Exaggerated object affordance and absent automatic

15 Pages·2013·1.26 MB·English

cortex xxx (2013) 1e15
Availableonlineatwww.sciencedirect.com
Journalhomepage:www.elsevier.com/locate/cortex
Research report
Exaggerated object affordance and absent automatic
inhibition in alien hand syndrome
Jennifer McBridea,b,*, Petroc Sumnerc, Stephen R. Jacksond, Nin Bajaje and
Masud Husainf,g
aInstituteofCognitiveNeuroscience,UniversityCollegeLondon,AlexandraHouse,London,UK
bInstituteofNeurology,UniversityCollegeLondon,AlexandraHouse,London,UK
cSchoolofPsychology,CardiffUniversity,TowerBuilding,ParkPlace,Cardiff,UK
dSchoolofPsychology,UniversityofNottingham,Nottingham,UK
eDepartmentofClinicalNeurology,NottinghamUniversityHospitalsNHSTrust,Queen’sMedicalCentre,Nottingham,UK
fNuffieldDepartmentofClinicalNeurosciences,UniversityofOxford,UK
gDepartmentofExperimentalPsychology,UniversityofOxford,UK
a r t i c l e i n f o a b s t r a c t
Articlehistory: Patients with alien hand syndrome (AHS) experience making apparently deliberate and
Received13September2012 purposeful movements with their hand against their will. However, the mechanisms
Reviewed17October2012 contributingtotheseinvoluntaryactionsremainpoorlyunderstood.Here,wedescribetwo
Revised6December2012 experimentalinvestigationsinapatientwithcorticobasalsyndrome(CBS)withalienhand
Accepted11January2013 behaviourinherrighthand.First,weshowthatresponseswiththealienhandaremade
ActioneditorSergioDellaSala significantly more quickly to images of objects which afford an action with that hand
Publishedonlinexxx comparedtoobjectswhichaffordanactionwiththeunaffectedhand.Thisfindingsug-
geststhatinvoluntarygraspingbehavioursinAHSmightbeduetoexaggerated,automatic
Keywords: motor activation evoked by objects which afford actions with that limb. Second, using
Alienlimb abackwards masked primingtask, we found normal automatic inhibition ofprimed re-
Objectaffordance sponsesinthepatient’sunaffectedhand,butimportantlytherewasnoevidenceofsuch
Automaticinhibition suppression in the alien limb. Taken together, these findings suggest that grasping be-
Maskedpriming haviours in AHS may result from exaggerated object affordance effects, which might
potentiallyarisefromdisruptedinhibitionofautomaticallyevokedresponses.
ª2013ElsevierLtd.Allrightsreserved.
1. Introduction releasingobjectsoncegrasped(seee.g.,BiranandChatterjee,
2004; Della Sala et al., 1991). Despite the fact that such in-
Althoughmosthealthyadultsfeelthattheyhaveagreatdeal dividuals make seemingly deliberate and purposeful move-
of control over their actions, some neurological patients do mentswiththeir“alien”hand,thereiscleardisparitybetween
not.Patientswithalienhandsyndrome(AHS)mayinvoluntarily actions performed by the alien limb and the intentions of
graspobjectsplacedwithintheirreach,experiencingdifficulty patients,whosubjectivelyreportthatthehandisnotunder
* Corresponding author. Institute of Cognitive Neuroscience, University College London, Alexandra House, 17 Queen Square, London
WC1N3AR,UK.
E-mailaddresses:[email protected],[email protected](J.McBride).
0010-9452/$eseefrontmatterª2013ElsevierLtd.Allrightsreserved.
http://dx.doi.org/10.1016/j.cortex.2013.01.004
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004
2 cortex xxx (2013) 1e15
theircontrol.Instead,theyreportthatthealienlimbbehaves irrelevanttotheobserver’stask.Ofcourse,inhealthypeople,
asthoughithasamindofitsownorisbeingcontrolledbyan objectsdonotalwayselicitactionstowardsthem;thatwould
external agent (e.g., Assal et al., 2007; Biran and Chatterjee, makepeopleentirelystimulus-bound.Hencethereisaneed
2004; Fitzgerald et al., 2007). Although these remarkable tosuppresssuchautomaticallyevokedaffordances.Indeedin
grasping behaviours in AHS are now well-documented, we healthyobservers,thereisnowcompellingevidencethatre-
understandverylittleaboutthemechanismsthatmightun- sponsesautomaticallyprimedbytheenvironmentcanalsobe
derliesuchbehaviour. automatically suppressed (for reviews see Eimer and
AHSisarelativelyraresyndrome(forareview,seeFisher, Schlaghecken,2003;McBrideetal.,2012a;Sumner,2007).
2000), so detailed investigation has been correspondingly Usingabackwardsmaskedprimingparadigm,Eimerand
sparse.Someofthemostdetailedexperimentalworkcomes Schlaghecken (1998) showed that participants’ responses to
from Riddoch and her colleagues (e.g., Humphreys and targets were typically speeded if targets were preceded by
Riddoch, 2000; Riddoch et al., 1998). They instructed a pa- a compatible prime (a prime associated with the same
tient with bilateral AHS to reach out and grasp a cup with responseasthetarget)comparedtowhentargetsfollowedan
ahand.Thepatientwasabletodothiscorrectlyaslongasthe incompatible prime (a prime associated with the opposite
cup’s handle was on the same side as the hand they were response to the target). Thus, a target directing a left hand
instructedtousetograspthecup.However,ifthehandlewas response is faster if preceded by a (backward-masked) left
on the opposite side, “interference” errors were generated prime relative to a right prime. However, when the interval
withthepatientreachingwithwhicheverhandmatchedthe between masked-prime and target is extended beyond
side the cup’s handle was on. For example, if instructed to w150 msec this usual positive compatibility effect (PCE)
graspacupwiththerighthandwhenthecup’shandlewasto actually reverses to produce a negative compatibility effect
the left, the patient would often erroneously grasp the cup (NCE;EimerandSchlaghecken,1998).Now,atargetdirecting
withthelefthand.Theseeffectsareunlikelytobeperceptual a left hand response is actually slower if it is preceded by
or attentional because fewer interference errors were made a(backward-masked)leftprimerelativetoarightprime.
