradshaw JL Tonge B Egan G OrsquoBoyle MW et alVisuospatial processing and the function of prefrontal-parietal networksin autism spectrum disorders a functional MRI study Am J Psychiatry2006 163 1440ndash3

South M Ozonoff S McMahon WM The relationship between executivefunctioning central coherence and repetitive behaviors in the high-functioning autism spectrum Autism 2007 11 437ndash51

Taylor SF Martis B Fitzgerald KD Welsh RC Abelson JL Liberzon Iet al Medial frontal cortex activity and loss-related responses to errorsJ Neurosci 2006 26 4063ndash70

Taylor SF Stern ER Gehring WJ Neural systems for error monitoringrecent findings and theoretical perspectives Neuroscientist 2007 13160ndash72

Thesen S Heid O Mueller E Schad LR Prospective acquisition correctionfor head motion with image-based tracking for real-time fMRI MagnReson Med 2000 44 457ndash65

Turner M Annotation Repetitive behaviour in autism areview of psychological research J Child Psychol Psychiatry 1999 40839ndash49

Ullsperger M von Cramon DY Neuroimaging of performance monitoringerror detection and beyond Cortex 2004 40 593ndash604

Ursu S Stenger VA Shear MK Jones MR Carter CS Overactiveaction monitoring in obsessive-compulsive disorder evidencefrom functional magnetic resonance imaging Psychol Sci 2003 14347ndash53

Meel CS Heslenfeld DJ Oosterlaan J Sergeant JA Adaptive controldeficits in attention-deficithyperactivity disorder (ADHD) the role oferror processing Psychiatry Res 2007 151 211ndash20

van Veen V Carter CS The timing of action-monitoring processes in theanterior cingulate cortex J Cogn Neurosci 2002 14 593ndash602

Wechsler D Wechsler Abbreviated Scale of intelligence San Antonio TXThe Psychological Corporation 1999

Whalen PJ Bush G McNally RJ Wilhelm S McInerney SC Jenike MAet al The emotional counting Stroop paradigm a functional magneticresonance imaging probe of the anterior cingulate affective division BiolPsychiatry 1998 44 1219ndash28

White K Ashton R Handedness assessment inventory Neuropsychologia1976 14 261ndash4

Zandt F Prior M Kyrios M Repetitive behaviour in children with highfunctioning autism and obsessive compulsive disorder J Autism DevDisord 2007 37 251ndash9

Abnormal response monitoring in autism Brain (2008) Page 15 of 15

rather than a blunted response to errors Reduceddiscrimination between response outcomes may interferewith adjusting responses to obtain the most favourableoutcome both with regard to subsequent trials and longerterm modification of the strength of stimulus-responsemappings This could contribute to behaviour that isstimulus-bound and repetitive rather than flexible andresponsive to contingencies This provides a plausibleexplanation for the relation between increased rACCactivity to correct trials and repetitive behaviourA complementary explanation has been offered for the

relation of increased error signals to the severity ofobsessions and compulsions in OCD (Gehring et al2000 Ursu et al 2003) A longstanding theory of OCDposits that inappropriate and exaggerated error signals leadto a pervasive sense of incompleteness and self-doubt andbehavioural repetition represents a vain attempt to reducethese error signals (Pitman 1987) According to this theoryexaggerated ACC activation to correct trials reflectsinappropriate error signalling and triggers a compulsionto repeat behaviours that were already successfully com-pleted (Maltby et al 2005) Similar to individuals withOCD ASD participants showed an exaggerated ACCresponse to correct trials (Ursu et al 2003 Maltby et al2005) While in OCD fMRI activation to correct trials wasmaximal in dACC (Maltby et al 2005) in ASD it was

maximal in rACC and was related to repetitive behaviourWhile dACC activity following errors is thought to reflecteither conflict or error detection (for review see Ullspergerand von Cramon 2004) rACC activity following errors hasbeen interpreted as appraisal of the affective or motivationalsalience of errors (van Veen and Carter 2002 Luu et al2003 Taylor et al 2006) Thus one interpretation of thisrelation is that it reflects an inappropriate affective responseto correct performance that triggers behavioural repetitionThis exaggerated affective response may occur even withawareness that one has performed correctly Regardless of itsbasis the relation between increased rACC responses tocorrect trials and repetitive behaviour suggests that aberrantresponse monitoring contributes to a defining feature of ASD

Both ASD and OCD are characterized by behaviour thatis rigid and not optimally responsive to contingency andthere is evidence of both phenomenological overlap anddistinctions between the obsessions and compulsions ofOCD and the restricted and repetitive behaviours of ASDboth in terms of subjective experience and type ofbehaviour (Baron-Cohen 1989 McDougle et al 1995Russell et al 2005 Zandt et al 2007) Interestingly a studyof multiplex families of children with autism found thatparents were significantly more likely to have obsessivendashcompulsive traits or disorder if the children had high ADI-Rrepetitive behaviour scores and this was due to loading

Fig 5 Relations of ACC function and structure to restricted repetitive behaviour in ASD (A) Statistical map of regression of rawBOLD signal on ADI-R repetitive behaviour score BOLD signal is from the correct antisaccade versus fixation contrast at 8 s Results aredisplayed on the inflated right medial cortical surface Scatter plot shows BOLD signal from the maximum vertex in right rACC foreach ASD participant on the y-axis and ADI-R score on the x-axis (B) Statistical map of regression of FA on ADI-R repetitive behaviourscore Results are displayed on the inflated left medial surface Scatter plot shows FA from the maximum vertex in left rACC for eachASD participant on the y-axis and ADI-R score on the x-axis The rACC is outlined in red and the dACC is outlined in blue

Page 10 of 15 Brain (2008) K NThakkar et al

on ideational but not sensorimotor items (Hollander et al2003) This finding supports the distinction betweenrepetition of movements and repetition of more complexideational behaviours (for review see Turner 1999) andsuggests that the latter may share underlying mechanisms inOCD and ASD In the present study both sensorimotorand ideational subscores contributed to the relations seenwith response monitoring activation and FA At presentrepetitive behaviours in autism are incompletely under-stood and neurobiologically valid dimensions are yet to bedelineated The present findings suggest that ACC abnorm-alities may contribute to repetitive behaviour in ASDpossibly on the basis of their contribution to aberrantresponse monitoring but linking ACC abnormalities tospecific dimensions of repetitive behaviour and identifyingthe underlying mechanisms requires further study withlarge samplesAbnormal response monitoring may also contribute to

deficient cognitive control of behaviour in ASD which inthe present study was reflected in an increased antisaccadeerror rate consistent with prior work (Manoach et al1997 2004 Minshew et al 1999 Goldberg et al 2002Luna et al 2007) and possibly in reduced dynamicperformance adjustment In spite of making more errorsASD participants responded more quickly than controls oncorrect trials suggesting a disruption in the speed-accuracytrade-off This apparent difficulty in modulating responsespeed to improve accuracy is unlikely to reflect reducederror awareness since ASD participants were as or morelikely than controls to self-correct errors Nor is it likely toreflect a blunted neural response to errors given both theincreased response to errors in left perigenual rACC andprevious work showing normal or increased amplitude ofthe ERN in ASD (Henderson et al 2006) Deficient use ofoutcomes to adjust performance in a continuous dynamicfashion may stem from response monitoring dysfunction inthe ACC andor from dysconnection of the ACC fromocular motor regions involved in saccade generationAberrant connectivity of ACC regions subserving responsemonitoring in ASD is suggested by our finding of reducedFA in the underlying WM This finding is consistent with aprevious report of reduced FA in ACC white matter inautism (Barnea-Goraly et al 2004) and with evidence ofdecreased functional connectivity of the ACC to otherregions during a task involving response inhibition (Kanaet al 2007)How might information concerning the outcome of the

prior response be communicated to ocular motor regionsresponsible for adjusting behaviour in the subsequent trialMonkey electrophysiology studies suggest a basis for speed-accuracy trade-offs in the ocular motor system Comparedwith prosaccades antisaccades are associated with reducedpre-target activity of saccade-related neurons in the frontaleye field (Everling et al 1999 Everling and Munoz 2000)and superior colliculus (Everling et al 1999) key regions insaccade generation Moreover lower pre-target frontal eye

field activity is correlated with increased saccadic latencyand fewer antisaccade errors Reduced preparatory neuralactivity in response to the instruction to perform anantisaccade presumably makes it more difficult for activityrelated to the incorrect but prepotent prosaccade to reachthe threshold for triggering an eye movement However theprice for this improved accuracy is that the activitygenerating the correct antisaccade also takes longer toreach threshold resulting in prolonged saccadic latenciesThus we hypothesize that the slowing of responses thatfollow an antisaccade error (ie post-error slowingNieuwenhuis et al 2001) stems from an ACC errorresponse that signals the frontal eye field to reduce itspreparatory activity during the subsequent trial Thishypothesis is consistent with previous work demonstratingthat other aspects of trial history affect preparatory neuralactivity in the frontal eye field during antisaccades(Manoach et al 2007b) How might error signals beconveyed to frontal eye field The use of outcomes todynamically adjust responses has been attributed to ACCacting with lateral prefrontal cortex (Gehring and Knight2000 Paus 2001 Garavan et al 2002) and lateralprefrontal cortex has been proposed to provide inhibitoryinput to frontal eye field during antisaccades (DeSouzaet al 2003) Thus in ASD aberrant communicationbetween response evaluation in the ACC and regionsinitiating compensatory adjustments possibly lateral pre-frontal cortex could result in a failure to optimally reducepreparatory firing rates in frontal eye field in response toa prior error This reduced dynamic performance adjust-ment could lead to faster responding and more errorsWe are presently testing this hypothesis using an anti-saccade paradigm that gives rise to post-error slowing(Rabbitt 1966) and will allow a more direct examinationof compensatory behaviour and its supporting neuralcircuitry Unfortunately although post-error slowing isseen when antisaccades are the only task (Polli et al 2006)the intermixed antisaccade and prosaccade paradigmemployed here gives rise to other inter-trial factors thatalso affect latency (eg task-switching and prior antisac-cades Cherkasova et al 2002 Fecteau et al 2004 Bartonet al 2006 Manoach et al 2007b) and may obscure theeffects of a prior error There is evidence of reduced post-error slowing in adults with high functioning autismhowever from a previous study that employed a memorysearch task (Bogte et al 2007) Generalizing beyond theocular motor system failures of dynamic performanceadjustment due to ACC dysfunction and dysconnectivitymay contribute to the rigid and perseverative behavioursthat characterize ASD

Repetitive behaviour scores in ASD were also associatedwith reduced FA in the WM underlying left subgenualrACC Subgenual rACC is part of a network that is thoughtto play an important role in generating autonomicresponses to motivationally salient stimuli evaluatingtheir significance in predicting reward and in using this

Abnormal response monitoring in autism Brain (2008) Page 11 of 15

information to guide behaviour (Damasio et al 1990Drevets 2000) An important consideration in interpretingthis relation of FA to repetition is that subgenual rACCregion was not functionally or structurally abnormal in thisASD sample This suggests that even normal variation in FAof the subgenual rACC may contribute to a tendency torepeat unreinforced stimulus-response mappings Moregenerally the relations we observed between behaviouralrepetition and ACC function and structure may not bespecific to ASD Accumulating evidence suggests that thethree symptom clusters that define autismmdashsocial impair-ment communication deficits and restricted repetitivebehaviourmdasharise from distinct genetic and cognitivemechanisms and may represent continuous traits in thepopulation (for reviews see Happe et al 2006 London2007) Previous studies have reported relations betweenindices of response monitoring and repetition in OCD(Gehring et al 2000 Ursu et al 2003) and in non-clinically ascertained samples of children and adults withobsessivendashcompulsive characteristics (Hajcak and Simons2002 Santesso et al 2006)Our findings also raise the question of whether the

response monitoring abnormalities that we observed arespecific to ASD Similar abnormalities have been reportedin OCD (Maltby et al 2005) and attention deficithyperactivity disorder (van Meel et al 2007) Whilemethodological differences preclude a direct comparisonof these findings we previously studied patients withschizophrenia using the identical paradigm (Polli et al2008) Schizophrenia is also a neurodevelopmental disorderthat is characterized by rigid and perseverative behaviourresponse monitoring abnormalities (Carter et al 2001)reduced FA in ACC WM (Manoach et al 2007a) and aconsistently elevated antisaccade error rate (Radant et al2007 Manoach et al 2002) Although both schizophreniaand ASD samples made more antisaccade errors thancontrols ASD participants performed correct trials fasterand schizophrenia patients performed more slowly Whileboth groups showed reduced discrimination between errorand correct trials in the ACC this occurred for differentreasons In ASD this reflected a hyperactive response tocorrect trials while in schizophrenia it reflected a bluntedACC response to errors consistent with numerous previousreports (Laurens et al 2003) These differences argue fordistinct behavioural and neural signatures of responsemonitoring deficits in schizophrenia and ASD Thesefindings along with those of recent neuroimaging studiessuggesting genetic mediation of response monitoring(Fallgatter et al 2004 Klein et al 2007 Kramer et al2007) support the candidacy of neuroimaging-basedindices of response monitoring as neurocognitive endophe-notypes Whether response monitoring abnormalities canbe distinguished in ASD OCD and other neuropsychiatricdisorders whether they are related to diagnosis or specifictraits and whether they are also present in unaffectedfamily members remains to be seen

Several additional limitations to and alternative inter-pretations of our findings merit consideration First mightthe hyperactive rACC response to correct antisaccades inASD reflect a less efficient response to conflict ratherthan response monitoring The ACC is thought to play arole in detecting conflict and engaging cognitive controlto resolve it (for review see Carter and van Veen 2007)Antisaccades involve a high degree of conflict as one mustchoose between two competing responses and inhibit theprepotent one Errors give rise to another type of conflictwhen the actual response is compared with the intendedone Thus both correct and erroneous antisaccadesengender conflict While it is not possible to entirely ruleout the conflict explanation several observations argueagainst it If this rACC region were detecting conflict itshould respond to both correct and error trials Insteadcontrols showed a significant response only to errors Inaddition the timing of the haemodynamic response tocorrect antisaccades in ASD which started amp4 s and peakedat amp8 s (Fig 3) argues for a role in response monitoringrather than successful conflict resolution In this paradigmdifferential fMRI activation for correct antisaccade versusprosaccade trials including in dACC presumably reflectingconflict detection response inhibition and saccade genera-tion peaks at amp4 s and is mostly gone by 8 s (Manoachet al 2007b see also Polli et al 2005 for a discussionof ACC regions involved in saccade generation versusresponse monitoring) Finally it is dorsal rather than rACCactivation that is associated with engaging cognitive controlon correct high-conflict and error trials (Carter and vanVeen 2007)

An important limitation is that the small sample size ofthe present study may have compromised our power todetect additional group differences and relations of interestparticularly in regions outside our a priori ACC regions ofinterest for which more stringent multiple comparisonscorrections were applied In spite of the relatively smallsample size our a priori hypotheses were confirmedIn addition four of our ASD participants were takingmedications that could affect ACC function A re-analysisof the fMRI data with only unmedicated ASD participantshowever supported our main findings of significantly lessdiscrimination between response outcomes in the rACCand greater activation to correct trials It is also important tonote that our sample was limited to high functioning adultswith ASD so it is not clear that our findings would generalizeto lower functioning or younger samples Although increasedantisaccade error rates are seen in autism as early as ages 8ndash12(Luna et al 2007) we limited our study to adults sincesaccadic inhibition may not fully develop until late adoles-cence (Klein and Foerster 2001) and larger samples would benecessary to discriminate between the effects of diagnosis andthose due to normal development Another limitationconcerns our interpretation of FA as reflecting connectivityFA reflects the degree of directional coherence of waterdiffusion in tissue and is therefore an indirect measure of

Page 12 of 15 Brain (2008) K NThakkar et al

WMmicrostructure While FA correlates with axon myelinat-ion (Harsan et al 2006) not all of its biophysicaldeterminants are fully understood (Beaulieu 2002) Thus itis not possible to attribute FA differences or the relations weobserved to a particular WM propertyIn summary the present study reports hyperactive

response monitoring in the rACC in ASD that is relatedto restricted repetitive behaviour and is accompanied byreduced FA in the underlying white matter This suggests astructural correlate to abnormal function that may com-promise connectivity in the circuitry subserving responsemonitoring These findings suggest that structural and fun-ctional abnormalities of the ACC compromise responsemonitoring and contribute to behavioural repetition inASD Impairments in evaluating and learning from errorsmay lead to behaviour that is rigid and repetitive ratherthan optimally guided by outcomes and may compromiseperformance across a wide range of tasks Illuminating theneural basis and clinical significance of response monitoringdeficits in ASD represents a fruitful avenue for furtherresearch

Supplementary materialSupplementary material is available at Brain online

AcknowledgementsThe authors wish to thank Mark Vangel for statisticalconsultation Robert Levy Kim Ono Christopher Shermanand Jeremy Young for help with manuscript preparationand Matt Cain and Jay Edelman for technical assistanceSome of these data were presented at the annual meetingof the Society for Neuroscience in Washington DC inNovember 2005 This study was supported by NationalInstitute for Mental Health [R01 MH67720 (DSM)MH72120 (FEP)] Mental Illness Neuroscience Discovery(MIND) Institute (DOE DE-FG02-99ER62764) TheNational Center for Research Resources (P41RR14075)Funding to pay the Open Access publication charges forthis article was provided by the National Institute forMental Health NRSA MH72120

ReferencesAshwin C Baron-Cohen S Wheelwright S OrsquoRiordan M Bullmore ETDifferential activation of the amygdala and the lsquosocial brainrsquo duringfearful face-processing in Asperger syndrome Neuropsychologia 200745 2ndash14

Barnea-Goraly N Kwon H Menon V Eliez S Lotspeich L Reiss ALWhite matter structure in autism preliminary evidence from diffusiontensor imaging Biol Psychiatry 2004 55 323ndash6