whenthetaskwastopointratherthantograsp,orwhenthe Aslongasappropriatestimuliareused(seeSchlaghecken
patientrespondedtoLEDsinsteadoftocups.Thesefindings etal.,2007;LlerasandEnns,2004;Sumner,2008),thisNCEcan
suggestthat,forthispatient,simpleobservationofagrasp- be interpreted as reflecting automatic suppression of the
ableobjectmightbesufficienttoelicittheassociatedmotor primed response (see e.g., Eimer and Schlaghecken, 2003;
plan for interacting with that object, even when the plan Ja´skowski,2007,2008,2009;Sumner,2007).Accordingtothese
conflictswithcurrentgoals(seealsoBlakemoreetal.,2002). sensorimotor accounts of the NCE, initial motor activation
Indeed,suchinvoluntarygraspingbehaviourinAHSmay evoked by the prime is subsequently suppressed when the
be related to the longstanding view that, even in healthy primeisremovedoranovelstimulus(themask)isaddedto
adults,viewingvisualobjectscanautomaticallyprimeactions the scene (e.g.,Boy et al., 2008; Ja´skowski, 2007, 2008, 2009).
intheobserver.AHSmightrepresentanexaggeratedformof This suppression means that it takes longer to initiate the
such automatic priming. Gibson (1979) described “affordan- suppressed response relative to a response which has not
ces”aspropertiesofobjectsintheenvironmentwhichprime beeninhibited,therebyproducingtheNCE.
an observer to act. For example, seeing a teapot with the SumnerandHusain(2008)suggestedthatsuchautomatic
handletotherightmightautomaticallyprimetheobserverto suppression of automatically evoked responses might be
reach out with the right hand to grasp the handle. Object crucialforgoal-directedbehaviourbecauseitfreesanorgan-
affordanceeffectssuchasthesehavebeenextensivelystud- ism from stimulus-bound responses, and provides a level
ied in healthy adults using stimulus-response compatibility playingfieldforalternativeactionstooccuraccordingtothe
paradigms(e.g.,ChoandProctor,2010;Derbyshireetal.,2006; current goals of an animal. Consistent with this proposal,
Iani et al., 2011; McBride et al., 2012b; Pellicano et al., 2010; Vainioandcolleagueshavereportedthatautomaticinhibition
Phillips and Ward, 2002; Tucker and Ellis, 1998, 2001). For isnotrestrictedtomasked-primeparadigms,butalsooccurs
example,TuckerandEllis(1998)presentedpicturesofobjects whenresponsesareaffordedbygraspablestimuli(e.g.,Vainio,
which healthy observers classified as upright or inverted as 2009;Vainioetal.,2011;VainioandMustonen,2011).
quickly and accurately as possible using a manual button Such considerations naturally raise the possibility of
press.Crucially, the objects could be presentedso that they grasping behaviour in AHS arising from disruption of auto-
maximallyaffordedaresponsewitheithertheleftortheright matic inhibitory mechanisms which, in healthy observers,
hand.Althoughthisleft/rightorientationwasirrelevanttothe halt inappropriate activation of responses afforded by the
participants’ task, responses were significantly faster and environment (see also Blakemore et al., 2002; Giovannetti
moreaccuratewhenparticipantsrespondedwithahandthat et al., 2005). At present, however, there is very little direct
wascongruentwiththe(task-irrelevant)responseaffordedby evidencetosupportthishypothesis,althoughtherearesome
theobject. suggestivepiecesofevidence.Inhealthyadults,thesupple-
Thesefindings, and themanyothers likethem (e.g.,Cho mentary motor area (SMA) in the medial frontal lobes is
and Proctor, 2010; Derbyshire et al., 2006; Iani et al., 2011; associatedbothwithsimplyviewinggraspableobjectswith-
McBrideetal.,2012b;Pellicanoetal.,2010;PhillipsandWard, outreachingforthem(e.g.,Gre`zesandDecety,2002)aswellas
2002; Tucker and Ellis, 1998, 2001), suggest that through with successful automatic inhibition of primed responses
experience observers associate objects with particular ac- indexedbytheNCE(e.g.,Boyetal.,2010a,2011;Sumneretal.,
tions, and that these actions can be (partially) evoked by 2007).Intriguingly,AHShaslongbeenassociatedwithdamage
perceptual processing of the object even when they are tothesesamemedialfrontalregions(e.g.,Bakheitetal.,2013;
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004
cortex xxx (2013) 1e15 3
MarchettiandDellaSala,1998).AHSisincreasinglyrecognised in the alien hand compared to the non-alien hand (and
in corticobasal syndrome (CBS, to distinguish it from the relative to healthy controls); and (ii) automatic inhibition
pathologicentity,corticobasaldegeneration, CBD; seeBoeve of automatically evoked responses is reduced in the alien
etal.,2003).CBSisarare(annualincidencerateshavebeen limb.
estimatedataround.02per100,000individuals;Winteretal.,
2010),slowlyprogressive,neurodegenerativeconditionwhich
affects cortical regions as well as the basal ganglia (e.g., 2. Case report
Fitzgeraldetal.,2007;Murrayetal.,2007;Riddochetal.,1998;
TiwariandAmar,2008).Interestingly,CBSisalsoassociated PatientSAwasa72-year-old,right-handedwomanwhofirst
with metabolic impairment in the SMA (e.g., Garraux et al., reportednoticinghersymptoms3yearspreviouslywhenshe
2000). hadafall.Atthattime,itwasobservedthatherspeechhad
Tothebestofourknowledge,patientswithAHShavenot a telegraphic quality. She developed progressive difficulty
previously been tested on object affordance “compatibility” speaking and writing, swallowing, and controlling her right
tasks, or paradigms designed to investigate automatic inhi- hand.Shebegantouseherrightarmlessfrequently.Although
bitionofprimedactions(e.g.,maskedpriming).Wemetwith shecouldvoluntarilymoveitifnecessary,therewasalackof
fourpatientswithCBS(seeTable1forasummaryofpatients’ spontaneous use. Soon, she began to experience difficulty
details),butunfortunatelythemotorsymptomsexperienced chopping vegetables using the right hand. She encountered
bythreeofthesepatientsweresoseverethattheywerenot problems with her right hand grip, but at that time had no
able to complete basic motor tasks. However, one patient, difficultylettingobjectsgo.Priortotesting,shenotedthather
PatientSA,wasabletomakespeededmanualresponseswith walking had slowed. She began to experience difficulties
either hand according to stimuli presented. Patient SA had standingfromaseatedposition.Therewasnofamilyhistory
AHS which affected her right hand (involuntary grasping ofneurodegenerativedisease.
movements to objects placedwithinherreach),and no evi- On examination, she had a profound expressive aphasia
denceofalienbehaviourinherlefthand(seeTable1). and impaired articulation. However, she was able to com-
Here we report results from two experiments conducted prehend 3-stage commands well. Visual fields were full to
with Patient SA. Experiment 1 was designed to investigate confrontation. There was no evidence of visual or tactile
whetherobjectaffordanceeffectswerestrongerinthealien extinction. Eye movements were full, but she was slow to
hand relative to the unaffected hand. Our second study initiate saccades, particularly towards the left compared to
comparedautomaticinhibitionofactioninthetwohands.If therightandtherewassomeevidenceofgazeimpersistence.
grasping behaviour in AHS arises because of disruption of SuchoculomotordeficitsarenotuncommoninCBSpatients.
normal automatic suppression of afforded responses, one Therewas no facial weaknessand palatal movements were
mightpredictthat(i)objectaffordanceeffectsareexaggerated normal.Therewasnopoutreflex.
Table1ePatientdetails.DetailsofthefourpatientswithCBS.PatientsFC,DH,andEFhadmotordeficitsthatwereso
severethattheywerenotabletoperformsimplemotortasks.Theexperimentsreportedherewereconductedonlywith
PatientSA.DIndicatesthatasymptomwaspresent,eindicatesthatthesymptomwasnotdetected.
PatientSA PatientFC PatientDH PatientEF
Demographics
Age(years) 72 70 61 59
Sex(M/F) F F M F
Manualsymptoms
Involuntarygrasping þRighthand þBothhands e þLefthand
Difficultyreleasing e þBothhands e e
Intermanualconflict e e e e
Armlevitation e þBothhands e e
Mirrormovements e e e e
Dystonia e þBothhands e e
Dyspraxia e þBothhands þSomeinlefthand þ
Tremor e e e e
Rigidity þRightarm þBothhands þEspeciallyleftarm þ
Impairedeye-movements
Voluntarysaccades
Horizontal Slowedtoleft Impaired e e
Vertical e Impaired Impaired e
Reflexivesaccades
Horizontal e Impaired e Impaired
Vertical e Impaired e e
Othersymptoms
Telegraphicspeech þ þ þ þ
Visualextinction e þ e e
Tactileextinction e þ þ e
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004
4 cortex xxx (2013) 1e15
There was rigidity of the right arm and poor fine finger caudateheadbilaterally.Thesefindingswouldbeconsistent
movements, but good strength throughout. The right hand with the clinical diagnosis of CBS. Selected images in Fig. 1
showed evidence of mild alien hand behaviour, with invol- demonstrate loss of volume of the left medial frontal and
untarygraspingofanyobjectthatwasbroughtclosetoit.The parietalcortexwithapathologicallywidenedcingulatesulcus
patientwasadamantthatshewasnotwillingthehandtodo (whitearrowhead);lossofcorticalvolumeadjacenttoawid-
this, and she could not stop this behaviour even when she ened intraparietal sulcus particularly involving the superior
madeanefforttodoso.Therewasnoevidenceofalienhand parietal lobe, most prominently on the left (yellow arrow-
behaviourinthelefthand. head); widened sulci over superior parietal and frontal re-
Examinationdidnotrevealanydystoniaorlimbapraxia, gions, including the left central sulcus (red arrowhead); and
aboveandbeyondtheproblemsassociatedwithfinecontrolof reductionincaudateheadvolumebilaterally(leftsidemarked
therighthandmovements.Therewasnoamorphosynthesis withgreenarrowhead).
in the left hand. When she walked, there was reduced arm SA completed the two different experiments on two dif-
swing,moreprominentlyontherightthanontheleft,butshe ferentdays,approximately4weeksapart.Theaffordancetask
had a good stride length and postural reflexes were intact. was performed first. This study was approved by the local
Therewasnoevidenceofsomeoftheotherbehaviourswhich humansubjectsethicscommitteeandthepatientgavewrit-
are common in AHS: no levitation of either arm, no mirror teninformedconsentpriortotesting.
movements,andnointermanualconflictbetweenthehands.
Overall, the clinical presentation was considered to be con-
sistentwithCBS. 3. Experiment 1jObject affordance task
Magnetic resonance imaging (MRI; Fig. 1) demonstrated
cortical atrophy,slightly more prominent over parietalthan Stimuli, task, response measurement and analysis follow
frontal regions and in the left hemisphere compared to the closely from those reported in McBride et al. (2012b) which
right. In addition, there was reduction in volume of the reporteddatafromyounghealthyindividuals.
Fig.1eMRIbrainscansofPatientSA.(A)Sagittalimagedemonstratinglossofvolumeoftheleftmedialfrontalandparietal
cortexwithapathologicallywidenedcingulatesulcus.(B)Coronalimagesshowinglossofcorticalvolumeparticularlyofthe
superiorparietallobeadjacenttoawidenedintraparietalsulcus,mostprominentlyontheleft,togetherwithreductionin
caudateheadvolumebilaterally.(C)Axialimagesdemonstratingwidenedsulcioversuperiorparietalandfrontalregions,
includingtheleftcentralsulcus.Whitearrowhead[cingulatesulcus;yellowarrowhead[leftintraparietalsulcus;red
arrowhead[leftcentralsulcus;greenarrowhead[caudatehead.
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004
cortex xxx (2013) 1e15 5
3.1. Stimuliandtask experimenterifshehaddifficultyrecognisinganyoftheob-
jects(shedidnotreportanydifficulty).Therewereanequal
Eachtrialbeganwithpresentationofablackfixationcrosson numberoftrialscontainingstimuliofeachcategory(kitchen
awhitebackgroundonaCRTmonitor(seeFig.2).Thiscross or toolbox), and an equal number of congruent and incon-
subtended1degree(cid:2)1degreeofvisualangle,andwaspre- gruenttrialswithtargetsofeachcategory(kitchenortoolbox)
sented in the centre of the screen for 1500 msec. Following in each block. Order of presentation was shuffled randomly
a blank interval (200 msec), an imageof a target object was and independently for each block, and which image of the
presented at screen centre for 2000 msec. Stimuli were pic- targetcategorywaspresentedwasdeterminedrandomlyand
tures of ten household objects taken from the Object independentlyonatrial-by-trialbasis.