Baron-Cohen S Do autistic children have obsessions and compulsionsBr J Clin Psychol 1989 28 (Pt 3) 193ndash200

Barton JJ Greenzang C Hefter R Edelman J Manoach DS Switchingplasticity and prediction in a saccadic task-switch paradigm Exp BrainRes 2006 168 76ndash87

Basser PJ Mattiello J LeBihan D MR diffusion tensor spectroscopy andimaging Biophys J 1994 66 259ndash67

Beaulieu C The basis of anisotropic water diffusion in the nervous system- a technical review NMR Biomed 2002 15 435ndash55

Blair JR Spreen O Predicting premorbid IQ a revision of the NationalAdult Reading Test Clin Neuropsychol 1989 3 129ndash136

Bogte H Flamma B van der Meere J van Engeland H Post-erroradaptation in adults with high functioning autism Neuropsychologia2007 45 1707ndash14

Breiter HC Rauch SL Kwong KK Baker JR Weisskoff RM Kennedy DNet al Functional magnetic resonance imaging of symptom provocationin obsessive-compulsive disorder Arch Gen Psychiatry 1996 53595ndash606

Buckner RL Goodman J Burock M Rotte M Koutstaal W Schacter Det al Functional-anatomic correlates of object priming in humansrevealed by rapid presentation event-related fMRI Neuron 1998 20285ndash96

Burock MA Dale AM Estimation and detection of event-related fMRIsignals with temporally correlated noise a statistically efficient andunbiased approach Hum Brain Mapp 2000 11 249ndash60

Bush G Whalen PJ Rosen BR Jenike MA McInerney SC Rauch SL Thecounting stroop an interference task specialized for functionalneuroimaging- -validation study with functional MRI Hum BrainMapp 1998 6 270ndash82

Carter CS MacDonald AW 3rd Ross LL Stenger VA Anterior cingulatecortex activity and impaired self-monitoring of performance in patientswith schizophrenia an event-related fMRI study Am J Psychiatry 2001158 1423ndash8

Carter CS van Veen V Anterior cingulate cortex and conflict detection anupdate of theory and data Cogn Affect Behav Neurosci 2007 7 367ndash79

Cherkasova MV Manoach DS Intriligator JM Barton JJS Antisaccadesand task-switching interactions in controlled processing Exp Brain Res2002 144 528ndash37

Collins DL Neelin P Peters TM Evans AC Automatic 3D intersubjectregistration of MR volumetric data in standardized Talairach spaceJ Comput Assist Tomogr 1994 18 192ndash205

Courchesne E Pierce K Why the frontal cortex in autism might be talkingonly to itself local over-connectivity but long-distance disconnectionCurr Opin Neurobiol 2005 15 225ndash30

Cox RW Jesmanowicz A Real-time 3D image registration for functionalMRI Magn Reson Med 1999 42 1014ndash8

Dale AM Optimal experimental design for event-related fMRI Hum BrainMapp 1999 8 109ndash40

Dale AM Fischl B Sereno MI Cortical surface-based analysisI Segmentation and surface reconstruction Neuroimage 1999 9 179ndash94

Damasio AR Tranel D Damasio H Individuals with sociopathic behaviorcaused by frontal damage fail to respond autonomically to social stimuliBehav Brain Res 1990 41 81ndash94

Dehaene S Posner MI Tucker DM Localization of a neural system forerror detection and compensation Psychol Sci 1994 5 303ndash5

DeSouza JF Menon RS Everling S Preparatory set associated with pro-saccades and anti-saccades in humans investigated with event-relatedFMRI J Neurophysiol 2003 89 1016ndash23

Devinsky O Morrell MJ Vogt BA Contributions of anterior cingulatecortex to behaviour Brain 1995 118 (Pt 1) 279ndash306

Dichter GS Belger A Social stimuli interfere with cognitive control inautism Neuroimage 2007 35 1219ndash30

Dougherty DD Baer L Cosgrove GR Cassem EH Price BHNierenberg AA et al Prospective long-term follow-up of 44 patientswho received cingulotomy for treatment-refractory obsessive-compulsivedisorder Am J Psychiatry 2002 159 269ndash75

Drevets WC Neuroimaging studies of mood disorders Biol Psychiatry2000 48 813ndash29

Everling S Dorris MC Klein RM Munoz DP Role of primate superiorcolliculus in preparation and execution of anti-saccades and pro-saccades J Neurosci 1999 19 2740ndash54

Everling S Munoz DP Neuronal correlates for preparatory set associatedwith pro-saccades and anti-saccades in the primate frontal eye fieldJ Neurosci 2000 20 387ndash400

Abnormal response monitoring in autism Brain (2008) Page 13 of 15

Falkenstein M Hohnsbein J Hoormann J Event-related correlates of errorsin reaction tasks In Karmos G Molnar M Csepe I Desmedt JE editorsPerspectives of event-related potentials research AmsterdamThe Netherlands Elsevier Science 1995 p 287ndash96

Fallgatter AJ Herrmann MJ Roemmler J Ehlis AC Wagener AHeidrich A et al Allelic variation of serotonin transporter functionmodulates the brain electrical response for error processingNeuropsychopharmacology 2004 29 1506ndash11

Fecteau JH Au C Armstrong IT Munoz DP Sensory biases producealternation advantage found in sequential saccadic eye movement tasksExp Brain Res 2004 159 84ndash91

Fischer B Breitmeyer B Mechanisms of visual attention revealed bysaccadic eye movements Neuropsychologia 1987 25 73ndash83

Fischl B Sereno MI Dale AM Cortical surface-based analysis II Inflationflattening and a surface-based coordinate system Neuroimage 1999a 9195ndash207

Fischl B Sereno MI Tootell RBH Dale AM High-resolution intersubjectaveraging and a coordinate system for the cortical surface Hum BrainMapp 1999b 8 272ndash84

Fischl B van der Kouwe A Destrieux C Halgren E Segonne F Salat DHet al Automatically parcellating the human cerebral cortex CerebCortex 2004 14 11ndash22

Fitzgerald KD Welsh RC Gehring WJ Abelson JL Himle JA Liberzon Iet al Error-related hyperactivity of the anterior cingulate cortex inobsessive-compulsive disorder Biol Psychiatry 2005 57 287ndash94

Garavan H Ross TJ Murphy K Roche RA Stein EA Dissociable executivefunctions in the dynamic control of behavior inhibition errordetection and correction NeuroImage 2002 17 1820ndash9

Gehring WJ Goss B Coles MG Meyer DE Donchin E A neuralsystem for error detection and compensation Psychol Sci 1993 4385ndash90

Gehring WJ Himle J Nisenson LG Action-monitoring dysfunction inobsessive-compulsive disorder Psychol Sci 2000 11 1ndash6

Gehring WJ Knight RT Prefrontal-cingulate interactions in actionmonitoring Nat Neurosci 2000 3 516ndash20

Goldberg MC Lasker AG Zee DS Garth E Tien A Landa RJ Deficits inthe initiation of eye movements in the absence of a visual target inadolescents with high functioning autism Neuropsychologia 2002 402039ndash49

Gomot M Bernard FA Davis MH Belmonte MK Ashwin CBullmore ET et al Change detection in children with autism anauditory event-related fMRI study Neuroimage 2006 29 475ndash84

Hajcak G Simons RF Error-related brain activity in obsessive-compulsiveundergraduates Psychiatry Res 2002 110 63ndash72

Hall GB Szechtman H Nahmias C Enhanced salience and emotionrecognition in autism a PET study Am J Psychiatry 2003 160 1439ndash41

Hallett PE Primary and secondary saccades to goals defined byinstructions Vision Res 1978 18 1279ndash96

Happe F Ronald A Plomin R Time to give up on a single explanation forautism Nat Neurosci 2006 9 1218ndash20

Harsan LA Poulet P Guignard B Steibel J Parizel N de Sousa PL et alBrain dysmyelination and recovery assessment by noninvasive in vivodiffusion tensor magnetic resonance imaging J Neurosci Res 2006 83392ndash402

Haznedar MM Buchsbaum MS Metzger M Solimando A Spiegel-Cohen J Hollander E Anterior cingulate gyrus volume and glucosemetabolism in autistic disorder Am J Psychiatry 1997 154 1047ndash50

Haznedar MM Buchsbaum MS Wei TC Hof PR Cartwright CBienstock CA et al Limbic circuitry in patients with autism spectrumdisorders studied with positron emission tomography and magneticresonance imaging Am J Psychiatry 2000 157 1994ndash2001

Henderson H Schwartz C Mundy P Burnette C Sutton S Zahka N et alResponse monitoring the error-related negativity and differences insocial behavior in autism Brain Cogn 2006 61 96ndash109

Hill EL Executive dysfunction in autism Trends Cogn Sci 2004 8 26ndash32

Hollander E King A Delaney K Smith CJ Silverman JM Obsessive-compulsive behaviors in parents of multiplex autism families PsychiatryRes 2003 117 11ndash6

Hollingshead AB Two factor index of social position New Haven CTYale University Press 1965

Holroyd CB Coles MG The neural basis of human error processingreinforcement learning dopamine and the error-related negativityPsychol Rev 2002 109 679ndash709

Jenkinson M Smith S A global optimisation method for robust affineregistration of brain images Med Image Anal 2001 5 143ndash56

Johannes S Wieringa BM Nager W Rada D Dengler R Emrich HMet al Discrepant target detection and action monitoring in obsessive-compulsive disorder Psychiatry Res 2001 108 101ndash10

Jones DK The effect of gradient sampling schemes on measures derivedfrom diffusion tensor MRI A Monte Carlo study Magn Reson Med2004 51 807ndash15

Kana RK Keller TA Minshew NJ Just MA Inhibitory control in high-functioning autism decreased activation and underconnectivity ininhibition networks Biol Psychiatry 2007 62 198ndash206

Kennedy DP Redcay E Courchesne E Failing to deactivate restingfunctional abnormalities in autism Proc Natl Acad Sci USA 2006 1038275ndash80

Klein C Foerster F Development of prosaccade and antisaccade taskperformance in participants aged 6 to 26 years Psychophysiology 200138 179ndash89

Klein TA Neumann J Reuter M Hennig J von Cramon DYUllsperger M Genetically determined differences in learning fromerrors Science 2007 318 1642ndash5

Kramer UM Cunillera T Camara E Marco-Pallares J Cucurell DNager W et al The impact of catechol-O-methyltransferase anddopamine D4 receptor genotypes on neurophysiological markers ofperformance monitoring J Neurosci 2007 27 14190ndash8

Laurens KR Ngan ET Bates AT Kiehl KA Liddle PF Rostral anteriorcingulate cortex dysfunction during error processing in schizophreniaBrain 2003 126 610ndash22

Levitt JG OrsquoNeill J Blanton RE Smalley S Fadale D McCracken JT et alProton magnetic resonance spectroscopic imaging of the brain inchildhood autism Biol Psychiatry 2003 54 1355ndash66

Leyfer OT Folstein SE Bacalman S Davis NO Dinh E Morgan J et alComorbid psychiatric disorders in children with autism interview devel-opment and rates of disorders J Autism Dev Disord 2006 36 849ndash61

London E The role of the neurobiologist in redefining the diagnosis ofautism Brain Pathol 2007 17 408ndash11

Lopez BR Lincoln AJ Ozonoff S Lai Z Examining the relationshipbetween executive functions and restricted repetitive symptoms ofAutistic Disorder J Autism Dev Disord 2005 35 445ndash60

Lord C Rutter M DiLavore PC Risi S Autism diagnostic observationschedule - WPS (ADOS-WPS) Los Angeles CA Western PsychologicalServices 1999

Luna B Doll SK Hegedus SJ Minshew NJ Sweeney JA Maturation ofexecutive function in autism Biol Psychiatry 2007 61 474ndash81

Luu P Tucker DM Derryberry D Reed M Poulsen C Electrophysiolog-ical responses to errors and feedback in the process of action regulationPsychol Sci 2003 14 47ndash53

Maltby N Tolin DF Worhunsky P OrsquoKeefe TM Kiehl KA Dysfunctionalaction monitoring hyperactivates frontal-striatal circuits in obsessive-compulsive disorder an event-related fMRI study Neuroimage 2005 24495ndash503

Manoach DS Ketwaroo GA Polli FE Thakkar KN Barton JJ Goff DCet al Reduced microstructural integrity of the white matter underlyinganterior cingulate cortex is associated with increased saccadic latency inschizophrenia Neuroimage 2007a 37 599ndash610

Manoach DS Lindgren KA Barton JJ Deficient saccadic inhibition inAspergerrsquos disorder and the social-emotional processing disorderJ Neurol Neurosurg Psychiatry 2004 75 1719ndash26

Page 14 of 15 Brain (2008) K NThakkar et al

Manoach DS Lindgren KA Cherkasova MV Goff DC Halpern EFIntriligator J et al Schizophrenic subjects show deficient inhibition butintact task-switching on saccadic tasks Biol Psychiatry 2002 51 816ndash26

Manoach DS Thakkar KN Cain MS Polli FE Edelman JA Fischl B et alNeural activity is modulated by trial history a functional magneticresonance imaging study of the effects of a previous antisaccadeJ Neurosci 2007b 27 1791ndash8

Manoach DS Weintraub S Daffner KR Scinto LF Deficient antisaccadesin the social-emotional processing disorder Neuroreport 1997 8 901ndash5

McDougle CJ Kresch LE Goodman WK Naylor ST Volkmar FRCohen DJ et al A case-controlled study of repetitive thoughts andbehavior in adults with autistic disorder and obsessive-compulsivedisorder Am J Psychiatry 1995 152 772ndash7

McPeek RM Skavenski AA Nakayama K Concurrent processing ofsaccades in visual search Vision Res 2000 40 2499ndash516

Miezin FM Maccotta L Ollinger JM Petersen SE Buckner RLCharacterizing the hemodynamic response effects of presentation ratesampling procedure and the possibility of ordering brain activity basedon relative timing Neuroimage 2000 11 735ndash59

Minshew NJ Luna B Sweeney JA Oculomotor evidence for neocorticalsystems but not cerebellar dysfunction in autism Neurology 1999 52917ndash22

Mokler A Fischer B The recognition and correction of involuntaryprosaccades in an antisaccade task Exp Brain Res 1999 125 511ndash6

Munoz DP Everling S Look away the anti-saccade task and the voluntarycontrol of eye movement Nat Rev Neurosci 2004 5 218ndash28

Murphy DG Daly E Schmitz N Toal F Murphy K Curran S et alCortical serotonin 5-HT2A receptor binding and social communicationin adults with Aspergerrsquos syndrome an in vivo SPECT study Am JPsychiatry 2006 163 934ndash6

Nieuwenhuis S Nielen MM Mol N Hajcak G Veltman DJ Performancemonitoring in obsessive-compulsive disorder Psychiatry Res 2005 134111ndash22

Nieuwenhuis S Ridderinkhof KR Blom J Band GP Kok A Error-relatedbrain potentials are differentially related to awareness of response errorsevidence from an antisaccade task Psychophysiology 2001 38 752ndash60

Ohnishi T Matsuda H Hashimoto T Kunihiro T Nishikawa M Uema Tet al Abnormal regional cerebral blood flow in childhood autism Brain2000 123 (Pt 9) 1838ndash44

Oldfield RC The assessment and analysis of handedness The EdinburghInventory Neuropsychologia 1971 9 97ndash113

Paus T Primate anterior cingulate cortex where motor control drive andcognition interface Nat Rev Neurosci 2001 2 417ndash24

Pfefferbaum A Sullivan EV Hedehus M Lim KO Adalsteinsson EMoseley M Age-related decline in brain white matter anisotropymeasured with spatially corrected echo-planar diffusion tensor imagingMagn Reson Med 2000 44 259ndash68

Pitman RK A cybernetic model of obsessive-compulsive psychopathologyCompr Psychiatry 1987 28 334ndash43

Polli FE Barton JJ Cain MS Thakkar KN Rauch SL Manoach DSRostral and dorsal anterior cingulate cortex make dissociable contribu-tions during antisaccade error commission Proc Natl Acad Sci USA2005 102 15700ndash5

Polli FE Barton JJ Thakkar KN Greve DN Goff DC Rauch SL et alReduced error-related activation in two anterior cingulate circuits is relatedto impaired performance in schizophrenia Brain 2008 131 971ndash86

Polli FE Barton JJ Vangel M Goff DC Iguchi L Manoach DSSchizophrenia patients show intact immediate error-related performanceadjustments on an antisaccade task Schizophr Res 2006 82 191ndash201

Rabbitt PM Errors and error correction in choice-response tasks J ExpPsychol 1966 71 264ndash72

Radant AD Dobie DJ Calkins ME Olincy A Braff DL Cadenhead KSet al Successful multi-site measurement of antisaccade performancedeficits in schizophrenia Schizophr Res 2007 89 320ndash9

Reese TG Heid O Weisskoff RM Wedeen VJ Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echoMagn Reson Med 2003 49 177ndash82

Russell J Jarrold C Error-correction problems in autism evidence for amonitoring impairment J Autism Dev Disord 1998 28 177ndash88

Russell AJ Mataix-Cols D Anson M Murphy DG Obsessions andcompulsions in Asperger syndrome and high-functioning autism Br JPsychiatry 2005 186 525ndash8

Rutter M Le Couteur A Lord C Autism diagnostic interview-revised LosAngeles CA Western Psychological Services 2003

Santesso DL Segalowitz SJ Schmidt LA Error-related electrocorticalresponses are enhanced in children with obsessive-compulsive behaviorsDev Neuropsychol 2006 29 431ndash45

Shafritz KM Dichter GS Baranek GT Belger A The neural circuitrymediating shifts in behavioral response and cognitive set in autism BiolPsychiatry 2007 63 974ndash80

Silk TJ Rinehart N Bradshaw JL Tonge B Egan G OrsquoBoyle MW et alVisuospatial processing and the function of prefrontal-parietal networksin autism spectrum disorders a functional MRI study Am J Psychiatry2006 163 1440ndash3

South M Ozonoff S McMahon WM The relationship between executivefunctioning central coherence and repetitive behaviors in the high-functioning autism spectrum Autism 2007 11 437ndash51

Taylor SF Martis B Fitzgerald KD Welsh RC Abelson JL Liberzon Iet al Medial frontal cortex activity and loss-related responses to errorsJ Neurosci 2006 26 4063ndash70

Taylor SF Stern ER Gehring WJ Neural systems for error monitoringrecent findings and theoretical perspectives Neuroscientist 2007 13160ndash72

Thesen S Heid O Mueller E Schad LR Prospective acquisition correctionfor head motion with image-based tracking for real-time fMRI MagnReson Med 2000 44 457ndash65

Turner M Annotation Repetitive behaviour in autism areview of psychological research J Child Psychol Psychiatry 1999 40839ndash49

Ullsperger M von Cramon DY Neuroimaging of performance monitoringerror detection and beyond Cortex 2004 40 593ndash604

Ursu S Stenger VA Shear MK Jones MR Carter CS Overactiveaction monitoring in obsessive-compulsive disorder evidencefrom functional magnetic resonance imaging Psychol Sci 2003 14347ndash53

Meel CS Heslenfeld DJ Oosterlaan J Sergeant JA Adaptive controldeficits in attention-deficithyperactivity disorder (ADHD) the role oferror processing Psychiatry Res 2007 151 211ndash20

van Veen V Carter CS The timing of action-monitoring processes in theanterior cingulate cortex J Cogn Neurosci 2002 14 593ndash602

Wechsler D Wechsler Abbreviated Scale of intelligence San Antonio TXThe Psychological Corporation 1999

Whalen PJ Bush G McNally RJ Wilhelm S McInerney SC Jenike MAet al The emotional counting Stroop paradigm a functional magneticresonance imaging probe of the anterior cingulate affective division BiolPsychiatry 1998 44 1219ndash28

White K Ashton R Handedness assessment inventory Neuropsychologia1976 14 261ndash4

Zandt F Prior M Kyrios M Repetitive behaviour in children with highfunctioning autism and obsessive compulsive disorder J Autism DevDisord 2007 37 251ndash9

Abnormal response monitoring in autism Brain (2008) Page 15 of 15

on ideational but not sensorimotor items (Hollander et al2003) This finding supports the distinction betweenrepetition of movements and repetition of more complexideational behaviours (for review see Turner 1999) andsuggests that the latter may share underlying mechanisms inOCD and ASD In the present study both sensorimotorand ideational subscores contributed to the relations seenwith response monitoring activation and FA At presentrepetitive behaviours in autism are incompletely under-stood and neurobiologically valid dimensions are yet to bedelineated The present findings suggest that ACC abnorm-alities may contribute to repetitive behaviour in ASDpossibly on the basis of their contribution to aberrantresponse monitoring but linking ACC abnormalities tospecific dimensions of repetitive behaviour and identifyingthe underlying mechanisms requires further study withlarge samplesAbnormal response monitoring may also contribute to

deficient cognitive control of behaviour in ASD which inthe present study was reflected in an increased antisaccadeerror rate consistent with prior work (Manoach et al1997 2004 Minshew et al 1999 Goldberg et al 2002Luna et al 2007) and possibly in reduced dynamicperformance adjustment In spite of making more errorsASD participants responded more quickly than controls oncorrect trials suggesting a disruption in the speed-accuracytrade-off This apparent difficulty in modulating responsespeed to improve accuracy is unlikely to reflect reducederror awareness since ASD participants were as or morelikely than controls to self-correct errors Nor is it likely toreflect a blunted neural response to errors given both theincreased response to errors in left perigenual rACC andprevious work showing normal or increased amplitude ofthe ERN in ASD (Henderson et al 2006) Deficient use ofoutcomes to adjust performance in a continuous dynamicfashion may stem from response monitoring dysfunction inthe ACC andor from dysconnection of the ACC fromocular motor regions involved in saccade generationAberrant connectivity of ACC regions subserving responsemonitoring in ASD is suggested by our finding of reducedFA in the underlying WM This finding is consistent with aprevious report of reduced FA in ACC white matter inautism (Barnea-Goraly et al 2004) and with evidence ofdecreased functional connectivity of the ACC to otherregions during a task involving response inhibition (Kanaet al 2007)How might information concerning the outcome of the

prior response be communicated to ocular motor regionsresponsible for adjusting behaviour in the subsequent trialMonkey electrophysiology studies suggest a basis for speed-accuracy trade-offs in the ocular motor system Comparedwith prosaccades antisaccades are associated with reducedpre-target activity of saccade-related neurons in the frontaleye field (Everling et al 1999 Everling and Munoz 2000)and superior colliculus (Everling et al 1999) key regions insaccade generation Moreover lower pre-target frontal eye

field activity is correlated with increased saccadic latencyand fewer antisaccade errors Reduced preparatory neuralactivity in response to the instruction to perform anantisaccade presumably makes it more difficult for activityrelated to the incorrect but prepotent prosaccade to reachthe threshold for triggering an eye movement However theprice for this improved accuracy is that the activitygenerating the correct antisaccade also takes longer toreach threshold resulting in prolonged saccadic latenciesThus we hypothesize that the slowing of responses thatfollow an antisaccade error (ie post-error slowingNieuwenhuis et al 2001) stems from an ACC errorresponse that signals the frontal eye field to reduce itspreparatory activity during the subsequent trial Thishypothesis is consistent with previous work demonstratingthat other aspects of trial history affect preparatory neuralactivity in the frontal eye field during antisaccades(Manoach et al 2007b) How might error signals beconveyed to frontal eye field The use of outcomes todynamically adjust responses has been attributed to ACCacting with lateral prefrontal cortex (Gehring and Knight2000 Paus 2001 Garavan et al 2002) and lateralprefrontal cortex has been proposed to provide inhibitoryinput to frontal eye field during antisaccades (DeSouzaet al 2003) Thus in ASD aberrant communicationbetween response evaluation in the ACC and regionsinitiating compensatory adjustments possibly lateral pre-frontal cortex could result in a failure to optimally reducepreparatory firing rates in frontal eye field in response toa prior error This reduced dynamic performance adjust-ment could lead to faster responding and more errorsWe are presently testing this hypothesis using an anti-saccade paradigm that gives rise to post-error slowing(Rabbitt 1966) and will allow a more direct examinationof compensatory behaviour and its supporting neuralcircuitry Unfortunately although post-error slowing isseen when antisaccades are the only task (Polli et al 2006)the intermixed antisaccade and prosaccade paradigmemployed here gives rise to other inter-trial factors thatalso affect latency (eg task-switching and prior antisac-cades Cherkasova et al 2002 Fecteau et al 2004 Bartonet al 2006 Manoach et al 2007b) and may obscure theeffects of a prior error There is evidence of reduced post-error slowing in adults with high functioning autismhowever from a previous study that employed a memorysearch task (Bogte et al 2007) Generalizing beyond theocular motor system failures of dynamic performanceadjustment due to ACC dysfunction and dysconnectivitymay contribute to the rigid and perseverative behavioursthat characterize ASD

Repetitive behaviour scores in ASD were also associatedwith reduced FA in the WM underlying left subgenualrACC Subgenual rACC is part of a network that is thoughtto play an important role in generating autonomicresponses to motivationally salient stimuli evaluatingtheir significance in predicting reward and in using this

Abnormal response monitoring in autism Brain (2008) Page 11 of 15

information to guide behaviour (Damasio et al 1990Drevets 2000) An important consideration in interpretingthis relation of FA to repetition is that subgenual rACCregion was not functionally or structurally abnormal in thisASD sample This suggests that even normal variation in FAof the subgenual rACC may contribute to a tendency torepeat unreinforced stimulus-response mappings Moregenerally the relations we observed between behaviouralrepetition and ACC function and structure may not bespecific to ASD Accumulating evidence suggests that thethree symptom clusters that define autismmdashsocial impair-ment communication deficits and restricted repetitivebehaviourmdasharise from distinct genetic and cognitivemechanisms and may represent continuous traits in thepopulation (for reviews see Happe et al 2006 London2007) Previous studies have reported relations betweenindices of response monitoring and repetition in OCD(Gehring et al 2000 Ursu et al 2003) and in non-clinically ascertained samples of children and adults withobsessivendashcompulsive characteristics (Hajcak and Simons2002 Santesso et al 2006)Our findings also raise the question of whether the

response monitoring abnormalities that we observed arespecific to ASD Similar abnormalities have been reportedin OCD (Maltby et al 2005) and attention deficithyperactivity disorder (van Meel et al 2007) Whilemethodological differences preclude a direct comparisonof these findings we previously studied patients withschizophrenia using the identical paradigm (Polli et al2008) Schizophrenia is also a neurodevelopmental disorderthat is characterized by rigid and perseverative behaviourresponse monitoring abnormalities (Carter et al 2001)reduced FA in ACC WM (Manoach et al 2007a) and aconsistently elevated antisaccade error rate (Radant et al2007 Manoach et al 2002) Although both schizophreniaand ASD samples made more antisaccade errors thancontrols ASD participants performed correct trials fasterand schizophrenia patients performed more slowly Whileboth groups showed reduced discrimination between errorand correct trials in the ACC this occurred for differentreasons In ASD this reflected a hyperactive response tocorrect trials while in schizophrenia it reflected a bluntedACC response to errors consistent with numerous previousreports (Laurens et al 2003) These differences argue fordistinct behavioural and neural signatures of responsemonitoring deficits in schizophrenia and ASD Thesefindings along with those of recent neuroimaging studiessuggesting genetic mediation of response monitoring(Fallgatter et al 2004 Klein et al 2007 Kramer et al2007) support the candidacy of neuroimaging-basedindices of response monitoring as neurocognitive endophe-notypes Whether response monitoring abnormalities canbe distinguished in ASD OCD and other neuropsychiatricdisorders whether they are related to diagnosis or specifictraits and whether they are also present in unaffectedfamily members remains to be seen

Several additional limitations to and alternative inter-pretations of our findings merit consideration First mightthe hyperactive rACC response to correct antisaccades inASD reflect a less efficient response to conflict ratherthan response monitoring The ACC is thought to play arole in detecting conflict and engaging cognitive controlto resolve it (for review see Carter and van Veen 2007)Antisaccades involve a high degree of conflict as one mustchoose between two competing responses and inhibit theprepotent one Errors give rise to another type of conflictwhen the actual response is compared with the intendedone Thus both correct and erroneous antisaccadesengender conflict While it is not possible to entirely ruleout the conflict explanation several observations argueagainst it If this rACC region were detecting conflict itshould respond to both correct and error trials Insteadcontrols showed a significant response only to errors Inaddition the timing of the haemodynamic response tocorrect antisaccades in ASD which started amp4 s and peakedat amp8 s (Fig 3) argues for a role in response monitoringrather than successful conflict resolution In this paradigmdifferential fMRI activation for correct antisaccade versusprosaccade trials including in dACC presumably reflectingconflict detection response inhibition and saccade genera-tion peaks at amp4 s and is mostly gone by 8 s (Manoachet al 2007b see also Polli et al 2005 for a discussionof ACC regions involved in saccade generation versusresponse monitoring) Finally it is dorsal rather than rACCactivation that is associated with engaging cognitive controlon correct high-conflict and error trials (Carter and vanVeen 2007)

An important limitation is that the small sample size ofthe present study may have compromised our power todetect additional group differences and relations of interestparticularly in regions outside our a priori ACC regions ofinterest for which more stringent multiple comparisonscorrections were applied In spite of the relatively smallsample size our a priori hypotheses were confirmedIn addition four of our ASD participants were takingmedications that could affect ACC function A re-analysisof the fMRI data with only unmedicated ASD participantshowever supported our main findings of significantly lessdiscrimination between response outcomes in the rACCand greater activation to correct trials It is also important tonote that our sample was limited to high functioning adultswith ASD so it is not clear that our findings would generalizeto lower functioning or younger samples Although increasedantisaccade error rates are seen in autism as early as ages 8ndash12(Luna et al 2007) we limited our study to adults sincesaccadic inhibition may not fully develop until late adoles-cence (Klein and Foerster 2001) and larger samples would benecessary to discriminate between the effects of diagnosis andthose due to normal development Another limitationconcerns our interpretation of FA as reflecting connectivityFA reflects the degree of directional coherence of waterdiffusion in tissue and is therefore an indirect measure of

Page 12 of 15 Brain (2008) K NThakkar et al

WMmicrostructure While FA correlates with axon myelinat-ion (Harsan et al 2006) not all of its biophysicaldeterminants are fully understood (Beaulieu 2002) Thus itis not possible to attribute FA differences or the relations weobserved to a particular WM propertyIn summary the present study reports hyperactive

response monitoring in the rACC in ASD that is relatedto restricted repetitive behaviour and is accompanied byreduced FA in the underlying white matter This suggests astructural correlate to abnormal function that may com-promise connectivity in the circuitry subserving responsemonitoring These findings suggest that structural and fun-ctional abnormalities of the ACC compromise responsemonitoring and contribute to behavioural repetition inASD Impairments in evaluating and learning from errorsmay lead to behaviour that is rigid and repetitive ratherthan optimally guided by outcomes and may compromiseperformance across a wide range of tasks Illuminating theneural basis and clinical significance of response monitoringdeficits in ASD represents a fruitful avenue for furtherresearch

Supplementary materialSupplementary material is available at Brain online

AcknowledgementsThe authors wish to thank Mark Vangel for statisticalconsultation Robert Levy Kim Ono Christopher Shermanand Jeremy Young for help with manuscript preparationand Matt Cain and Jay Edelman for technical assistanceSome of these data were presented at the annual meetingof the Society for Neuroscience in Washington DC inNovember 2005 This study was supported by NationalInstitute for Mental Health [R01 MH67720 (DSM)MH72120 (FEP)] Mental Illness Neuroscience Discovery(MIND) Institute (DOE DE-FG02-99ER62764) TheNational Center for Research Resources (P41RR14075)Funding to pay the Open Access publication charges forthis article was provided by the National Institute forMental Health NRSA MH72120

ReferencesAshwin C Baron-Cohen S Wheelwright S OrsquoRiordan M Bullmore ETDifferential activation of the amygdala and the lsquosocial brainrsquo duringfearful face-processing in Asperger syndrome Neuropsychologia 200745 2ndash14

Barnea-Goraly N Kwon H Menon V Eliez S Lotspeich L Reiss ALWhite matter structure in autism preliminary evidence from diffusiontensor imaging Biol Psychiatry 2004 55 323ndash6

Baron-Cohen S Do autistic children have obsessions and compulsionsBr J Clin Psychol 1989 28 (Pt 3) 193ndash200

Barton JJ Greenzang C Hefter R Edelman J Manoach DS Switchingplasticity and prediction in a saccadic task-switch paradigm Exp BrainRes 2006 168 76ndash87

Basser PJ Mattiello J LeBihan D MR diffusion tensor spectroscopy andimaging Biophys J 1994 66 259ndash67

Beaulieu C The basis of anisotropic water diffusion in the nervous system- a technical review NMR Biomed 2002 15 435ndash55

Blair JR Spreen O Predicting premorbid IQ a revision of the NationalAdult Reading Test Clin Neuropsychol 1989 3 129ndash136

Bogte H Flamma B van der Meere J van Engeland H Post-erroradaptation in adults with high functioning autism Neuropsychologia2007 45 1707ndash14

Breiter HC Rauch SL Kwong KK Baker JR Weisskoff RM Kennedy DNet al Functional magnetic resonance imaging of symptom provocationin obsessive-compulsive disorder Arch Gen Psychiatry 1996 53595ndash606

Buckner RL Goodman J Burock M Rotte M Koutstaal W Schacter Det al Functional-anatomic correlates of object priming in humansrevealed by rapid presentation event-related fMRI Neuron 1998 20285ndash96

Burock MA Dale AM Estimation and detection of event-related fMRIsignals with temporally correlated noise a statistically efficient andunbiased approach Hum Brain Mapp 2000 11 249ndash60

Bush G Whalen PJ Rosen BR Jenike MA McInerney SC Rauch SL Thecounting stroop an interference task specialized for functionalneuroimaging- -validation study with functional MRI Hum BrainMapp 1998 6 270ndash82

Carter CS MacDonald AW 3rd Ross LL Stenger VA Anterior cingulatecortex activity and impaired self-monitoring of performance in patientswith schizophrenia an event-related fMRI study Am J Psychiatry 2001158 1423ndash8

Carter CS van Veen V Anterior cingulate cortex and conflict detection anupdate of theory and data Cogn Affect Behav Neurosci 2007 7 367ndash79

Cherkasova MV Manoach DS Intriligator JM Barton JJS Antisaccadesand task-switching interactions in controlled processing Exp Brain Res2002 144 528ndash37

Collins DL Neelin P Peters TM Evans AC Automatic 3D intersubjectregistration of MR volumetric data in standardized Talairach spaceJ Comput Assist Tomogr 1994 18 192ndash205

Courchesne E Pierce K Why the frontal cortex in autism might be talkingonly to itself local over-connectivity but long-distance disconnectionCurr Opin Neurobiol 2005 15 225ndash30

Cox RW Jesmanowicz A Real-time 3D image registration for functionalMRI Magn Reson Med 1999 42 1014ndash8

Dale AM Optimal experimental design for event-related fMRI Hum BrainMapp 1999 8 109ndash40

Dale AM Fischl B Sereno MI Cortical surface-based analysisI Segmentation and surface reconstruction Neuroimage 1999 9 179ndash94

Damasio AR Tranel D Damasio H Individuals with sociopathic behaviorcaused by frontal damage fail to respond autonomically to social stimuliBehav Brain Res 1990 41 81ndash94

Dehaene S Posner MI Tucker DM Localization of a neural system forerror detection and compensation Psychol Sci 1994 5 303ndash5