Databank (courtesy of Michael J. Tarr, Brown University,
http://www.tarrlab.org/)andVerfailleandBousten’s3Dobject 3.2. Apparatus
database (see Verfaillie and Boutsen, 1995; Boutsen et al.,
1998). Objects were matched for orientation. Five objects Stimuliweredisplayedona21inchCRTmonitor(1024(cid:2)768)
belonged in a kitchen (fork, frying pan, knife, saucepan, whichtheparticipantviewedbinocularlyfromadistanceof
spoon),andfiveinatoolbox(chisel,pliers,saw,screwdriver, 60 cm. Stimulus timing and presentation was locked to the
spanner).Imagessubtended10.6e17.3degreesofvisualangle screen refresh rate of 100 Hz. Stimuli were presented using
horizontally, and 2.8e5.3 degrees of visual angle vertically. aPCrunningPresentationsoftware(version13.1;http://www.
Objectswereorientedwiththeirhandlesaffordinganaction neurobs.com).
withtheleftorrighthand. Responsesweremeasuredusingtwospeciallydesignedde-
The participant was instructed to respond by making vices,constructedfromarolledaneroidsphygmomanometer
ashort,sharpsqueezeofagripforcemeasuringdevice(details cuff(Boso-clinicusI,ref:030-0-111),inflatedto20mmHg,con-
below) with the left hand for kitchen objects, and with the nectedtoapressuretransducer.Onedevicewasheldineach
right hand for toolbox objects. Therefore, depending on the hand, and the participant was instructed to make their re-
orientationoftheobjectpresented,theobjectcouldaffordan sponsesbymakingashort,sharpsqueezeoftherolledcuffand
actionthatwaseither“congruent”or“incongruent”withthe then release their grip. Grip force was converted to voltage
requiredresponse.Thenexttrialbeganfollowingablankin- which was digitised and stored using a LabJack U3 HV data
terval(1000msec). acquisitiondevicewithDAQFactorysoftware.Dataweresam-
Before the experiment began, the participant practiced pledat1000Hz.Theparticipantwasencouragedtorespondas
makingresponseswhileobservingtheoutputfromthepres- quicklyaspossiblewhilemaintainingahighlevelofaccuracy,
sure transducers on a computer screen. Following a short butnoresponsefeedbackwasgivenduringtheexperiment.
practiceblock(12trials)PatientSAcompletedtwosessionson
the same day, each containing 4 blocks of 64 trials each, 3.3. Dataanalysis
totalling512trialsafterpractice.Therewasanopportunityto
restbetweenblocks.Allobjectswerepresentedatleastonce Continuous force recordings were locked to stimulus onset
during practice, and Patient SA was instructed to tell the andepochedintoperiodsof2500msec,beginning500msec
Fig.2eSequenceofeventsinatypicalaffordanceexperimenttrial.PatientSAwasinstructedtomakeasqueezeresponse
withthelefthandiftheobjectpresentedbelongedinakitchen,andtomakeasqueezeresponsewiththerighthandifthe
objectbelongedinatoolbox(thelatterisdepictedhere).Eachobjectcouldbepresentedintwopossibleorientations,sothat
itaffordedanactionwiththeleftorrighthand(alltrialtypeswereequiprobable).Objectorientationwasirrelevanttothe
task.
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004
6 cortex xxx (2013) 1e15
before target onset. Data were smoothed using a simple 5- than3SDsfromeachparticipants’meanRTforthathandfor
point moving average to reduce high-frequency noise. The thatcondition).Theelderlyhealthycontrolshadfasteroverall
resulting waveforms were baseline corrected on a trial-by- RTs (mean ¼ 609 msec) and showed a smaller congruency
trialbasisaccordingtotheaveragebaselineactivityforeach effect [mean ¼ 14 msec; congruency effect was reliable in
responsedeviceduringthe200msecpre-stimulusperiodon elderly controls: t(24) ¼ 3.15, p ¼ .004] than for Patient SA’s
eachtrial. alienhand.1TodirectlycomparetheperformanceofPatient
A response (either correct or incorrect) was said to have SA’s alien hand to that of healthy elderly controls, we con-
occurred in a trial if at any point after the target stimulus vertedtheoverallmeanRTandaffordanceeffectforthealien
onsetuntiltheendofthetrial,twocriteriaweresatisfied:(i) handtoz-scores,calculatedaccordingtotheelderlycontrols’
the force measured was greater than 3 SDs from the mean samplemeansandSDs.Thez-scoresfortheaffordanceeffect
forcemeasuredduringthepre-stimulusbaselineperiodand and overall RT shown for Patient SA’s alien hand were 2.82
that was followedby at least 18/20pointsthat also reached and4.24,respectively.Asthesearebothbeyondthe95%limits
thisthreshold;and(ii)therewasanincreaseinresponsebyat (two-tailed) of the controls’ distributions (95% limits are
least .01 V over the following 100 points or less. Response indicatedbyaz-scoreof1.96),itisunlikelythatPatientSA’s
onset time (RT) was defined as the first point that satisfied effectsaresimplyanextremecaseinthenormalelderlydis-
these criteria. Peak response was determined as the max- tribution,andthattheseeffectsareduetoage.2
imum amplitude of the response made in a trial that was To investigate how often differences like those exhibited
surroundedbypointsoneithersidewiththesameorlower bySA’salienlimbexistinhealthycontrols,weanalysedthe
amplitude. individual affordance effects for left and right hands in the
Outliersweredefinedasanyresponsesgreaterthanthree younghealthycontrolspreviouslyreportedbyMcBrideetal.
standarddeviations(SDs)awayfromthemeanresponsetime (2012a), plus the previously unpublished data from elderly
forthathand,inthatcondition(congruentorincongruent)in healthy controls, mentioned above. None of these healthy
thattestingsession.Remainingcorrectresponsetimeswere adultsshowedthesamepatternofeffectsshownbySA,with
enteredintoa2(hand)(cid:2)2(congruency)(cid:2)2(session)factorial a significant interaction between the effects of hand and
analysisofvariance(ANOVA).Therewasnosignificanteffect congruency,andasignificantasymmetryinoverallRT.