DeSouza JF Menon RS Everling S Preparatory set associated with pro-saccades and anti-saccades in humans investigated with event-relatedFMRI J Neurophysiol 2003 89 1016ndash23

Devinsky O Morrell MJ Vogt BA Contributions of anterior cingulatecortex to behaviour Brain 1995 118 (Pt 1) 279ndash306

Dichter GS Belger A Social stimuli interfere with cognitive control inautism Neuroimage 2007 35 1219ndash30

Dougherty DD Baer L Cosgrove GR Cassem EH Price BHNierenberg AA et al Prospective long-term follow-up of 44 patientswho received cingulotomy for treatment-refractory obsessive-compulsivedisorder Am J Psychiatry 2002 159 269ndash75

Drevets WC Neuroimaging studies of mood disorders Biol Psychiatry2000 48 813ndash29

Everling S Dorris MC Klein RM Munoz DP Role of primate superiorcolliculus in preparation and execution of anti-saccades and pro-saccades J Neurosci 1999 19 2740ndash54

Everling S Munoz DP Neuronal correlates for preparatory set associatedwith pro-saccades and anti-saccades in the primate frontal eye fieldJ Neurosci 2000 20 387ndash400

Abnormal response monitoring in autism Brain (2008) Page 13 of 15

Falkenstein M Hohnsbein J Hoormann J Event-related correlates of errorsin reaction tasks In Karmos G Molnar M Csepe I Desmedt JE editorsPerspectives of event-related potentials research AmsterdamThe Netherlands Elsevier Science 1995 p 287ndash96

Fallgatter AJ Herrmann MJ Roemmler J Ehlis AC Wagener AHeidrich A et al Allelic variation of serotonin transporter functionmodulates the brain electrical response for error processingNeuropsychopharmacology 2004 29 1506ndash11

Fecteau JH Au C Armstrong IT Munoz DP Sensory biases producealternation advantage found in sequential saccadic eye movement tasksExp Brain Res 2004 159 84ndash91

Fischer B Breitmeyer B Mechanisms of visual attention revealed bysaccadic eye movements Neuropsychologia 1987 25 73ndash83

Fischl B Sereno MI Dale AM Cortical surface-based analysis II Inflationflattening and a surface-based coordinate system Neuroimage 1999a 9195ndash207

Fischl B Sereno MI Tootell RBH Dale AM High-resolution intersubjectaveraging and a coordinate system for the cortical surface Hum BrainMapp 1999b 8 272ndash84

Fischl B van der Kouwe A Destrieux C Halgren E Segonne F Salat DHet al Automatically parcellating the human cerebral cortex CerebCortex 2004 14 11ndash22

Fitzgerald KD Welsh RC Gehring WJ Abelson JL Himle JA Liberzon Iet al Error-related hyperactivity of the anterior cingulate cortex inobsessive-compulsive disorder Biol Psychiatry 2005 57 287ndash94

Garavan H Ross TJ Murphy K Roche RA Stein EA Dissociable executivefunctions in the dynamic control of behavior inhibition errordetection and correction NeuroImage 2002 17 1820ndash9

Gehring WJ Goss B Coles MG Meyer DE Donchin E A neuralsystem for error detection and compensation Psychol Sci 1993 4385ndash90

Gehring WJ Himle J Nisenson LG Action-monitoring dysfunction inobsessive-compulsive disorder Psychol Sci 2000 11 1ndash6

Gehring WJ Knight RT Prefrontal-cingulate interactions in actionmonitoring Nat Neurosci 2000 3 516ndash20

Goldberg MC Lasker AG Zee DS Garth E Tien A Landa RJ Deficits inthe initiation of eye movements in the absence of a visual target inadolescents with high functioning autism Neuropsychologia 2002 402039ndash49

Gomot M Bernard FA Davis MH Belmonte MK Ashwin CBullmore ET et al Change detection in children with autism anauditory event-related fMRI study Neuroimage 2006 29 475ndash84

Hajcak G Simons RF Error-related brain activity in obsessive-compulsiveundergraduates Psychiatry Res 2002 110 63ndash72

Hall GB Szechtman H Nahmias C Enhanced salience and emotionrecognition in autism a PET study Am J Psychiatry 2003 160 1439ndash41

Hallett PE Primary and secondary saccades to goals defined byinstructions Vision Res 1978 18 1279ndash96

Happe F Ronald A Plomin R Time to give up on a single explanation forautism Nat Neurosci 2006 9 1218ndash20

Harsan LA Poulet P Guignard B Steibel J Parizel N de Sousa PL et alBrain dysmyelination and recovery assessment by noninvasive in vivodiffusion tensor magnetic resonance imaging J Neurosci Res 2006 83392ndash402

Haznedar MM Buchsbaum MS Metzger M Solimando A Spiegel-Cohen J Hollander E Anterior cingulate gyrus volume and glucosemetabolism in autistic disorder Am J Psychiatry 1997 154 1047ndash50

Haznedar MM Buchsbaum MS Wei TC Hof PR Cartwright CBienstock CA et al Limbic circuitry in patients with autism spectrumdisorders studied with positron emission tomography and magneticresonance imaging Am J Psychiatry 2000 157 1994ndash2001

Henderson H Schwartz C Mundy P Burnette C Sutton S Zahka N et alResponse monitoring the error-related negativity and differences insocial behavior in autism Brain Cogn 2006 61 96ndash109

Hill EL Executive dysfunction in autism Trends Cogn Sci 2004 8 26ndash32

Hollander E King A Delaney K Smith CJ Silverman JM Obsessive-compulsive behaviors in parents of multiplex autism families PsychiatryRes 2003 117 11ndash6

Hollingshead AB Two factor index of social position New Haven CTYale University Press 1965

Holroyd CB Coles MG The neural basis of human error processingreinforcement learning dopamine and the error-related negativityPsychol Rev 2002 109 679ndash709

Jenkinson M Smith S A global optimisation method for robust affineregistration of brain images Med Image Anal 2001 5 143ndash56

Johannes S Wieringa BM Nager W Rada D Dengler R Emrich HMet al Discrepant target detection and action monitoring in obsessive-compulsive disorder Psychiatry Res 2001 108 101ndash10

Jones DK The effect of gradient sampling schemes on measures derivedfrom diffusion tensor MRI A Monte Carlo study Magn Reson Med2004 51 807ndash15

Kana RK Keller TA Minshew NJ Just MA Inhibitory control in high-functioning autism decreased activation and underconnectivity ininhibition networks Biol Psychiatry 2007 62 198ndash206

Kennedy DP Redcay E Courchesne E Failing to deactivate restingfunctional abnormalities in autism Proc Natl Acad Sci USA 2006 1038275ndash80

Klein C Foerster F Development of prosaccade and antisaccade taskperformance in participants aged 6 to 26 years Psychophysiology 200138 179ndash89

Klein TA Neumann J Reuter M Hennig J von Cramon DYUllsperger M Genetically determined differences in learning fromerrors Science 2007 318 1642ndash5

Kramer UM Cunillera T Camara E Marco-Pallares J Cucurell DNager W et al The impact of catechol-O-methyltransferase anddopamine D4 receptor genotypes on neurophysiological markers ofperformance monitoring J Neurosci 2007 27 14190ndash8

Laurens KR Ngan ET Bates AT Kiehl KA Liddle PF Rostral anteriorcingulate cortex dysfunction during error processing in schizophreniaBrain 2003 126 610ndash22

Levitt JG OrsquoNeill J Blanton RE Smalley S Fadale D McCracken JT et alProton magnetic resonance spectroscopic imaging of the brain inchildhood autism Biol Psychiatry 2003 54 1355ndash66

Leyfer OT Folstein SE Bacalman S Davis NO Dinh E Morgan J et alComorbid psychiatric disorders in children with autism interview devel-opment and rates of disorders J Autism Dev Disord 2006 36 849ndash61

London E The role of the neurobiologist in redefining the diagnosis ofautism Brain Pathol 2007 17 408ndash11

Lopez BR Lincoln AJ Ozonoff S Lai Z Examining the relationshipbetween executive functions and restricted repetitive symptoms ofAutistic Disorder J Autism Dev Disord 2005 35 445ndash60

Lord C Rutter M DiLavore PC Risi S Autism diagnostic observationschedule - WPS (ADOS-WPS) Los Angeles CA Western PsychologicalServices 1999

Luna B Doll SK Hegedus SJ Minshew NJ Sweeney JA Maturation ofexecutive function in autism Biol Psychiatry 2007 61 474ndash81

Luu P Tucker DM Derryberry D Reed M Poulsen C Electrophysiolog-ical responses to errors and feedback in the process of action regulationPsychol Sci 2003 14 47ndash53

Maltby N Tolin DF Worhunsky P OrsquoKeefe TM Kiehl KA Dysfunctionalaction monitoring hyperactivates frontal-striatal circuits in obsessive-compulsive disorder an event-related fMRI study Neuroimage 2005 24495ndash503

Manoach DS Ketwaroo GA Polli FE Thakkar KN Barton JJ Goff DCet al Reduced microstructural integrity of the white matter underlyinganterior cingulate cortex is associated with increased saccadic latency inschizophrenia Neuroimage 2007a 37 599ndash610

Manoach DS Lindgren KA Barton JJ Deficient saccadic inhibition inAspergerrsquos disorder and the social-emotional processing disorderJ Neurol Neurosurg Psychiatry 2004 75 1719ndash26

Page 14 of 15 Brain (2008) K NThakkar et al

Manoach DS Lindgren KA Cherkasova MV Goff DC Halpern EFIntriligator J et al Schizophrenic subjects show deficient inhibition butintact task-switching on saccadic tasks Biol Psychiatry 2002 51 816ndash26

Manoach DS Thakkar KN Cain MS Polli FE Edelman JA Fischl B et alNeural activity is modulated by trial history a functional magneticresonance imaging study of the effects of a previous antisaccadeJ Neurosci 2007b 27 1791ndash8

Manoach DS Weintraub S Daffner KR Scinto LF Deficient antisaccadesin the social-emotional processing disorder Neuroreport 1997 8 901ndash5

McDougle CJ Kresch LE Goodman WK Naylor ST Volkmar FRCohen DJ et al A case-controlled study of repetitive thoughts andbehavior in adults with autistic disorder and obsessive-compulsivedisorder Am J Psychiatry 1995 152 772ndash7

McPeek RM Skavenski AA Nakayama K Concurrent processing ofsaccades in visual search Vision Res 2000 40 2499ndash516

Miezin FM Maccotta L Ollinger JM Petersen SE Buckner RLCharacterizing the hemodynamic response effects of presentation ratesampling procedure and the possibility of ordering brain activity basedon relative timing Neuroimage 2000 11 735ndash59

Minshew NJ Luna B Sweeney JA Oculomotor evidence for neocorticalsystems but not cerebellar dysfunction in autism Neurology 1999 52917ndash22

Mokler A Fischer B The recognition and correction of involuntaryprosaccades in an antisaccade task Exp Brain Res 1999 125 511ndash6

Munoz DP Everling S Look away the anti-saccade task and the voluntarycontrol of eye movement Nat Rev Neurosci 2004 5 218ndash28

Murphy DG Daly E Schmitz N Toal F Murphy K Curran S et alCortical serotonin 5-HT2A receptor binding and social communicationin adults with Aspergerrsquos syndrome an in vivo SPECT study Am JPsychiatry 2006 163 934ndash6

Nieuwenhuis S Nielen MM Mol N Hajcak G Veltman DJ Performancemonitoring in obsessive-compulsive disorder Psychiatry Res 2005 134111ndash22

Nieuwenhuis S Ridderinkhof KR Blom J Band GP Kok A Error-relatedbrain potentials are differentially related to awareness of response errorsevidence from an antisaccade task Psychophysiology 2001 38 752ndash60

Ohnishi T Matsuda H Hashimoto T Kunihiro T Nishikawa M Uema Tet al Abnormal regional cerebral blood flow in childhood autism Brain2000 123 (Pt 9) 1838ndash44

Oldfield RC The assessment and analysis of handedness The EdinburghInventory Neuropsychologia 1971 9 97ndash113

Paus T Primate anterior cingulate cortex where motor control drive andcognition interface Nat Rev Neurosci 2001 2 417ndash24

Pfefferbaum A Sullivan EV Hedehus M Lim KO Adalsteinsson EMoseley M Age-related decline in brain white matter anisotropymeasured with spatially corrected echo-planar diffusion tensor imagingMagn Reson Med 2000 44 259ndash68

Pitman RK A cybernetic model of obsessive-compulsive psychopathologyCompr Psychiatry 1987 28 334ndash43

Polli FE Barton JJ Cain MS Thakkar KN Rauch SL Manoach DSRostral and dorsal anterior cingulate cortex make dissociable contribu-tions during antisaccade error commission Proc Natl Acad Sci USA2005 102 15700ndash5

Polli FE Barton JJ Thakkar KN Greve DN Goff DC Rauch SL et alReduced error-related activation in two anterior cingulate circuits is relatedto impaired performance in schizophrenia Brain 2008 131 971ndash86

Polli FE Barton JJ Vangel M Goff DC Iguchi L Manoach DSSchizophrenia patients show intact immediate error-related performanceadjustments on an antisaccade task Schizophr Res 2006 82 191ndash201

Rabbitt PM Errors and error correction in choice-response tasks J ExpPsychol 1966 71 264ndash72

Radant AD Dobie DJ Calkins ME Olincy A Braff DL Cadenhead KSet al Successful multi-site measurement of antisaccade performancedeficits in schizophrenia Schizophr Res 2007 89 320ndash9

Reese TG Heid O Weisskoff RM Wedeen VJ Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echoMagn Reson Med 2003 49 177ndash82

Russell J Jarrold C Error-correction problems in autism evidence for amonitoring impairment J Autism Dev Disord 1998 28 177ndash88

Russell AJ Mataix-Cols D Anson M Murphy DG Obsessions andcompulsions in Asperger syndrome and high-functioning autism Br JPsychiatry 2005 186 525ndash8

Rutter M Le Couteur A Lord C Autism diagnostic interview-revised LosAngeles CA Western Psychological Services 2003

Santesso DL Segalowitz SJ Schmidt LA Error-related electrocorticalresponses are enhanced in children with obsessive-compulsive behaviorsDev Neuropsychol 2006 29 431ndash45

Shafritz KM Dichter GS Baranek GT Belger A The neural circuitrymediating shifts in behavioral response and cognitive set in autism BiolPsychiatry 2007 63 974ndash80

Silk TJ Rinehart N Bradshaw JL Tonge B Egan G OrsquoBoyle MW et alVisuospatial processing and the function of prefrontal-parietal networksin autism spectrum disorders a functional MRI study Am J Psychiatry2006 163 1440ndash3

South M Ozonoff S McMahon WM The relationship between executivefunctioning central coherence and repetitive behaviors in the high-functioning autism spectrum Autism 2007 11 437ndash51

Taylor SF Martis B Fitzgerald KD Welsh RC Abelson JL Liberzon Iet al Medial frontal cortex activity and loss-related responses to errorsJ Neurosci 2006 26 4063ndash70

Taylor SF Stern ER Gehring WJ Neural systems for error monitoringrecent findings and theoretical perspectives Neuroscientist 2007 13160ndash72

Thesen S Heid O Mueller E Schad LR Prospective acquisition correctionfor head motion with image-based tracking for real-time fMRI MagnReson Med 2000 44 457ndash65

Turner M Annotation Repetitive behaviour in autism areview of psychological research J Child Psychol Psychiatry 1999 40839ndash49

Ullsperger M von Cramon DY Neuroimaging of performance monitoringerror detection and beyond Cortex 2004 40 593ndash604

Ursu S Stenger VA Shear MK Jones MR Carter CS Overactiveaction monitoring in obsessive-compulsive disorder evidencefrom functional magnetic resonance imaging Psychol Sci 2003 14347ndash53

Meel CS Heslenfeld DJ Oosterlaan J Sergeant JA Adaptive controldeficits in attention-deficithyperactivity disorder (ADHD) the role oferror processing Psychiatry Res 2007 151 211ndash20

van Veen V Carter CS The timing of action-monitoring processes in theanterior cingulate cortex J Cogn Neurosci 2002 14 593ndash602

Wechsler D Wechsler Abbreviated Scale of intelligence San Antonio TXThe Psychological Corporation 1999

Whalen PJ Bush G McNally RJ Wilhelm S McInerney SC Jenike MAet al The emotional counting Stroop paradigm a functional magneticresonance imaging probe of the anterior cingulate affective division BiolPsychiatry 1998 44 1219ndash28

White K Ashton R Handedness assessment inventory Neuropsychologia1976 14 261ndash4

Zandt F Prior M Kyrios M Repetitive behaviour in children with highfunctioning autism and obsessive compulsive disorder J Autism DevDisord 2007 37 251ndash9

Abnormal response monitoring in autism Brain (2008) Page 15 of 15

information to guide behaviour (Damasio et al 1990Drevets 2000) An important consideration in interpretingthis relation of FA to repetition is that subgenual rACCregion was not functionally or structurally abnormal in thisASD sample This suggests that even normal variation in FAof the subgenual rACC may contribute to a tendency torepeat unreinforced stimulus-response mappings Moregenerally the relations we observed between behaviouralrepetition and ACC function and structure may not bespecific to ASD Accumulating evidence suggests that thethree symptom clusters that define autismmdashsocial impair-ment communication deficits and restricted repetitivebehaviourmdasharise from distinct genetic and cognitivemechanisms and may represent continuous traits in thepopulation (for reviews see Happe et al 2006 London2007) Previous studies have reported relations betweenindices of response monitoring and repetition in OCD(Gehring et al 2000 Ursu et al 2003) and in non-clinically ascertained samples of children and adults withobsessivendashcompulsive characteristics (Hajcak and Simons2002 Santesso et al 2006)Our findings also raise the question of whether the

response monitoring abnormalities that we observed arespecific to ASD Similar abnormalities have been reportedin OCD (Maltby et al 2005) and attention deficithyperactivity disorder (van Meel et al 2007) Whilemethodological differences preclude a direct comparisonof these findings we previously studied patients withschizophrenia using the identical paradigm (Polli et al2008) Schizophrenia is also a neurodevelopmental disorderthat is characterized by rigid and perseverative behaviourresponse monitoring abnormalities (Carter et al 2001)reduced FA in ACC WM (Manoach et al 2007a) and aconsistently elevated antisaccade error rate (Radant et al2007 Manoach et al 2002) Although both schizophreniaand ASD samples made more antisaccade errors thancontrols ASD participants performed correct trials fasterand schizophrenia patients performed more slowly Whileboth groups showed reduced discrimination between errorand correct trials in the ACC this occurred for differentreasons In ASD this reflected a hyperactive response tocorrect trials while in schizophrenia it reflected a bluntedACC response to errors consistent with numerous previousreports (Laurens et al 2003) These differences argue fordistinct behavioural and neural signatures of responsemonitoring deficits in schizophrenia and ASD Thesefindings along with those of recent neuroimaging studiessuggesting genetic mediation of response monitoring(Fallgatter et al 2004 Klein et al 2007 Kramer et al2007) support the candidacy of neuroimaging-basedindices of response monitoring as neurocognitive endophe-notypes Whether response monitoring abnormalities canbe distinguished in ASD OCD and other neuropsychiatricdisorders whether they are related to diagnosis or specifictraits and whether they are also present in unaffectedfamily members remains to be seen

Several additional limitations to and alternative inter-pretations of our findings merit consideration First mightthe hyperactive rACC response to correct antisaccades inASD reflect a less efficient response to conflict ratherthan response monitoring The ACC is thought to play arole in detecting conflict and engaging cognitive controlto resolve it (for review see Carter and van Veen 2007)Antisaccades involve a high degree of conflict as one mustchoose between two competing responses and inhibit theprepotent one Errors give rise to another type of conflictwhen the actual response is compared with the intendedone Thus both correct and erroneous antisaccadesengender conflict While it is not possible to entirely ruleout the conflict explanation several observations argueagainst it If this rACC region were detecting conflict itshould respond to both correct and error trials Insteadcontrols showed a significant response only to errors Inaddition the timing of the haemodynamic response tocorrect antisaccades in ASD which started amp4 s and peakedat amp8 s (Fig 3) argues for a role in response monitoringrather than successful conflict resolution In this paradigmdifferential fMRI activation for correct antisaccade versusprosaccade trials including in dACC presumably reflectingconflict detection response inhibition and saccade genera-tion peaks at amp4 s and is mostly gone by 8 s (Manoachet al 2007b see also Polli et al 2005 for a discussionof ACC regions involved in saccade generation versusresponse monitoring) Finally it is dorsal rather than rACCactivation that is associated with engaging cognitive controlon correct high-conflict and error trials (Carter and vanVeen 2007)

An important limitation is that the small sample size ofthe present study may have compromised our power todetect additional group differences and relations of interestparticularly in regions outside our a priori ACC regions ofinterest for which more stringent multiple comparisonscorrections were applied In spite of the relatively smallsample size our a priori hypotheses were confirmedIn addition four of our ASD participants were takingmedications that could affect ACC function A re-analysisof the fMRI data with only unmedicated ASD participantshowever supported our main findings of significantly lessdiscrimination between response outcomes in the rACCand greater activation to correct trials It is also important tonote that our sample was limited to high functioning adultswith ASD so it is not clear that our findings would generalizeto lower functioning or younger samples Although increasedantisaccade error rates are seen in autism as early as ages 8ndash12(Luna et al 2007) we limited our study to adults sincesaccadic inhibition may not fully develop until late adoles-cence (Klein and Foerster 2001) and larger samples would benecessary to discriminate between the effects of diagnosis andthose due to normal development Another limitationconcerns our interpretation of FA as reflecting connectivityFA reflects the degree of directional coherence of waterdiffusion in tissue and is therefore an indirect measure of

Page 12 of 15 Brain (2008) K NThakkar et al

WMmicrostructure While FA correlates with axon myelinat-ion (Harsan et al 2006) not all of its biophysicaldeterminants are fully understood (Beaulieu 2002) Thus itis not possible to attribute FA differences or the relations weobserved to a particular WM propertyIn summary the present study reports hyperactive

response monitoring in the rACC in ASD that is relatedto restricted repetitive behaviour and is accompanied byreduced FA in the underlying white matter This suggests astructural correlate to abnormal function that may com-promise connectivity in the circuitry subserving responsemonitoring These findings suggest that structural and fun-ctional abnormalities of the ACC compromise responsemonitoring and contribute to behavioural repetition inASD Impairments in evaluating and learning from errorsmay lead to behaviour that is rigid and repetitive ratherthan optimally guided by outcomes and may compromiseperformance across a wide range of tasks Illuminating theneural basis and clinical significance of response monitoringdeficits in ASD represents a fruitful avenue for furtherresearch

Supplementary materialSupplementary material is available at Brain online

AcknowledgementsThe authors wish to thank Mark Vangel for statisticalconsultation Robert Levy Kim Ono Christopher Shermanand Jeremy Young for help with manuscript preparationand Matt Cain and Jay Edelman for technical assistanceSome of these data were presented at the annual meetingof the Society for Neuroscience in Washington DC inNovember 2005 This study was supported by NationalInstitute for Mental Health [R01 MH67720 (DSM)MH72120 (FEP)] Mental Illness Neuroscience Discovery(MIND) Institute (DOE DE-FG02-99ER62764) TheNational Center for Research Resources (P41RR14075)Funding to pay the Open Access publication charges forthis article was provided by the National Institute forMental Health NRSA MH72120

ReferencesAshwin C Baron-Cohen S Wheelwright S OrsquoRiordan M Bullmore ETDifferential activation of the amygdala and the lsquosocial brainrsquo duringfearful face-processing in Asperger syndrome Neuropsychologia 200745 2ndash14

Barnea-Goraly N Kwon H Menon V Eliez S Lotspeich L Reiss ALWhite matter structure in autism preliminary evidence from diffusiontensor imaging Biol Psychiatry 2004 55 323ndash6

Baron-Cohen S Do autistic children have obsessions and compulsionsBr J Clin Psychol 1989 28 (Pt 3) 193ndash200

Barton JJ Greenzang C Hefter R Edelman J Manoach DS Switchingplasticity and prediction in a saccadic task-switch paradigm Exp BrainRes 2006 168 76ndash87

Basser PJ Mattiello J LeBihan D MR diffusion tensor spectroscopy andimaging Biophys J 1994 66 259ndash67

Beaulieu C The basis of anisotropic water diffusion in the nervous system- a technical review NMR Biomed 2002 15 435ndash55

Blair JR Spreen O Predicting premorbid IQ a revision of the NationalAdult Reading Test Clin Neuropsychol 1989 3 129ndash136

Bogte H Flamma B van der Meere J van Engeland H Post-erroradaptation in adults with high functioning autism Neuropsychologia2007 45 1707ndash14

Breiter HC Rauch SL Kwong KK Baker JR Weisskoff RM Kennedy DNet al Functional magnetic resonance imaging of symptom provocationin obsessive-compulsive disorder Arch Gen Psychiatry 1996 53595ndash606

Buckner RL Goodman J Burock M Rotte M Koutstaal W Schacter Det al Functional-anatomic correlates of object priming in humansrevealed by rapid presentation event-related fMRI Neuron 1998 20285ndash96

Burock MA Dale AM Estimation and detection of event-related fMRIsignals with temporally correlated noise a statistically efficient andunbiased approach Hum Brain Mapp 2000 11 249ndash60

Bush G Whalen PJ Rosen BR Jenike MA McInerney SC Rauch SL Thecounting stroop an interference task specialized for functionalneuroimaging- -validation study with functional MRI Hum BrainMapp 1998 6 270ndash82

Carter CS MacDonald AW 3rd Ross LL Stenger VA Anterior cingulatecortex activity and impaired self-monitoring of performance in patientswith schizophrenia an event-related fMRI study Am J Psychiatry 2001158 1423ndash8

Carter CS van Veen V Anterior cingulate cortex and conflict detection anupdate of theory and data Cogn Affect Behav Neurosci 2007 7 367ndash79

Cherkasova MV Manoach DS Intriligator JM Barton JJS Antisaccadesand task-switching interactions in controlled processing Exp Brain Res2002 144 528ndash37

Collins DL Neelin P Peters TM Evans AC Automatic 3D intersubjectregistration of MR volumetric data in standardized Talairach spaceJ Comput Assist Tomogr 1994 18 192ndash205

Courchesne E Pierce K Why the frontal cortex in autism might be talkingonly to itself local over-connectivity but long-distance disconnectionCurr Opin Neurobiol 2005 15 225ndash30

Cox RW Jesmanowicz A Real-time 3D image registration for functionalMRI Magn Reson Med 1999 42 1014ndash8

Dale AM Optimal experimental design for event-related fMRI Hum BrainMapp 1999 8 109ndash40

Dale AM Fischl B Sereno MI Cortical surface-based analysisI Segmentation and surface reconstruction Neuroimage 1999 9 179ndash94

Damasio AR Tranel D Damasio H Individuals with sociopathic behaviorcaused by frontal damage fail to respond autonomically to social stimuliBehav Brain Res 1990 41 81ndash94

Dehaene S Posner MI Tucker DM Localization of a neural system forerror detection and compensation Psychol Sci 1994 5 303ndash5

DeSouza JF Menon RS Everling S Preparatory set associated with pro-saccades and anti-saccades in humans investigated with event-relatedFMRI J Neurophysiol 2003 89 1016ndash23

Devinsky O Morrell MJ Vogt BA Contributions of anterior cingulatecortex to behaviour Brain 1995 118 (Pt 1) 279ndash306

Dichter GS Belger A Social stimuli interfere with cognitive control inautism Neuroimage 2007 35 1219ndash30

Dougherty DD Baer L Cosgrove GR Cassem EH Price BHNierenberg AA et al Prospective long-term follow-up of 44 patientswho received cingulotomy for treatment-refractory obsessive-compulsivedisorder Am J Psychiatry 2002 159 269ndash75

Drevets WC Neuroimaging studies of mood disorders Biol Psychiatry2000 48 813ndash29

Everling S Dorris MC Klein RM Munoz DP Role of primate superiorcolliculus in preparation and execution of anti-saccades and pro-saccades J Neurosci 1999 19 2740ndash54

Everling S Munoz DP Neuronal correlates for preparatory set associatedwith pro-saccades and anti-saccades in the primate frontal eye fieldJ Neurosci 2000 20 387ndash400

Abnormal response monitoring in autism Brain (2008) Page 13 of 15

Falkenstein M Hohnsbein J Hoormann J Event-related correlates of errorsin reaction tasks In Karmos G Molnar M Csepe I Desmedt JE editorsPerspectives of event-related potentials research AmsterdamThe Netherlands Elsevier Science 1995 p 287ndash96

Fallgatter AJ Herrmann MJ Roemmler J Ehlis AC Wagener AHeidrich A et al Allelic variation of serotonin transporter functionmodulates the brain electrical response for error processingNeuropsychopharmacology 2004 29 1506ndash11

Fecteau JH Au C Armstrong IT Munoz DP Sensory biases producealternation advantage found in sequential saccadic eye movement tasksExp Brain Res 2004 159 84ndash91

Fischer B Breitmeyer B Mechanisms of visual attention revealed bysaccadic eye movements Neuropsychologia 1987 25 73ndash83

Fischl B Sereno MI Dale AM Cortical surface-based analysis II Inflationflattening and a surface-based coordinate system Neuroimage 1999a 9195ndash207

Fischl B Sereno MI Tootell RBH Dale AM High-resolution intersubjectaveraging and a coordinate system for the cortical surface Hum BrainMapp 1999b 8 272ndash84

Fischl B van der Kouwe A Destrieux C Halgren E Segonne F Salat DHet al Automatically parcellating the human cerebral cortex CerebCortex 2004 14 11ndash22

Fitzgerald KD Welsh RC Gehring WJ Abelson JL Himle JA Liberzon Iet al Error-related hyperactivity of the anterior cingulate cortex inobsessive-compulsive disorder Biol Psychiatry 2005 57 287ndash94

Garavan H Ross TJ Murphy K Roche RA Stein EA Dissociable executivefunctions in the dynamic control of behavior inhibition errordetection and correction NeuroImage 2002 17 1820ndash9

Gehring WJ Goss B Coles MG Meyer DE Donchin E A neuralsystem for error detection and compensation Psychol Sci 1993 4385ndash90

Gehring WJ Himle J Nisenson LG Action-monitoring dysfunction inobsessive-compulsive disorder Psychol Sci 2000 11 1ndash6

Gehring WJ Knight RT Prefrontal-cingulate interactions in actionmonitoring Nat Neurosci 2000 3 516ndash20

Goldberg MC Lasker AG Zee DS Garth E Tien A Landa RJ Deficits inthe initiation of eye movements in the absence of a visual target inadolescents with high functioning autism Neuropsychologia 2002 402039ndash49

Gomot M Bernard FA Davis MH Belmonte MK Ashwin CBullmore ET et al Change detection in children with autism anauditory event-related fMRI study Neuroimage 2006 29 475ndash84

Hajcak G Simons RF Error-related brain activity in obsessive-compulsiveundergraduates Psychiatry Res 2002 110 63ndash72

Hall GB Szechtman H Nahmias C Enhanced salience and emotionrecognition in autism a PET study Am J Psychiatry 2003 160 1439ndash41

Hallett PE Primary and secondary saccades to goals defined byinstructions Vision Res 1978 18 1279ndash96

Happe F Ronald A Plomin R Time to give up on a single explanation forautism Nat Neurosci 2006 9 1218ndash20

Harsan LA Poulet P Guignard B Steibel J Parizel N de Sousa PL et alBrain dysmyelination and recovery assessment by noninvasive in vivodiffusion tensor magnetic resonance imaging J Neurosci Res 2006 83392ndash402

Haznedar MM Buchsbaum MS Metzger M Solimando A Spiegel-Cohen J Hollander E Anterior cingulate gyrus volume and glucosemetabolism in autistic disorder Am J Psychiatry 1997 154 1047ndash50

Haznedar MM Buchsbaum MS Wei TC Hof PR Cartwright CBienstock CA et al Limbic circuitry in patients with autism spectrumdisorders studied with positron emission tomography and magneticresonance imaging Am J Psychiatry 2000 157 1994ndash2001

Henderson H Schwartz C Mundy P Burnette C Sutton S Zahka N et alResponse monitoring the error-related negativity and differences insocial behavior in autism Brain Cogn 2006 61 96ndash109

Hill EL Executive dysfunction in autism Trends Cogn Sci 2004 8 26ndash32

Hollander E King A Delaney K Smith CJ Silverman JM Obsessive-compulsive behaviors in parents of multiplex autism families PsychiatryRes 2003 117 11ndash6

Hollingshead AB Two factor index of social position New Haven CTYale University Press 1965

Holroyd CB Coles MG The neural basis of human error processingreinforcement learning dopamine and the error-related negativityPsychol Rev 2002 109 679ndash709

Jenkinson M Smith S A global optimisation method for robust affineregistration of brain images Med Image Anal 2001 5 143ndash56

Johannes S Wieringa BM Nager W Rada D Dengler R Emrich HMet al Discrepant target detection and action monitoring in obsessive-compulsive disorder Psychiatry Res 2001 108 101ndash10

Jones DK The effect of gradient sampling schemes on measures derivedfrom diffusion tensor MRI A Monte Carlo study Magn Reson Med2004 51 807ndash15

Kana RK Keller TA Minshew NJ Just MA Inhibitory control in high-functioning autism decreased activation and underconnectivity ininhibition networks Biol Psychiatry 2007 62 198ndash206

Kennedy DP Redcay E Courchesne E Failing to deactivate restingfunctional abnormalities in autism Proc Natl Acad Sci USA 2006 1038275ndash80

Klein C Foerster F Development of prosaccade and antisaccade taskperformance in participants aged 6 to 26 years Psychophysiology 200138 179ndash89

Klein TA Neumann J Reuter M Hennig J von Cramon DYUllsperger M Genetically determined differences in learning fromerrors Science 2007 318 1642ndash5

Kramer UM Cunillera T Camara E Marco-Pallares J Cucurell DNager W et al The impact of catechol-O-methyltransferase anddopamine D4 receptor genotypes on neurophysiological markers ofperformance monitoring J Neurosci 2007 27 14190ndash8

Laurens KR Ngan ET Bates AT Kiehl KA Liddle PF Rostral anteriorcingulate cortex dysfunction during error processing in schizophreniaBrain 2003 126 610ndash22

Levitt JG OrsquoNeill J Blanton RE Smalley S Fadale D McCracken JT et alProton magnetic resonance spectroscopic imaging of the brain inchildhood autism Biol Psychiatry 2003 54 1355ndash66

Leyfer OT Folstein SE Bacalman S Davis NO Dinh E Morgan J et alComorbid psychiatric disorders in children with autism interview devel-opment and rates of disorders J Autism Dev Disord 2006 36 849ndash61

London E The role of the neurobiologist in redefining the diagnosis ofautism Brain Pathol 2007 17 408ndash11

Lopez BR Lincoln AJ Ozonoff S Lai Z Examining the relationshipbetween executive functions and restricted repetitive symptoms ofAutistic Disorder J Autism Dev Disord 2005 35 445ndash60

Lord C Rutter M DiLavore PC Risi S Autism diagnostic observationschedule - WPS (ADOS-WPS) Los Angeles CA Western PsychologicalServices 1999

Luna B Doll SK Hegedus SJ Minshew NJ Sweeney JA Maturation ofexecutive function in autism Biol Psychiatry 2007 61 474ndash81

Luu P Tucker DM Derryberry D Reed M Poulsen C Electrophysiolog-ical responses to errors and feedback in the process of action regulationPsychol Sci 2003 14 47ndash53