of session (morning or afternoon) on RTs, and the effect of However,overallRTsinSA’salienhandwerelongerthan
session did not interact with any of the other factors (all those recorded in the non-alien hand, as well as those
F’s < 1). Therefore, subsequent analyses collapse across reported in young and elderly controls. Therefore, we per-
session. formedfurtheranalysestoinvestigatethepossibilitythatthe
differenceincongruencyeffectacrossPatientSA’shandswas
simply attributable to the difference in baseline RT. We re-
4. Results and discussion plotted the congruency effect as a function of RT in a delta
plot(seevandenWildenbergetal.,2010,forareviewofthis
The key motivation in conducting Experiment 1 was to techniqueanditsadvantages).Foreachhandseparately,un-
examinewhetherresponseswithPatientSA’salienhandwere trimmed(includingthose trialsconsidered“outliers”forthe
moresusceptibletoprimingbyobjectaffordancesrelativeto ANOVAanalysis)correctRTsweredividedaccordingtotrial
responseswithhernon-alienhand.Herresponsesweregen- congruency (congruent or incongruent), rank-ordered, and
erally slower than those reported for healthy adults on this then divided into eight bins of equal size. On two trials, no
task(seeMcBrideetal.,2012b).Moreover,SA’sleft(non-alien) correct response was detected. Data for these trials were
hand responses were significantly faster than right (alien) replacedwiththemeancorrectRTforthathandandcondition
hand responses [see Fig. 3; left mean ¼ 836 msec vs (thisisameanstokeepthetotalnumberoftrialsthesamein
right¼1090msec,F(1,497)¼307.47,p<.001].Furthermore, each condition and dividable by 8, to avoid problems asso-
stimuliwhichaffordedacongruentresponseproducedfaster ciatedwithunequalbinsizes).ThemeanRTineachbinfor
reactions than stimuli which afforded an incongruent each condition was then calculated and the difference be-
response [incongruent mean ¼ 983 msec; congruent tweenincongruentandcongruenttrialsisplottedagainstthe
mean¼944msec;F(1,497)¼7.13,p¼.008].Importantly,the mean RT for that bin (see Fig. 3B), giving the size of the
congruencyeffectwasmuchlargerforthealienthanforthe
non-alien hand [significant congruency (cid:2) hand interaction:
F(1,497)¼6.62,p¼.010].ThisinteractionisshowninFig.3A. 1We have also examined the median affordance effects from
untrimmed RTs. For Patient SA: median affordance effect non-
Thecongruencyeffectshowninthealienhand(76msec)
alien hand ¼ 7 msec; median affordance effect alien
wasseveraltimeslargerthanwehavefoundusingidentical hand ¼ 67 msec; for healthy elderly controls: mean of median
apparatus in healthy young controls (mean of median affordanceeffect¼13msec.
RTs¼16msec,seeMcBrideetal.,2012b).Wealsohaveasyet 2Recently,CrawfordandGarthwaite(2012)havesuggestedan
unpublished data on this task from elderly healthy controls alternativemethodfortestingwhetheraneffectshownbyasingle
(N¼26;aged54e75years;meanage¼64years;onepartic- caseislikelytohavebeendrawnfromthecontroldistributionby
ipant,whoshowedanaverageaffordanceeffectof(cid:3)111msec, conducting a t-test following Crawford and Howell’s (1998)
method. For completeness, we have also conducted this test,
wasremovedasanoutlier).AsforPatientSA,wecalculated whichalsoindicatesthattheaffordanceeffect[t(24)¼2.76,p<.01]
each elderly control’s mean RT for each condition, after andoverallRT[t(24)¼4.16,p<.01]forPatientSA’salienhandare
removingoutliers(followingthesamecriteriaforoutlierse- reliablydifferentfromthoseshowninelderlycontrols(testsare
lectionasforPatientSA,weremovedanyRTthatwasgreater two-tailed).
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004
cortex xxx (2013) 1e15 7
A B
Fig.3eResultsfromaffordancetask.(A)MeanRTs(trimmedatD/L3SDsfromthemeanforeachhandineachcondition)
forresponsesmadewitheachhandtotargetstimulithatwereincongruent(greybars)orcongruent(whitebars)withthe
responseaffordedbytheobject.ErrorbarsdenoteD/L1SEM.(B)MeanoctilecongruencyeffectsasafunctionofmeanRT
forthatoctile,calculatedseparatelyfortheleft(greyline)andright(blackline)hands.Datafromthelongestandmost
variableRTbinareincludedforcompleteness,andareshownhereasdottedlinesandopensymbols.*DenotesBonferroni-
correctedt-testp<.05.
congruency effect for that part of the RT distribution. Un- t(24)¼.55,p¼.59;elderlycontrols’meanaffordanceeffectfor
trimmed RTs such as these typically have long “tails” pro- kitchen stimuli ¼ 12 msec; for toolbox stimuli ¼ 16 msec;
duced from a number of slow outlier responses (see also t(24) ¼ .570, p ¼ .574]. Therefore, there is no indication that
Maylor et al., 2011). The increased variability and reduced thereisanyreliabledifferenceintheaffordanceselicitedby
reliabilityoflongRTsmeanthatitisdifficulttodrawmean- differentstimulustypes.
ingfulconclusionsfromthelastbin,butitisincludedinthe As noted in the methods, the particular object presented
figureforcompleteness(dottedlines). was determined randomly and independently for each trial
If the difference in congruency effects across the two (while the number of trials in each condition was held
handswassimplyinlinewithdifferencesinbaselineRT,then constant). Therefore, perhaps the very large affordance
wemightexpectsimilarcongruencyeffectsinthosepartsof effectshowninPatientSA’sright(alien)handisduetoasubset
theRTdistributionwhichoverlapacrossthetwohands(i.e., of toolbox stimuli which by chance appeared more (or less)
for responses which onset between approximately often than the others. To investigate this possibility, we cal-
800e1000msecafterthestimulusappeared).However,there culated how often each particular toolbox object was pre-
isclearseparationbetweenthecongruencyeffectsshownby sented. Four out of the five objects appeared 54 or 55 times
theleftandrighthandsinthispartoftheRTdistribution,soit each,andoneitem(thechisel)waspresented39times.Reas-
seems unlikely that the interaction between the effects of suringly,therewasnoreliableinteractionbetweentheafford-
handandcongruencyisbeingdrivenbydifferencesinbase- ance effect and the particular toolbox exemplar presented
lineRT.Thus,PatientSAshowsasignificantlylargerafford- [congruency (cid:2) object interaction: F(4, 239) ¼ 1.20, p ¼ .31].