Maltby N Tolin DF Worhunsky P OrsquoKeefe TM Kiehl KA Dysfunctionalaction monitoring hyperactivates frontal-striatal circuits in obsessive-compulsive disorder an event-related fMRI study Neuroimage 2005 24495ndash503

Manoach DS Ketwaroo GA Polli FE Thakkar KN Barton JJ Goff DCet al Reduced microstructural integrity of the white matter underlyinganterior cingulate cortex is associated with increased saccadic latency inschizophrenia Neuroimage 2007a 37 599ndash610

Manoach DS Lindgren KA Barton JJ Deficient saccadic inhibition inAspergerrsquos disorder and the social-emotional processing disorderJ Neurol Neurosurg Psychiatry 2004 75 1719ndash26

Page 14 of 15 Brain (2008) K NThakkar et al

Manoach DS Lindgren KA Cherkasova MV Goff DC Halpern EFIntriligator J et al Schizophrenic subjects show deficient inhibition butintact task-switching on saccadic tasks Biol Psychiatry 2002 51 816ndash26

Manoach DS Thakkar KN Cain MS Polli FE Edelman JA Fischl B et alNeural activity is modulated by trial history a functional magneticresonance imaging study of the effects of a previous antisaccadeJ Neurosci 2007b 27 1791ndash8

Manoach DS Weintraub S Daffner KR Scinto LF Deficient antisaccadesin the social-emotional processing disorder Neuroreport 1997 8 901ndash5

McDougle CJ Kresch LE Goodman WK Naylor ST Volkmar FRCohen DJ et al A case-controlled study of repetitive thoughts andbehavior in adults with autistic disorder and obsessive-compulsivedisorder Am J Psychiatry 1995 152 772ndash7

McPeek RM Skavenski AA Nakayama K Concurrent processing ofsaccades in visual search Vision Res 2000 40 2499ndash516

Miezin FM Maccotta L Ollinger JM Petersen SE Buckner RLCharacterizing the hemodynamic response effects of presentation ratesampling procedure and the possibility of ordering brain activity basedon relative timing Neuroimage 2000 11 735ndash59

Minshew NJ Luna B Sweeney JA Oculomotor evidence for neocorticalsystems but not cerebellar dysfunction in autism Neurology 1999 52917ndash22

Mokler A Fischer B The recognition and correction of involuntaryprosaccades in an antisaccade task Exp Brain Res 1999 125 511ndash6

Munoz DP Everling S Look away the anti-saccade task and the voluntarycontrol of eye movement Nat Rev Neurosci 2004 5 218ndash28

Murphy DG Daly E Schmitz N Toal F Murphy K Curran S et alCortical serotonin 5-HT2A receptor binding and social communicationin adults with Aspergerrsquos syndrome an in vivo SPECT study Am JPsychiatry 2006 163 934ndash6

Nieuwenhuis S Nielen MM Mol N Hajcak G Veltman DJ Performancemonitoring in obsessive-compulsive disorder Psychiatry Res 2005 134111ndash22

Nieuwenhuis S Ridderinkhof KR Blom J Band GP Kok A Error-relatedbrain potentials are differentially related to awareness of response errorsevidence from an antisaccade task Psychophysiology 2001 38 752ndash60

Ohnishi T Matsuda H Hashimoto T Kunihiro T Nishikawa M Uema Tet al Abnormal regional cerebral blood flow in childhood autism Brain2000 123 (Pt 9) 1838ndash44

Oldfield RC The assessment and analysis of handedness The EdinburghInventory Neuropsychologia 1971 9 97ndash113

Paus T Primate anterior cingulate cortex where motor control drive andcognition interface Nat Rev Neurosci 2001 2 417ndash24

Pfefferbaum A Sullivan EV Hedehus M Lim KO Adalsteinsson EMoseley M Age-related decline in brain white matter anisotropymeasured with spatially corrected echo-planar diffusion tensor imagingMagn Reson Med 2000 44 259ndash68

Pitman RK A cybernetic model of obsessive-compulsive psychopathologyCompr Psychiatry 1987 28 334ndash43

Polli FE Barton JJ Cain MS Thakkar KN Rauch SL Manoach DSRostral and dorsal anterior cingulate cortex make dissociable contribu-tions during antisaccade error commission Proc Natl Acad Sci USA2005 102 15700ndash5

Polli FE Barton JJ Thakkar KN Greve DN Goff DC Rauch SL et alReduced error-related activation in two anterior cingulate circuits is relatedto impaired performance in schizophrenia Brain 2008 131 971ndash86

Polli FE Barton JJ Vangel M Goff DC Iguchi L Manoach DSSchizophrenia patients show intact immediate error-related performanceadjustments on an antisaccade task Schizophr Res 2006 82 191ndash201

Rabbitt PM Errors and error correction in choice-response tasks J ExpPsychol 1966 71 264ndash72

Radant AD Dobie DJ Calkins ME Olincy A Braff DL Cadenhead KSet al Successful multi-site measurement of antisaccade performancedeficits in schizophrenia Schizophr Res 2007 89 320ndash9

Reese TG Heid O Weisskoff RM Wedeen VJ Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echoMagn Reson Med 2003 49 177ndash82

Russell J Jarrold C Error-correction problems in autism evidence for amonitoring impairment J Autism Dev Disord 1998 28 177ndash88

Russell AJ Mataix-Cols D Anson M Murphy DG Obsessions andcompulsions in Asperger syndrome and high-functioning autism Br JPsychiatry 2005 186 525ndash8

Rutter M Le Couteur A Lord C Autism diagnostic interview-revised LosAngeles CA Western Psychological Services 2003

Santesso DL Segalowitz SJ Schmidt LA Error-related electrocorticalresponses are enhanced in children with obsessive-compulsive behaviorsDev Neuropsychol 2006 29 431ndash45

Shafritz KM Dichter GS Baranek GT Belger A The neural circuitrymediating shifts in behavioral response and cognitive set in autism BiolPsychiatry 2007 63 974ndash80

Silk TJ Rinehart N Bradshaw JL Tonge B Egan G OrsquoBoyle MW et alVisuospatial processing and the function of prefrontal-parietal networksin autism spectrum disorders a functional MRI study Am J Psychiatry2006 163 1440ndash3

South M Ozonoff S McMahon WM The relationship between executivefunctioning central coherence and repetitive behaviors in the high-functioning autism spectrum Autism 2007 11 437ndash51

Taylor SF Martis B Fitzgerald KD Welsh RC Abelson JL Liberzon Iet al Medial frontal cortex activity and loss-related responses to errorsJ Neurosci 2006 26 4063ndash70

Taylor SF Stern ER Gehring WJ Neural systems for error monitoringrecent findings and theoretical perspectives Neuroscientist 2007 13160ndash72

Thesen S Heid O Mueller E Schad LR Prospective acquisition correctionfor head motion with image-based tracking for real-time fMRI MagnReson Med 2000 44 457ndash65

Turner M Annotation Repetitive behaviour in autism areview of psychological research J Child Psychol Psychiatry 1999 40839ndash49

Ullsperger M von Cramon DY Neuroimaging of performance monitoringerror detection and beyond Cortex 2004 40 593ndash604

Ursu S Stenger VA Shear MK Jones MR Carter CS Overactiveaction monitoring in obsessive-compulsive disorder evidencefrom functional magnetic resonance imaging Psychol Sci 2003 14347ndash53

Meel CS Heslenfeld DJ Oosterlaan J Sergeant JA Adaptive controldeficits in attention-deficithyperactivity disorder (ADHD) the role oferror processing Psychiatry Res 2007 151 211ndash20

van Veen V Carter CS The timing of action-monitoring processes in theanterior cingulate cortex J Cogn Neurosci 2002 14 593ndash602

Wechsler D Wechsler Abbreviated Scale of intelligence San Antonio TXThe Psychological Corporation 1999

Whalen PJ Bush G McNally RJ Wilhelm S McInerney SC Jenike MAet al The emotional counting Stroop paradigm a functional magneticresonance imaging probe of the anterior cingulate affective division BiolPsychiatry 1998 44 1219ndash28

White K Ashton R Handedness assessment inventory Neuropsychologia1976 14 261ndash4

Zandt F Prior M Kyrios M Repetitive behaviour in children with highfunctioning autism and obsessive compulsive disorder J Autism DevDisord 2007 37 251ndash9

Abnormal response monitoring in autism Brain (2008) Page 15 of 15

WMmicrostructure While FA correlates with axon myelinat-ion (Harsan et al 2006) not all of its biophysicaldeterminants are fully understood (Beaulieu 2002) Thus itis not possible to attribute FA differences or the relations weobserved to a particular WM propertyIn summary the present study reports hyperactive

response monitoring in the rACC in ASD that is relatedto restricted repetitive behaviour and is accompanied byreduced FA in the underlying white matter This suggests astructural correlate to abnormal function that may com-promise connectivity in the circuitry subserving responsemonitoring These findings suggest that structural and fun-ctional abnormalities of the ACC compromise responsemonitoring and contribute to behavioural repetition inASD Impairments in evaluating and learning from errorsmay lead to behaviour that is rigid and repetitive ratherthan optimally guided by outcomes and may compromiseperformance across a wide range of tasks Illuminating theneural basis and clinical significance of response monitoringdeficits in ASD represents a fruitful avenue for furtherresearch

Supplementary materialSupplementary material is available at Brain online

AcknowledgementsThe authors wish to thank Mark Vangel for statisticalconsultation Robert Levy Kim Ono Christopher Shermanand Jeremy Young for help with manuscript preparationand Matt Cain and Jay Edelman for technical assistanceSome of these data were presented at the annual meetingof the Society for Neuroscience in Washington DC inNovember 2005 This study was supported by NationalInstitute for Mental Health [R01 MH67720 (DSM)MH72120 (FEP)] Mental Illness Neuroscience Discovery(MIND) Institute (DOE DE-FG02-99ER62764) TheNational Center for Research Resources (P41RR14075)Funding to pay the Open Access publication charges forthis article was provided by the National Institute forMental Health NRSA MH72120

ReferencesAshwin C Baron-Cohen S Wheelwright S OrsquoRiordan M Bullmore ETDifferential activation of the amygdala and the lsquosocial brainrsquo duringfearful face-processing in Asperger syndrome Neuropsychologia 200745 2ndash14

Barnea-Goraly N Kwon H Menon V Eliez S Lotspeich L Reiss ALWhite matter structure in autism preliminary evidence from diffusiontensor imaging Biol Psychiatry 2004 55 323ndash6

Baron-Cohen S Do autistic children have obsessions and compulsionsBr J Clin Psychol 1989 28 (Pt 3) 193ndash200

Barton JJ Greenzang C Hefter R Edelman J Manoach DS Switchingplasticity and prediction in a saccadic task-switch paradigm Exp BrainRes 2006 168 76ndash87

Basser PJ Mattiello J LeBihan D MR diffusion tensor spectroscopy andimaging Biophys J 1994 66 259ndash67

Beaulieu C The basis of anisotropic water diffusion in the nervous system- a technical review NMR Biomed 2002 15 435ndash55

Blair JR Spreen O Predicting premorbid IQ a revision of the NationalAdult Reading Test Clin Neuropsychol 1989 3 129ndash136

Bogte H Flamma B van der Meere J van Engeland H Post-erroradaptation in adults with high functioning autism Neuropsychologia2007 45 1707ndash14

Breiter HC Rauch SL Kwong KK Baker JR Weisskoff RM Kennedy DNet al Functional magnetic resonance imaging of symptom provocationin obsessive-compulsive disorder Arch Gen Psychiatry 1996 53595ndash606

Buckner RL Goodman J Burock M Rotte M Koutstaal W Schacter Det al Functional-anatomic correlates of object priming in humansrevealed by rapid presentation event-related fMRI Neuron 1998 20285ndash96

Burock MA Dale AM Estimation and detection of event-related fMRIsignals with temporally correlated noise a statistically efficient andunbiased approach Hum Brain Mapp 2000 11 249ndash60

Bush G Whalen PJ Rosen BR Jenike MA McInerney SC Rauch SL Thecounting stroop an interference task specialized for functionalneuroimaging- -validation study with functional MRI Hum BrainMapp 1998 6 270ndash82

Carter CS MacDonald AW 3rd Ross LL Stenger VA Anterior cingulatecortex activity and impaired self-monitoring of performance in patientswith schizophrenia an event-related fMRI study Am J Psychiatry 2001158 1423ndash8

Carter CS van Veen V Anterior cingulate cortex and conflict detection anupdate of theory and data Cogn Affect Behav Neurosci 2007 7 367ndash79

Cherkasova MV Manoach DS Intriligator JM Barton JJS Antisaccadesand task-switching interactions in controlled processing Exp Brain Res2002 144 528ndash37

Collins DL Neelin P Peters TM Evans AC Automatic 3D intersubjectregistration of MR volumetric data in standardized Talairach spaceJ Comput Assist Tomogr 1994 18 192ndash205

Courchesne E Pierce K Why the frontal cortex in autism might be talkingonly to itself local over-connectivity but long-distance disconnectionCurr Opin Neurobiol 2005 15 225ndash30

Cox RW Jesmanowicz A Real-time 3D image registration for functionalMRI Magn Reson Med 1999 42 1014ndash8

Dale AM Optimal experimental design for event-related fMRI Hum BrainMapp 1999 8 109ndash40

Dale AM Fischl B Sereno MI Cortical surface-based analysisI Segmentation and surface reconstruction Neuroimage 1999 9 179ndash94

Damasio AR Tranel D Damasio H Individuals with sociopathic behaviorcaused by frontal damage fail to respond autonomically to social stimuliBehav Brain Res 1990 41 81ndash94

Dehaene S Posner MI Tucker DM Localization of a neural system forerror detection and compensation Psychol Sci 1994 5 303ndash5

DeSouza JF Menon RS Everling S Preparatory set associated with pro-saccades and anti-saccades in humans investigated with event-relatedFMRI J Neurophysiol 2003 89 1016ndash23

Devinsky O Morrell MJ Vogt BA Contributions of anterior cingulatecortex to behaviour Brain 1995 118 (Pt 1) 279ndash306

Dichter GS Belger A Social stimuli interfere with cognitive control inautism Neuroimage 2007 35 1219ndash30

Dougherty DD Baer L Cosgrove GR Cassem EH Price BHNierenberg AA et al Prospective long-term follow-up of 44 patientswho received cingulotomy for treatment-refractory obsessive-compulsivedisorder Am J Psychiatry 2002 159 269ndash75

Drevets WC Neuroimaging studies of mood disorders Biol Psychiatry2000 48 813ndash29

Everling S Dorris MC Klein RM Munoz DP Role of primate superiorcolliculus in preparation and execution of anti-saccades and pro-saccades J Neurosci 1999 19 2740ndash54

Everling S Munoz DP Neuronal correlates for preparatory set associatedwith pro-saccades and anti-saccades in the primate frontal eye fieldJ Neurosci 2000 20 387ndash400

Abnormal response monitoring in autism Brain (2008) Page 13 of 15

Falkenstein M Hohnsbein J Hoormann J Event-related correlates of errorsin reaction tasks In Karmos G Molnar M Csepe I Desmedt JE editorsPerspectives of event-related potentials research AmsterdamThe Netherlands Elsevier Science 1995 p 287ndash96

Fallgatter AJ Herrmann MJ Roemmler J Ehlis AC Wagener AHeidrich A et al Allelic variation of serotonin transporter functionmodulates the brain electrical response for error processingNeuropsychopharmacology 2004 29 1506ndash11

Fecteau JH Au C Armstrong IT Munoz DP Sensory biases producealternation advantage found in sequential saccadic eye movement tasksExp Brain Res 2004 159 84ndash91

Fischer B Breitmeyer B Mechanisms of visual attention revealed bysaccadic eye movements Neuropsychologia 1987 25 73ndash83

Fischl B Sereno MI Dale AM Cortical surface-based analysis II Inflationflattening and a surface-based coordinate system Neuroimage 1999a 9195ndash207

Fischl B Sereno MI Tootell RBH Dale AM High-resolution intersubjectaveraging and a coordinate system for the cortical surface Hum BrainMapp 1999b 8 272ndash84

Fischl B van der Kouwe A Destrieux C Halgren E Segonne F Salat DHet al Automatically parcellating the human cerebral cortex CerebCortex 2004 14 11ndash22

Fitzgerald KD Welsh RC Gehring WJ Abelson JL Himle JA Liberzon Iet al Error-related hyperactivity of the anterior cingulate cortex inobsessive-compulsive disorder Biol Psychiatry 2005 57 287ndash94

Garavan H Ross TJ Murphy K Roche RA Stein EA Dissociable executivefunctions in the dynamic control of behavior inhibition errordetection and correction NeuroImage 2002 17 1820ndash9

Gehring WJ Goss B Coles MG Meyer DE Donchin E A neuralsystem for error detection and compensation Psychol Sci 1993 4385ndash90

Gehring WJ Himle J Nisenson LG Action-monitoring dysfunction inobsessive-compulsive disorder Psychol Sci 2000 11 1ndash6

Gehring WJ Knight RT Prefrontal-cingulate interactions in actionmonitoring Nat Neurosci 2000 3 516ndash20

Goldberg MC Lasker AG Zee DS Garth E Tien A Landa RJ Deficits inthe initiation of eye movements in the absence of a visual target inadolescents with high functioning autism Neuropsychologia 2002 402039ndash49

Gomot M Bernard FA Davis MH Belmonte MK Ashwin CBullmore ET et al Change detection in children with autism anauditory event-related fMRI study Neuroimage 2006 29 475ndash84

Hajcak G Simons RF Error-related brain activity in obsessive-compulsiveundergraduates Psychiatry Res 2002 110 63ndash72

Hall GB Szechtman H Nahmias C Enhanced salience and emotionrecognition in autism a PET study Am J Psychiatry 2003 160 1439ndash41

Hallett PE Primary and secondary saccades to goals defined byinstructions Vision Res 1978 18 1279ndash96

Happe F Ronald A Plomin R Time to give up on a single explanation forautism Nat Neurosci 2006 9 1218ndash20

Harsan LA Poulet P Guignard B Steibel J Parizel N de Sousa PL et alBrain dysmyelination and recovery assessment by noninvasive in vivodiffusion tensor magnetic resonance imaging J Neurosci Res 2006 83392ndash402

Haznedar MM Buchsbaum MS Metzger M Solimando A Spiegel-Cohen J Hollander E Anterior cingulate gyrus volume and glucosemetabolism in autistic disorder Am J Psychiatry 1997 154 1047ndash50

Haznedar MM Buchsbaum MS Wei TC Hof PR Cartwright CBienstock CA et al Limbic circuitry in patients with autism spectrumdisorders studied with positron emission tomography and magneticresonance imaging Am J Psychiatry 2000 157 1994ndash2001

Henderson H Schwartz C Mundy P Burnette C Sutton S Zahka N et alResponse monitoring the error-related negativity and differences insocial behavior in autism Brain Cogn 2006 61 96ndash109

Hill EL Executive dysfunction in autism Trends Cogn Sci 2004 8 26ndash32

Hollander E King A Delaney K Smith CJ Silverman JM Obsessive-compulsive behaviors in parents of multiplex autism families PsychiatryRes 2003 117 11ndash6

Hollingshead AB Two factor index of social position New Haven CTYale University Press 1965

Holroyd CB Coles MG The neural basis of human error processingreinforcement learning dopamine and the error-related negativityPsychol Rev 2002 109 679ndash709

Jenkinson M Smith S A global optimisation method for robust affineregistration of brain images Med Image Anal 2001 5 143ndash56

Johannes S Wieringa BM Nager W Rada D Dengler R Emrich HMet al Discrepant target detection and action monitoring in obsessive-compulsive disorder Psychiatry Res 2001 108 101ndash10

Jones DK The effect of gradient sampling schemes on measures derivedfrom diffusion tensor MRI A Monte Carlo study Magn Reson Med2004 51 807ndash15

Kana RK Keller TA Minshew NJ Just MA Inhibitory control in high-functioning autism decreased activation and underconnectivity ininhibition networks Biol Psychiatry 2007 62 198ndash206

Kennedy DP Redcay E Courchesne E Failing to deactivate restingfunctional abnormalities in autism Proc Natl Acad Sci USA 2006 1038275ndash80

Klein C Foerster F Development of prosaccade and antisaccade taskperformance in participants aged 6 to 26 years Psychophysiology 200138 179ndash89

Klein TA Neumann J Reuter M Hennig J von Cramon DYUllsperger M Genetically determined differences in learning fromerrors Science 2007 318 1642ndash5

Kramer UM Cunillera T Camara E Marco-Pallares J Cucurell DNager W et al The impact of catechol-O-methyltransferase anddopamine D4 receptor genotypes on neurophysiological markers ofperformance monitoring J Neurosci 2007 27 14190ndash8

Laurens KR Ngan ET Bates AT Kiehl KA Liddle PF Rostral anteriorcingulate cortex dysfunction during error processing in schizophreniaBrain 2003 126 610ndash22

Levitt JG OrsquoNeill J Blanton RE Smalley S Fadale D McCracken JT et alProton magnetic resonance spectroscopic imaging of the brain inchildhood autism Biol Psychiatry 2003 54 1355ndash66

Leyfer OT Folstein SE Bacalman S Davis NO Dinh E Morgan J et alComorbid psychiatric disorders in children with autism interview devel-opment and rates of disorders J Autism Dev Disord 2006 36 849ndash61

London E The role of the neurobiologist in redefining the diagnosis ofautism Brain Pathol 2007 17 408ndash11

Lopez BR Lincoln AJ Ozonoff S Lai Z Examining the relationshipbetween executive functions and restricted repetitive symptoms ofAutistic Disorder J Autism Dev Disord 2005 35 445ndash60

Lord C Rutter M DiLavore PC Risi S Autism diagnostic observationschedule - WPS (ADOS-WPS) Los Angeles CA Western PsychologicalServices 1999

Luna B Doll SK Hegedus SJ Minshew NJ Sweeney JA Maturation ofexecutive function in autism Biol Psychiatry 2007 61 474ndash81

Luu P Tucker DM Derryberry D Reed M Poulsen C Electrophysiolog-ical responses to errors and feedback in the process of action regulationPsychol Sci 2003 14 47ndash53

Maltby N Tolin DF Worhunsky P OrsquoKeefe TM Kiehl KA Dysfunctionalaction monitoring hyperactivates frontal-striatal circuits in obsessive-compulsive disorder an event-related fMRI study Neuroimage 2005 24495ndash503

Manoach DS Ketwaroo GA Polli FE Thakkar KN Barton JJ Goff DCet al Reduced microstructural integrity of the white matter underlyinganterior cingulate cortex is associated with increased saccadic latency inschizophrenia Neuroimage 2007a 37 599ndash610

Manoach DS Lindgren KA Barton JJ Deficient saccadic inhibition inAspergerrsquos disorder and the social-emotional processing disorderJ Neurol Neurosurg Psychiatry 2004 75 1719ndash26

Page 14 of 15 Brain (2008) K NThakkar et al

Manoach DS Lindgren KA Cherkasova MV Goff DC Halpern EFIntriligator J et al Schizophrenic subjects show deficient inhibition butintact task-switching on saccadic tasks Biol Psychiatry 2002 51 816ndash26

Manoach DS Thakkar KN Cain MS Polli FE Edelman JA Fischl B et alNeural activity is modulated by trial history a functional magneticresonance imaging study of the effects of a previous antisaccadeJ Neurosci 2007b 27 1791ndash8

Manoach DS Weintraub S Daffner KR Scinto LF Deficient antisaccadesin the social-emotional processing disorder Neuroreport 1997 8 901ndash5

McDougle CJ Kresch LE Goodman WK Naylor ST Volkmar FRCohen DJ et al A case-controlled study of repetitive thoughts andbehavior in adults with autistic disorder and obsessive-compulsivedisorder Am J Psychiatry 1995 152 772ndash7

McPeek RM Skavenski AA Nakayama K Concurrent processing ofsaccades in visual search Vision Res 2000 40 2499ndash516

Miezin FM Maccotta L Ollinger JM Petersen SE Buckner RLCharacterizing the hemodynamic response effects of presentation ratesampling procedure and the possibility of ordering brain activity basedon relative timing Neuroimage 2000 11 735ndash59

Minshew NJ Luna B Sweeney JA Oculomotor evidence for neocorticalsystems but not cerebellar dysfunction in autism Neurology 1999 52917ndash22

Mokler A Fischer B The recognition and correction of involuntaryprosaccades in an antisaccade task Exp Brain Res 1999 125 511ndash6

Munoz DP Everling S Look away the anti-saccade task and the voluntarycontrol of eye movement Nat Rev Neurosci 2004 5 218ndash28

Murphy DG Daly E Schmitz N Toal F Murphy K Curran S et alCortical serotonin 5-HT2A receptor binding and social communicationin adults with Aspergerrsquos syndrome an in vivo SPECT study Am JPsychiatry 2006 163 934ndash6

Nieuwenhuis S Nielen MM Mol N Hajcak G Veltman DJ Performancemonitoring in obsessive-compulsive disorder Psychiatry Res 2005 134111ndash22

Nieuwenhuis S Ridderinkhof KR Blom J Band GP Kok A Error-relatedbrain potentials are differentially related to awareness of response errorsevidence from an antisaccade task Psychophysiology 2001 38 752ndash60

Ohnishi T Matsuda H Hashimoto T Kunihiro T Nishikawa M Uema Tet al Abnormal regional cerebral blood flow in childhood autism Brain2000 123 (Pt 9) 1838ndash44

Oldfield RC The assessment and analysis of handedness The EdinburghInventory Neuropsychologia 1971 9 97ndash113

Paus T Primate anterior cingulate cortex where motor control drive andcognition interface Nat Rev Neurosci 2001 2 417ndash24

Pfefferbaum A Sullivan EV Hedehus M Lim KO Adalsteinsson EMoseley M Age-related decline in brain white matter anisotropymeasured with spatially corrected echo-planar diffusion tensor imagingMagn Reson Med 2000 44 259ndash68

Pitman RK A cybernetic model of obsessive-compulsive psychopathologyCompr Psychiatry 1987 28 334ndash43

Polli FE Barton JJ Cain MS Thakkar KN Rauch SL Manoach DSRostral and dorsal anterior cingulate cortex make dissociable contribu-tions during antisaccade error commission Proc Natl Acad Sci USA2005 102 15700ndash5

Polli FE Barton JJ Thakkar KN Greve DN Goff DC Rauch SL et alReduced error-related activation in two anterior cingulate circuits is relatedto impaired performance in schizophrenia Brain 2008 131 971ndash86

Polli FE Barton JJ Vangel M Goff DC Iguchi L Manoach DSSchizophrenia patients show intact immediate error-related performanceadjustments on an antisaccade task Schizophr Res 2006 82 191ndash201

Rabbitt PM Errors and error correction in choice-response tasks J ExpPsychol 1966 71 264ndash72

Radant AD Dobie DJ Calkins ME Olincy A Braff DL Cadenhead KSet al Successful multi-site measurement of antisaccade performancedeficits in schizophrenia Schizophr Res 2007 89 320ndash9

Reese TG Heid O Weisskoff RM Wedeen VJ Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echoMagn Reson Med 2003 49 177ndash82

Russell J Jarrold C Error-correction problems in autism evidence for amonitoring impairment J Autism Dev Disord 1998 28 177ndash88

Russell AJ Mataix-Cols D Anson M Murphy DG Obsessions andcompulsions in Asperger syndrome and high-functioning autism Br JPsychiatry 2005 186 525ndash8

Rutter M Le Couteur A Lord C Autism diagnostic interview-revised LosAngeles CA Western Psychological Services 2003

Santesso DL Segalowitz SJ Schmidt LA Error-related electrocorticalresponses are enhanced in children with obsessive-compulsive behaviorsDev Neuropsychol 2006 29 431ndash45

Shafritz KM Dichter GS Baranek GT Belger A The neural circuitrymediating shifts in behavioral response and cognitive set in autism BiolPsychiatry 2007 63 974ndash80

Silk TJ Rinehart N Bradshaw JL Tonge B Egan G OrsquoBoyle MW et alVisuospatial processing and the function of prefrontal-parietal networksin autism spectrum disorders a functional MRI study Am J Psychiatry2006 163 1440ndash3

South M Ozonoff S McMahon WM The relationship between executivefunctioning central coherence and repetitive behaviors in the high-functioning autism spectrum Autism 2007 11 437ndash51

Taylor SF Martis B Fitzgerald KD Welsh RC Abelson JL Liberzon Iet al Medial frontal cortex activity and loss-related responses to errorsJ Neurosci 2006 26 4063ndash70

Taylor SF Stern ER Gehring WJ Neural systems for error monitoringrecent findings and theoretical perspectives Neuroscientist 2007 13160ndash72

Thesen S Heid O Mueller E Schad LR Prospective acquisition correctionfor head motion with image-based tracking for real-time fMRI MagnReson Med 2000 44 457ndash65

Turner M Annotation Repetitive behaviour in autism areview of psychological research J Child Psychol Psychiatry 1999 40839ndash49

Ullsperger M von Cramon DY Neuroimaging of performance monitoringerror detection and beyond Cortex 2004 40 593ndash604

Ursu S Stenger VA Shear MK Jones MR Carter CS Overactiveaction monitoring in obsessive-compulsive disorder evidencefrom functional magnetic resonance imaging Psychol Sci 2003 14347ndash53

Meel CS Heslenfeld DJ Oosterlaan J Sergeant JA Adaptive controldeficits in attention-deficithyperactivity disorder (ADHD) the role oferror processing Psychiatry Res 2007 151 211ndash20

van Veen V Carter CS The timing of action-monitoring processes in theanterior cingulate cortex J Cogn Neurosci 2002 14 593ndash602

Wechsler D Wechsler Abbreviated Scale of intelligence San Antonio TXThe Psychological Corporation 1999

Whalen PJ Bush G McNally RJ Wilhelm S McInerney SC Jenike MAet al The emotional counting Stroop paradigm a functional magneticresonance imaging probe of the anterior cingulate affective division BiolPsychiatry 1998 44 1219ndash28

White K Ashton R Handedness assessment inventory Neuropsychologia1976 14 261ndash4

Zandt F Prior M Kyrios M Repetitive behaviour in children with highfunctioning autism and obsessive compulsive disorder J Autism DevDisord 2007 37 251ndash9

Abnormal response monitoring in autism Brain (2008) Page 15 of 15

Falkenstein M Hohnsbein J Hoormann J Event-related correlates of errorsin reaction tasks In Karmos G Molnar M Csepe I Desmedt JE editorsPerspectives of event-related potentials research AmsterdamThe Netherlands Elsevier Science 1995 p 287ndash96

Fallgatter AJ Herrmann MJ Roemmler J Ehlis AC Wagener AHeidrich A et al Allelic variation of serotonin transporter functionmodulates the brain electrical response for error processingNeuropsychopharmacology 2004 29 1506ndash11

Fecteau JH Au C Armstrong IT Munoz DP Sensory biases producealternation advantage found in sequential saccadic eye movement tasksExp Brain Res 2004 159 84ndash91

Fischer B Breitmeyer B Mechanisms of visual attention revealed bysaccadic eye movements Neuropsychologia 1987 25 73ndash83

Fischl B Sereno MI Dale AM Cortical surface-based analysis II Inflationflattening and a surface-based coordinate system Neuroimage 1999a 9195ndash207

Fischl B Sereno MI Tootell RBH Dale AM High-resolution intersubjectaveraging and a coordinate system for the cortical surface Hum BrainMapp 1999b 8 272ndash84

Fischl B van der Kouwe A Destrieux C Halgren E Segonne F Salat DHet al Automatically parcellating the human cerebral cortex CerebCortex 2004 14 11ndash22

Fitzgerald KD Welsh RC Gehring WJ Abelson JL Himle JA Liberzon Iet al Error-related hyperactivity of the anterior cingulate cortex inobsessive-compulsive disorder Biol Psychiatry 2005 57 287ndash94

Garavan H Ross TJ Murphy K Roche RA Stein EA Dissociable executivefunctions in the dynamic control of behavior inhibition errordetection and correction NeuroImage 2002 17 1820ndash9

Gehring WJ Goss B Coles MG Meyer DE Donchin E A neuralsystem for error detection and compensation Psychol Sci 1993 4385ndash90

Gehring WJ Himle J Nisenson LG Action-monitoring dysfunction inobsessive-compulsive disorder Psychol Sci 2000 11 1ndash6

Gehring WJ Knight RT Prefrontal-cingulate interactions in actionmonitoring Nat Neurosci 2000 3 516ndash20

Goldberg MC Lasker AG Zee DS Garth E Tien A Landa RJ Deficits inthe initiation of eye movements in the absence of a visual target inadolescents with high functioning autism Neuropsychologia 2002 402039ndash49

Gomot M Bernard FA Davis MH Belmonte MK Ashwin CBullmore ET et al Change detection in children with autism anauditory event-related fMRI study Neuroimage 2006 29 475ndash84

Hajcak G Simons RF Error-related brain activity in obsessive-compulsiveundergraduates Psychiatry Res 2002 110 63ndash72

Hall GB Szechtman H Nahmias C Enhanced salience and emotionrecognition in autism a PET study Am J Psychiatry 2003 160 1439ndash41

Hallett PE Primary and secondary saccades to goals defined byinstructions Vision Res 1978 18 1279ndash96

Happe F Ronald A Plomin R Time to give up on a single explanation forautism Nat Neurosci 2006 9 1218ndash20

Harsan LA Poulet P Guignard B Steibel J Parizel N de Sousa PL et alBrain dysmyelination and recovery assessment by noninvasive in vivodiffusion tensor magnetic resonance imaging J Neurosci Res 2006 83392ndash402

Haznedar MM Buchsbaum MS Metzger M Solimando A Spiegel-Cohen J Hollander E Anterior cingulate gyrus volume and glucosemetabolism in autistic disorder Am J Psychiatry 1997 154 1047ndash50

Haznedar MM Buchsbaum MS Wei TC Hof PR Cartwright CBienstock CA et al Limbic circuitry in patients with autism spectrumdisorders studied with positron emission tomography and magneticresonance imaging Am J Psychiatry 2000 157 1994ndash2001

Henderson H Schwartz C Mundy P Burnette C Sutton S Zahka N et alResponse monitoring the error-related negativity and differences insocial behavior in autism Brain Cogn 2006 61 96ndash109

Hill EL Executive dysfunction in autism Trends Cogn Sci 2004 8 26ndash32

Hollander E King A Delaney K Smith CJ Silverman JM Obsessive-compulsive behaviors in parents of multiplex autism families PsychiatryRes 2003 117 11ndash6

Hollingshead AB Two factor index of social position New Haven CTYale University Press 1965

Holroyd CB Coles MG The neural basis of human error processingreinforcement learning dopamine and the error-related negativityPsychol Rev 2002 109 679ndash709

Jenkinson M Smith S A global optimisation method for robust affineregistration of brain images Med Image Anal 2001 5 143ndash56

Johannes S Wieringa BM Nager W Rada D Dengler R Emrich HMet al Discrepant target detection and action monitoring in obsessive-compulsive disorder Psychiatry Res 2001 108 101ndash10

Jones DK The effect of gradient sampling schemes on measures derivedfrom diffusion tensor MRI A Monte Carlo study Magn Reson Med2004 51 807ndash15

Kana RK Keller TA Minshew NJ Just MA Inhibitory control in high-functioning autism decreased activation and underconnectivity ininhibition networks Biol Psychiatry 2007 62 198ndash206