ancecongruencyeffectwhenmakingresponseswithheralien Furthermore, we repeated the analysis of correct RTs after
(right)hand,comparedtohernon-alien(left)hand,suggesting removingthosetrialswhichcontainedtherelativelyinfrequent
thatanobject’saffordancehadanexaggeratedeffectonher exemplar (the chisel). The affordance effect shown for the
alienlimbcomparedtotheunaffectedhand. remaining toolbox items remains very large and statistically
The stimulus-response mappings Patient SA used in significant (incongruent mean ¼ 1122 msec; congruent
Experiment 1 were held constant over the course of the mean¼1064msec;congruencyeffect¼58msec,p¼.03).
experiment.ThiswastopreventanypossibledifficultiesPa- Errorswereveryinfrequent(anabove-thresholdresponse
tient SA might have experienced with task-switching if we was made by the erroneous hand on only 10/512 trials e
had changed the stimulus-response mapping part-way approximately2%ofalltrials).Thiserrorrateissimilartothat
through the experiment (see Alvarez and Emory, 2006, for whichweobservedinyoung(approximately5%),andelderly
discussion).Toexaminewhetherthereisanydifferenceinthe (approximately3%)healthycontrols.Oftheseerrorsmadeby
affordance effects normally produced by different stimulus PatientSA,8/10weremadebytheright(alien)handwhenthe
types, we analysed affordance effectsto these same stimuli taskrequiredaresponsewiththelefthand.Sixerrorswere
fromyoung(previouslyreportedinMcBrideetal.,2012b)and detected by the alien limb in response to affordance incon-
elderly(previouslyunpublished)healthycontrolparticipants, gruent trials (in other words, when the object presented
where stimulus-response mapping was counterbalanced requiredalefthandresponse,butwasorientedsuchthatit
across participants. Young and elderly healthy controls affordedaright-handresponse),and2errorsinresponseto
showed comparable affordance effectsfor kitchen and tool- affordancecongruenttrials.Errorswerenotconfinedtoone
box stimuli [young controls’ mean affordance effect for particularstimulus,andinsteadwerespreadacross7differ-
kitchenstimuli¼18msec;fortoolboxstimuli¼15msec;no ent exemplars. As errors were so infrequent, they were not
reliable difference of stimulus type on affordance effect: analysedanyfurther.
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004
8 cortex xxx (2013) 1e15
SAcompleted8blocks(4ofeachSOAcondition)of84trials
5. Experiment 2jMasked priming task
each,makingatotalof672trials.
Aschematicofthestimuliandtimingsforthistaskcanbe
InExperiment2,weusedabackwardsmaskedprimingtask
seen in Fig. 4. Note that the total presentation time of each
(adapted from Sumner et al., 2007) to investigate automatic
stimulus (prime, mask, target) was the same in both SOA
inhibitionofresponsesthathadbeenautomaticallyprimedin
conditions.
thealienandnon-alienhands.
Thetargetstimulusappearedafterthemaskhadonset,and
In order to be sure of producing automatic priming and
waseitheraleft-,orright-pointingdoublearrowhead(sothatit
inhibitionofresponses,itwasnecessarytochangetheinter-
waseithercompatibleorincompatiblewiththeprimestimu-
val between masked-prime and target. There are several
lus). The target appeared in one of three possible locations,
methodsreportedin theliteratureto achievethis.Onepos-
centred 5 degrees of visual angle to the left, to the right, or
sibilitywouldbetopresentthetargetstimulusoncethemask
abovethecentreofthescreen.Theparticipantwasinstructed
had offset, and change the duration of the mask. However,
toignorethetarget’sposition,andtorespondtothedirection
shortermaskswouldbeexpectedtomasktheprimestimulus
ofthisarrowheadbysqueezingwitheitherthelefthand(for
lesswell,whichcouldhavestrongeffectsontheprimingof
left-pointingtargets)ortherighthand(forright-pointingtar-
responses. Alternatively, some researchers have usedmeta-
gets) as quickly and accurately as possible. In each block of
contrast masking e that is, to use a stimulus which masks
trials there were an equal number of trials with each target
theprimebysurroundingit.However,suchmasksareprob-
type(left-,andright-pointing)ineachpossibleposition(left-,
lematicbecausemaskscanactasprimestimuliintheirown
right-, above-centre), with each prime type (compatible and
right e as masks of this type typically contain elements of
incompatible), presented in randomly shuffled order deter-
both possible primes, any NCE obtained using such a mask
mined independently for each block. The target stimulus
may not be produced by response inhibition, but by mask-
remained on the screen for 200 msec. There was a blank
inducedprimingoftheresponseoppositethatevokedbythe
intertrialinterval(ITI)of2500msecbeforethenexttrialbegan
prime (see “object updating” e.g., Lleras and Enns, 2004;
withthefixationcross.Datarecordingandanalysisprocedures
Sumner,2008).Aswewereinterestedintheeffectsofauto-
werethesameasthoseusedintheaffordanceexperiment.