Kennedy DP Redcay E Courchesne E Failing to deactivate restingfunctional abnormalities in autism Proc Natl Acad Sci USA 2006 1038275ndash80

Klein C Foerster F Development of prosaccade and antisaccade taskperformance in participants aged 6 to 26 years Psychophysiology 200138 179ndash89

Klein TA Neumann J Reuter M Hennig J von Cramon DYUllsperger M Genetically determined differences in learning fromerrors Science 2007 318 1642ndash5

Kramer UM Cunillera T Camara E Marco-Pallares J Cucurell DNager W et al The impact of catechol-O-methyltransferase anddopamine D4 receptor genotypes on neurophysiological markers ofperformance monitoring J Neurosci 2007 27 14190ndash8

Laurens KR Ngan ET Bates AT Kiehl KA Liddle PF Rostral anteriorcingulate cortex dysfunction during error processing in schizophreniaBrain 2003 126 610ndash22

Levitt JG OrsquoNeill J Blanton RE Smalley S Fadale D McCracken JT et alProton magnetic resonance spectroscopic imaging of the brain inchildhood autism Biol Psychiatry 2003 54 1355ndash66

Leyfer OT Folstein SE Bacalman S Davis NO Dinh E Morgan J et alComorbid psychiatric disorders in children with autism interview devel-opment and rates of disorders J Autism Dev Disord 2006 36 849ndash61

London E The role of the neurobiologist in redefining the diagnosis ofautism Brain Pathol 2007 17 408ndash11

Lopez BR Lincoln AJ Ozonoff S Lai Z Examining the relationshipbetween executive functions and restricted repetitive symptoms ofAutistic Disorder J Autism Dev Disord 2005 35 445ndash60

Lord C Rutter M DiLavore PC Risi S Autism diagnostic observationschedule - WPS (ADOS-WPS) Los Angeles CA Western PsychologicalServices 1999

Luna B Doll SK Hegedus SJ Minshew NJ Sweeney JA Maturation ofexecutive function in autism Biol Psychiatry 2007 61 474ndash81

Luu P Tucker DM Derryberry D Reed M Poulsen C Electrophysiolog-ical responses to errors and feedback in the process of action regulationPsychol Sci 2003 14 47ndash53

Maltby N Tolin DF Worhunsky P OrsquoKeefe TM Kiehl KA Dysfunctionalaction monitoring hyperactivates frontal-striatal circuits in obsessive-compulsive disorder an event-related fMRI study Neuroimage 2005 24495ndash503

Manoach DS Ketwaroo GA Polli FE Thakkar KN Barton JJ Goff DCet al Reduced microstructural integrity of the white matter underlyinganterior cingulate cortex is associated with increased saccadic latency inschizophrenia Neuroimage 2007a 37 599ndash610

Manoach DS Lindgren KA Barton JJ Deficient saccadic inhibition inAspergerrsquos disorder and the social-emotional processing disorderJ Neurol Neurosurg Psychiatry 2004 75 1719ndash26

Page 14 of 15 Brain (2008) K NThakkar et al

Manoach DS Lindgren KA Cherkasova MV Goff DC Halpern EFIntriligator J et al Schizophrenic subjects show deficient inhibition butintact task-switching on saccadic tasks Biol Psychiatry 2002 51 816ndash26

Manoach DS Thakkar KN Cain MS Polli FE Edelman JA Fischl B et alNeural activity is modulated by trial history a functional magneticresonance imaging study of the effects of a previous antisaccadeJ Neurosci 2007b 27 1791ndash8

Manoach DS Weintraub S Daffner KR Scinto LF Deficient antisaccadesin the social-emotional processing disorder Neuroreport 1997 8 901ndash5

McDougle CJ Kresch LE Goodman WK Naylor ST Volkmar FRCohen DJ et al A case-controlled study of repetitive thoughts andbehavior in adults with autistic disorder and obsessive-compulsivedisorder Am J Psychiatry 1995 152 772ndash7

McPeek RM Skavenski AA Nakayama K Concurrent processing ofsaccades in visual search Vision Res 2000 40 2499ndash516

Miezin FM Maccotta L Ollinger JM Petersen SE Buckner RLCharacterizing the hemodynamic response effects of presentation ratesampling procedure and the possibility of ordering brain activity basedon relative timing Neuroimage 2000 11 735ndash59

Minshew NJ Luna B Sweeney JA Oculomotor evidence for neocorticalsystems but not cerebellar dysfunction in autism Neurology 1999 52917ndash22

Mokler A Fischer B The recognition and correction of involuntaryprosaccades in an antisaccade task Exp Brain Res 1999 125 511ndash6

Munoz DP Everling S Look away the anti-saccade task and the voluntarycontrol of eye movement Nat Rev Neurosci 2004 5 218ndash28

Murphy DG Daly E Schmitz N Toal F Murphy K Curran S et alCortical serotonin 5-HT2A receptor binding and social communicationin adults with Aspergerrsquos syndrome an in vivo SPECT study Am JPsychiatry 2006 163 934ndash6

Nieuwenhuis S Nielen MM Mol N Hajcak G Veltman DJ Performancemonitoring in obsessive-compulsive disorder Psychiatry Res 2005 134111ndash22

Nieuwenhuis S Ridderinkhof KR Blom J Band GP Kok A Error-relatedbrain potentials are differentially related to awareness of response errorsevidence from an antisaccade task Psychophysiology 2001 38 752ndash60

Ohnishi T Matsuda H Hashimoto T Kunihiro T Nishikawa M Uema Tet al Abnormal regional cerebral blood flow in childhood autism Brain2000 123 (Pt 9) 1838ndash44

Oldfield RC The assessment and analysis of handedness The EdinburghInventory Neuropsychologia 1971 9 97ndash113

Paus T Primate anterior cingulate cortex where motor control drive andcognition interface Nat Rev Neurosci 2001 2 417ndash24

Pfefferbaum A Sullivan EV Hedehus M Lim KO Adalsteinsson EMoseley M Age-related decline in brain white matter anisotropymeasured with spatially corrected echo-planar diffusion tensor imagingMagn Reson Med 2000 44 259ndash68

Pitman RK A cybernetic model of obsessive-compulsive psychopathologyCompr Psychiatry 1987 28 334ndash43

Polli FE Barton JJ Cain MS Thakkar KN Rauch SL Manoach DSRostral and dorsal anterior cingulate cortex make dissociable contribu-tions during antisaccade error commission Proc Natl Acad Sci USA2005 102 15700ndash5

Polli FE Barton JJ Thakkar KN Greve DN Goff DC Rauch SL et alReduced error-related activation in two anterior cingulate circuits is relatedto impaired performance in schizophrenia Brain 2008 131 971ndash86

Polli FE Barton JJ Vangel M Goff DC Iguchi L Manoach DSSchizophrenia patients show intact immediate error-related performanceadjustments on an antisaccade task Schizophr Res 2006 82 191ndash201

Rabbitt PM Errors and error correction in choice-response tasks J ExpPsychol 1966 71 264ndash72

Radant AD Dobie DJ Calkins ME Olincy A Braff DL Cadenhead KSet al Successful multi-site measurement of antisaccade performancedeficits in schizophrenia Schizophr Res 2007 89 320ndash9

Reese TG Heid O Weisskoff RM Wedeen VJ Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echoMagn Reson Med 2003 49 177ndash82

Russell J Jarrold C Error-correction problems in autism evidence for amonitoring impairment J Autism Dev Disord 1998 28 177ndash88

Russell AJ Mataix-Cols D Anson M Murphy DG Obsessions andcompulsions in Asperger syndrome and high-functioning autism Br JPsychiatry 2005 186 525ndash8

Rutter M Le Couteur A Lord C Autism diagnostic interview-revised LosAngeles CA Western Psychological Services 2003

Santesso DL Segalowitz SJ Schmidt LA Error-related electrocorticalresponses are enhanced in children with obsessive-compulsive behaviorsDev Neuropsychol 2006 29 431ndash45

Shafritz KM Dichter GS Baranek GT Belger A The neural circuitrymediating shifts in behavioral response and cognitive set in autism BiolPsychiatry 2007 63 974ndash80

Silk TJ Rinehart N Bradshaw JL Tonge B Egan G OrsquoBoyle MW et alVisuospatial processing and the function of prefrontal-parietal networksin autism spectrum disorders a functional MRI study Am J Psychiatry2006 163 1440ndash3

South M Ozonoff S McMahon WM The relationship between executivefunctioning central coherence and repetitive behaviors in the high-functioning autism spectrum Autism 2007 11 437ndash51

Taylor SF Martis B Fitzgerald KD Welsh RC Abelson JL Liberzon Iet al Medial frontal cortex activity and loss-related responses to errorsJ Neurosci 2006 26 4063ndash70

Taylor SF Stern ER Gehring WJ Neural systems for error monitoringrecent findings and theoretical perspectives Neuroscientist 2007 13160ndash72

Thesen S Heid O Mueller E Schad LR Prospective acquisition correctionfor head motion with image-based tracking for real-time fMRI MagnReson Med 2000 44 457ndash65

Turner M Annotation Repetitive behaviour in autism areview of psychological research J Child Psychol Psychiatry 1999 40839ndash49

Ullsperger M von Cramon DY Neuroimaging of performance monitoringerror detection and beyond Cortex 2004 40 593ndash604

Ursu S Stenger VA Shear MK Jones MR Carter CS Overactiveaction monitoring in obsessive-compulsive disorder evidencefrom functional magnetic resonance imaging Psychol Sci 2003 14347ndash53

Meel CS Heslenfeld DJ Oosterlaan J Sergeant JA Adaptive controldeficits in attention-deficithyperactivity disorder (ADHD) the role oferror processing Psychiatry Res 2007 151 211ndash20

van Veen V Carter CS The timing of action-monitoring processes in theanterior cingulate cortex J Cogn Neurosci 2002 14 593ndash602

Wechsler D Wechsler Abbreviated Scale of intelligence San Antonio TXThe Psychological Corporation 1999

Whalen PJ Bush G McNally RJ Wilhelm S McInerney SC Jenike MAet al The emotional counting Stroop paradigm a functional magneticresonance imaging probe of the anterior cingulate affective division BiolPsychiatry 1998 44 1219ndash28

White K Ashton R Handedness assessment inventory Neuropsychologia1976 14 261ndash4

Zandt F Prior M Kyrios M Repetitive behaviour in children with highfunctioning autism and obsessive compulsive disorder J Autism DevDisord 2007 37 251ndash9

Abnormal response monitoring in autism Brain (2008) Page 15 of 15

Manoach DS Lindgren KA Cherkasova MV Goff DC Halpern EFIntriligator J et al Schizophrenic subjects show deficient inhibition butintact task-switching on saccadic tasks Biol Psychiatry 2002 51 816ndash26

Manoach DS Thakkar KN Cain MS Polli FE Edelman JA Fischl B et alNeural activity is modulated by trial history a functional magneticresonance imaging study of the effects of a previous antisaccadeJ Neurosci 2007b 27 1791ndash8

Manoach DS Weintraub S Daffner KR Scinto LF Deficient antisaccadesin the social-emotional processing disorder Neuroreport 1997 8 901ndash5

McDougle CJ Kresch LE Goodman WK Naylor ST Volkmar FRCohen DJ et al A case-controlled study of repetitive thoughts andbehavior in adults with autistic disorder and obsessive-compulsivedisorder Am J Psychiatry 1995 152 772ndash7

McPeek RM Skavenski AA Nakayama K Concurrent processing ofsaccades in visual search Vision Res 2000 40 2499ndash516

Miezin FM Maccotta L Ollinger JM Petersen SE Buckner RLCharacterizing the hemodynamic response effects of presentation ratesampling procedure and the possibility of ordering brain activity basedon relative timing Neuroimage 2000 11 735ndash59

Minshew NJ Luna B Sweeney JA Oculomotor evidence for neocorticalsystems but not cerebellar dysfunction in autism Neurology 1999 52917ndash22

Mokler A Fischer B The recognition and correction of involuntaryprosaccades in an antisaccade task Exp Brain Res 1999 125 511ndash6

Munoz DP Everling S Look away the anti-saccade task and the voluntarycontrol of eye movement Nat Rev Neurosci 2004 5 218ndash28

Murphy DG Daly E Schmitz N Toal F Murphy K Curran S et alCortical serotonin 5-HT2A receptor binding and social communicationin adults with Aspergerrsquos syndrome an in vivo SPECT study Am JPsychiatry 2006 163 934ndash6

Nieuwenhuis S Nielen MM Mol N Hajcak G Veltman DJ Performancemonitoring in obsessive-compulsive disorder Psychiatry Res 2005 134111ndash22

Nieuwenhuis S Ridderinkhof KR Blom J Band GP Kok A Error-relatedbrain potentials are differentially related to awareness of response errorsevidence from an antisaccade task Psychophysiology 2001 38 752ndash60

Ohnishi T Matsuda H Hashimoto T Kunihiro T Nishikawa M Uema Tet al Abnormal regional cerebral blood flow in childhood autism Brain2000 123 (Pt 9) 1838ndash44

Oldfield RC The assessment and analysis of handedness The EdinburghInventory Neuropsychologia 1971 9 97ndash113

Paus T Primate anterior cingulate cortex where motor control drive andcognition interface Nat Rev Neurosci 2001 2 417ndash24

Pfefferbaum A Sullivan EV Hedehus M Lim KO Adalsteinsson EMoseley M Age-related decline in brain white matter anisotropymeasured with spatially corrected echo-planar diffusion tensor imagingMagn Reson Med 2000 44 259ndash68

Pitman RK A cybernetic model of obsessive-compulsive psychopathologyCompr Psychiatry 1987 28 334ndash43

Polli FE Barton JJ Cain MS Thakkar KN Rauch SL Manoach DSRostral and dorsal anterior cingulate cortex make dissociable contribu-tions during antisaccade error commission Proc Natl Acad Sci USA2005 102 15700ndash5

Polli FE Barton JJ Thakkar KN Greve DN Goff DC Rauch SL et alReduced error-related activation in two anterior cingulate circuits is relatedto impaired performance in schizophrenia Brain 2008 131 971ndash86

Polli FE Barton JJ Vangel M Goff DC Iguchi L Manoach DSSchizophrenia patients show intact immediate error-related performanceadjustments on an antisaccade task Schizophr Res 2006 82 191ndash201

Rabbitt PM Errors and error correction in choice-response tasks J ExpPsychol 1966 71 264ndash72

Radant AD Dobie DJ Calkins ME Olincy A Braff DL Cadenhead KSet al Successful multi-site measurement of antisaccade performancedeficits in schizophrenia Schizophr Res 2007 89 320ndash9

Reese TG Heid O Weisskoff RM Wedeen VJ Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echoMagn Reson Med 2003 49 177ndash82

Russell J Jarrold C Error-correction problems in autism evidence for amonitoring impairment J Autism Dev Disord 1998 28 177ndash88

Russell AJ Mataix-Cols D Anson M Murphy DG Obsessions andcompulsions in Asperger syndrome and high-functioning autism Br JPsychiatry 2005 186 525ndash8

Rutter M Le Couteur A Lord C Autism diagnostic interview-revised LosAngeles CA Western Psychological Services 2003

Santesso DL Segalowitz SJ Schmidt LA Error-related electrocorticalresponses are enhanced in children with obsessive-compulsive behaviorsDev Neuropsychol 2006 29 431ndash45

Shafritz KM Dichter GS Baranek GT Belger A The neural circuitrymediating shifts in behavioral response and cognitive set in autism BiolPsychiatry 2007 63 974ndash80

Silk TJ Rinehart N Bradshaw JL Tonge B Egan G OrsquoBoyle MW et alVisuospatial processing and the function of prefrontal-parietal networksin autism spectrum disorders a functional MRI study Am J Psychiatry2006 163 1440ndash3

South M Ozonoff S McMahon WM The relationship between executivefunctioning central coherence and repetitive behaviors in the high-functioning autism spectrum Autism 2007 11 437ndash51

Taylor SF Martis B Fitzgerald KD Welsh RC Abelson JL Liberzon Iet al Medial frontal cortex activity and loss-related responses to errorsJ Neurosci 2006 26 4063ndash70

Taylor SF Stern ER Gehring WJ Neural systems for error monitoringrecent findings and theoretical perspectives Neuroscientist 2007 13160ndash72

Thesen S Heid O Mueller E Schad LR Prospective acquisition correctionfor head motion with image-based tracking for real-time fMRI MagnReson Med 2000 44 457ndash65

Turner M Annotation Repetitive behaviour in autism areview of psychological research J Child Psychol Psychiatry 1999 40839ndash49

Ullsperger M von Cramon DY Neuroimaging of performance monitoringerror detection and beyond Cortex 2004 40 593ndash604

Ursu S Stenger VA Shear MK Jones MR Carter CS Overactiveaction monitoring in obsessive-compulsive disorder evidencefrom functional magnetic resonance imaging Psychol Sci 2003 14347ndash53

Meel CS Heslenfeld DJ Oosterlaan J Sergeant JA Adaptive controldeficits in attention-deficithyperactivity disorder (ADHD) the role oferror processing Psychiatry Res 2007 151 211ndash20

van Veen V Carter CS The timing of action-monitoring processes in theanterior cingulate cortex J Cogn Neurosci 2002 14 593ndash602

Wechsler D Wechsler Abbreviated Scale of intelligence San Antonio TXThe Psychological Corporation 1999

Whalen PJ Bush G McNally RJ Wilhelm S McInerney SC Jenike MAet al The emotional counting Stroop paradigm a functional magneticresonance imaging probe of the anterior cingulate affective division BiolPsychiatry 1998 44 1219ndash28

White K Ashton R Handedness assessment inventory Neuropsychologia1976 14 261ndash4

Zandt F Prior M Kyrios M Repetitive behaviour in children with highfunctioning autism and obsessive compulsive disorder J Autism DevDisord 2007 37 251ndash9

Abnormal response monitoring in autism Brain (2008) Page 15 of 15