maticresponseinhibition,wesoughttoavoidthispossibility. Left-andright-pointingdoublearrowheads(e.g.,“<<”and
Thus, we followed a well-established, standard method “>>”)servedasprimesandtargets.Thelinesmakingupthese
reportedpreviouslyintheliteraturewhichisknowntoreliably
stimuliwereeach1degreeofvisualanglelong,andthelines
producePCEsandNCEsandwhichkeepsthedurationsofeach ineacharrowheadhadanangularseparationof60(cid:4)(30(cid:4)above
stimulus (prime, mask, and target) constant. We presented
and below the horizontal). Masks were constructed of 30
masks and primes simultaneously in short stimulus onset pseudo-randomlyorientatedlinesarrangedintoa6(cid:2)5grid
asynchrony(SOA)conditions,andintroducedablankscreen
centred over the centre of the screen. To prevent any per-
betweenmaskandtargetinlongSOAconditions(seee.g.,Boy
ceptual interactions between prime and mask modulating
etal.,2010a;2010b;BoyandSumner,2010;Schlagheckenetal.,
priming effects (see “object updating” accounts of the NCE
2006,2003;SchlagheckenandEimer,2002;Schlagheckenand
e.g., Lleras and Enns, 2004) lines in the mask avoided any
Maylor,2005).Itispossiblethatdifferencesintheshort-and
orientationwithin5degreesofthelinesmakinguptheprime
long-SOAtrialsequencemayaffectglobalRTseforexample
andtarget.Thelinesinthemaskwerebetween1.5and3de-
theoffsetofthemaskinthelongSOAconditionmayserveas
grees of visual angle long. Line length and orientation were
awarningsignalthatthetargetisabouttoappearandthus
determinedrandomlywithintheselimitsandindependently
speedresponsesinthelongSOAcondition.However,assuch
for each line in the mask. Thus, the mask was between
effectsareexpectedtohaveaglobalinfluenceonRTs,andnot 3.5 (cid:2) 3.5e5.5 (cid:2) 5.5 degrees of visual angle, centred on the
affect one condition (compatible or incompatible) or hand
centre of the screen. A new mask was constructed on each
(alienornon-alien)morethantheother,theyshouldnotbe
trialtopreventperceptuallearningofthemaskwhichcouldin
able to account for any differences in compatibility effect
turnleadtoincreasedprimeidentification(e.g.,Schlaghecken
showninthedifferenthands.
etal.,2008).SuchmaskshavebeenshownnottoinvokeNCEs
by object updating (Sumner, 2008) or by perceptual in-
5.1. Stimuliandtask teractions (Boy and Sumner, 2010), and thus any NCEs
observedcanbeattributedtomotorinhibition.
Eachtrialbeganwithpresentationofawhitefixationcrosson
amid-greybackground.Thiscrosssubtended1degree(cid:2)1de- 5.2. Proceduretodeterminethresholdforprime
greeof visual angle,and waspresentedin the centreofthe perception
screenfor500msec.Followingablankintervalof200msec,
theprimeappearedinthecentreofthescreenandremained Priortothemainexperiment,thedurationoftheprimewas
for50msec(seebelowforhowthisdurationwasdetermined). settobelowthethresholdforconsciousperception(50msec
Theprimewasthenreplacedwiththemaskwhichremained duration) using a psychophysical staircase procedure. Here,
onthescreenfor100msec.Twomask-targetSOAswereused oneachtrialaprimeandmaskwerepresentedwithnotarget,
inthisexperiment;20msec(shortSOA,whichwasexpected and the participant was instructed to make a 2-alternative
to produce a PCE) and 150 msec (long SOA, which was forced-choice button-press according to the direction of the
expectedtoproduceanNCE).SOAconditionswerepresented primestimulus.Theparticipantwasinstructedtomaketheir
inalternatingblocks,startingwithalongSOAblock.Patient best guess if they were unsure of prime direction, to
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004
cortex xxx (2013) 1e15 9
A B
Fig.4eStimulussequenceandrelevanttimingsinthemasked-primetask.Inthisexample,asprimeandtargetare
associatedwiththesameresponse,bothtrialsshownarecompatibletrials.Thepatientmadespeededsqueezeresponses
accordingtothedirectionofthetargetpresentedoneachtrial(alltrialtypeswereequiprobable).(A)Showsthetrial
sequenceforshortSOAblocks,and(B)showsthetrialsequenceforlongSOAblocks.Thetargetarrowheadsappeared
5degreesofvisualangleeithertotheleft,right,orabovethecentreofthescreen.Alltrialtypeswereequiprobable.Aleft
handsqueezeresponsetoleft-pointingtarget,primecompatibletrialsareshownhere.Notethatthemaskandtargetwere
eachpresentedforthesametotaltimeinbothshortandlongSOAconditions.
concentrateonbeingaccurate,andthatspeedwasunimpor- This method of analysis is essentially like the delta plot
tant for this part of the task. The prime duration began at method used to analyse the data in Experiment 1. Trials in
120msec,andthenwasvariedaccordingtoafixed-step,1-up/ whichnocorrectresponsewasdetectedwerereplacedwith
2-downprocedure:Aftertwoconsecutivecorrectresponsesto themeancorrectresponsetimeforthathand,condition,and
primespresentedatthesameduration,primedurationwas mask-targetSOA(thisisameanstokeepthetotalnumberof
reduced by 10 msec on the next trial; after an incorrect trialsthesameineachconditionanddividableby8,toavoid
response it was increased by 10 msec, within a range of problemsassociatedwithunequalbinsizes).Then,response
10e200 msec. This staircase procedure terminated after 10 timeswerere-calculatedrelativetotheprimeonset(weadded
“reversals”. The fastest prime duration was 60 msec (which theprime-targetSOAforeachtrialtotheRTforthattrial),and
waspresentedtwice,andtheprimewasincorrectlyidentified rank-orderedforeachhand(leftorright)foreachcondition
onthesecondpresentation),andthemeanprimedurationat (incompatible or compatible) across SOA conditions. This
the reversals was 84 msec. Thus, for the remainder of the meantthattherewassomere-shufflingofresponsesacross
experimenttheprimedurationwassetto50msec,whichwas SOAconditions(becauseaslowresponseonashortSOAtrial
the faster than the fastest prime duration measured during may have a longer prime-locked response time than a fast
thestaircase(andwasnotreliablyidentified),andfasterthan response on a long SOA trial). These prime-locked response
theaveragedurationofthereversals. times were divided into 8 bins of equal size. The average
compatibility effect (average incompatible RT (cid:3) average
5.3. Prime-lockedanalysisofRTs compatibleRT)foreachbinforeachhandwascalculated,and
plottedrelativetothemeanRTforthatbinandhand.Lastly,
We followed the method described in Schlaghecken et al. thestatisticalsignificanceofthecompatibilityeffectineach
(2011) to analyse RTs in a masked priming task relative to bin was determined by conducting a Bonferroni-corrected
prime onset rather than to target onset. Schlaghecken and unpairedt-testontheresponsetimesineachbin.
colleaguesnotedthattheincreasedtrial-to-trialvariabilityin
olderadults’RTsmayobscuretheprimingeffectsthatwould
be revealed by traditional analyses which separate target- 6. Results and discussion
locked RTs according to prime-target compatibility and
mask-target SOA on each trial. In fact, Schlaghecken et al. The results for the masked-prime experiment are shown in
(2011) showed that calculating RTs relative to prime onset Fig.5.TheunaffectedlefthandshowedthepatternofRTef-
could reveal a reliable NCE in older participants’ RTs when fects that would be expected from healthy individuals in
thesewerenotshownbytraditionalanalyses. amaskedprimingtask(e.g.,SchlagheckenandEimer,2002).
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004
10 cortex xxx (2013) 1e15
Schlaghecken et al. (2012) have recently suggested that
prime-locked distributional analyses like those performed
herecanproduce‘significant’effectsinsomelatencybinsby
chance.Herewedonotrelyonsearchingforsignificantbins,
butrathercomparethewholepatternbetweenthealienand
non-alienhands.Nevertheless,wehavealsotestedthepos-
sibilitythatthepatternshownbythealienhandcouldariseby
chance from a ‘healthy’ distribution of data. We pooled the
prime-locked RT data from the non-alien hand across com-
patibleandincompatibleconditionsandrandomlyre-labelled
trials as incompatible and compatible. We then re-ran the
distributional analyses described here. After repeating this
process100times,noneofthe100randomlyre-sampleddata
Fig.5eResultsfrommaskedprimingtask.Meanoctile setsshowedthesamereliablePCEin6/8RTbinsasshownby
compatibilityeffectsasafunctionofmeanRT(relativeto thealienhand(andonly3outof100showedareliablePCEin
primeonset)forthatoctile,calculatedseparatelyfortheleft anyof5/8bins).Thus,wesuggestthatitisveryunlikelythat
(greyline)andright(blackline)handsfollowingthe responsesfromPatientSA’salienhandactuallybelongtothe
proceduredescribedinSchlagheckenetal.(2011).Data same distribution as that of her non-alien hand, and only
fromthelongestandmostvariableRTbinareincludedfor showedaconsistentlysignificantPCEduetochance.
completeness,andareshownhereasdottedlinesand Thus,thereisnoevidenceofautomaticmotorinhibitionof
opensymbols.Theleft(non-alien)handshowsapatternof primed responses, indexed by the NCE, for responses made
compatibilityeffectswhichissimilartothatreported withthealienhand.Itisunlikelythatthisdisruptedinhibition
elsewhereinhealthycontrols(PCEfollowedbyNCE).The ismerelyduetoageornon-specificeffectsofdisease,because
right(alien)handshowsastrongPCE,andthereisno reliableinhibitionisshownforresponsesmadewiththeleft
evidencethatthisturnsnegativeastheprime-response (non-alien)hand.
intervalincreases.*DenotesBonferroni-correctedt-test
p<.05. 6.1. Spatialcongruencyeffects
The design of the masked priming experiment required the
targetstimulustobepresentedinadifferentlocationtothe
For fast responses (which occurred most quickly after the primeandmask(toavoidspatialandtemporaloverlappingof
prime was presented), RTs were faster for compatible trials stimuli in the short SOA condition). Thus, on each trial the
relative to incompatible trials (a PCE). For responses that targetwaspresentedtotheleft,totheright,orabovecentral
occurred later, responses were faster on incompatible trials fixation.Thisspatialaspectofthetargetstimulusmighthave
relativetocompatibletrials(aNCE).Thereissomeevidence affectedperformanceinamannersimilartotheSimoneffect
thattheprimingeffectmayhavereturnedtopositiveagainat (e.g.,LuandProctor,1995,forareview)andthespatialStroop
thetailendofthedistributioninbin8,whichisalsoconsistent effect (e.g., Banich et al., 2000). Thus, the design of this
withpreviousstudies(seee.g.,SumnerandBrandwood,2008), experimentprovidesanopportunitytoinvestigateanyeffect
butthis effectmay have beenskewed by outliersin thetail ofspatialcongruencyinPatientSA.
end of the distribution, and did not reach statistical signifi- Afterremovalofanyresponsewhichoccurredþ/(cid:3)3SDs
cance(Bonferroni-correctedp>.1). away from the mean of that condition, we calculated mean
A very different pattern emerged in the RTs for the re- RTs for each hand for spatially incongruent, neutral, and
sponses made with the alien hand. Here, responses were congruenttrials.ThesemeanRTsare showninFig.6A.The
consistentlyfasterforcompatibletrialsrelativetoincompat- expected location congruency effects were observed: re-
ible trials (a PCE), and there was no evidence of this effect sponseswerefastestwhenthetargetappearedinalocation
turningnegative(NCE),evenatlaterpointsinthedistribution. thatwascongruentwiththerequiredresponse,andslowest
WhilethereisasmalldifferenceintheoverallRTsfortheleft when the target appeared in a location that was congruent
and the right hands, responses made with the left hand with the response opposite that required to the target
showedasignificantNCEbyaround850msecaftertheprime {incongruent condition; [F(2, 627) ¼ 7.37, p ¼ .001]}. Also, as
hadonset,whereasright-handresponsesstillshowedasig- expected,responsesmadewiththeleft(non-alien)handwere
nificant PCE at 1050 msec.Thus, these distributions suggest significantlyfasterthanresponsesmadewiththeright(alien)
thatthis difference incompatibilityeffectisnotlikelyto be hand[F(1,627)¼51.12,p<.001].Importantly,theinteraction
duetotheslightlylongerrightthanlefthandresponses. betweentheeffectsofhandandcongruencydidnotapproach
Furthermore,we suggestthat it is unlikelythat the inhi- statisticalsignificance[F(2,627)<1].
bition has simply been delayed in right-hand responses. Asnotedabove,adeltaplotcanbeamoresensitivewayof
Mayloretal.(2011)reportedreliableNCEsforelderlypartici- examiningRTeffectsthancomparingaverageRTs.Therefore,
pantsforresponseswhichoccurredby500msecafterprime we have plotted the spatial congruency effect (incongruent
onset,whereasPatientSAhereshowedaprimingeffectthat RT(cid:3)congruentRT)over8RTbins(seeFig.6B)accordingtothe
wasstillpositiveforresponseswhichwererecordedaround procedures described above. The pattern of spatial congru-
1050msecaftertheprimehadonset. encyeffectswassimilarforbothhands,andtheeffectdidnot
Pleasecitethisarticleinpressas:McBrideJ,etal.,Exaggeratedobjectaffordanceandabsentautomaticinhibitioninalienhand
syndrome,Cortex(2013),http://dx.doi.org/10.1016/j.cortex.2013.01.004