perspectivetaking,fantasy,empathicconcern,andpersonaldistressDavis,199 - PDF document

perspectivetaking,fantasy,empathicconcern,andpersonaldistressDavis,1996ScanningmethodandprocedureSubjectstookpartinfoursequentialfMRIsessions.Eachsessionconsistedof10blocks,twoofeachofthefivefollowingconditionsinwhichsubjectshadtowatchandassess:(1)righthandsinpainfulsituations,(2)righthandsinneutralsituations,(3)rightfeetinpainfulsituations,(4)rightfeetinneutralsituations,or(5)baselinetrialsshowingstaticcrosses.Eachblockconsistedoffour8-strialsofthesamecondition(picture=3.5s,blankscreen=0.5s,ratingscale=3.0s,blankscreen=1.0s),andeachpicturewasfollowedbyavisualanalogueratingscalerangingfromNoPainWorstPossiblePain,exceptforthebaselinetrialswherethescalevalues.Inthefirstfourconditions,subjectswereinstructedtoratetheintensityofpaintheythoughtthepersonwouldfeelineachsituation.Inthebaselinetrials,subjectswereaskedtomovethecursorinordertoreproducetheintersectionofthetwolines,whichwasnotsymmetricalandvariedrandomlyintermsofthepositionatwhichtheverticalandhorizontallinescrossed.Thus,attheendofeachtrial,theyusedatwo-buttonresponseboxundertheirlefthandtomoveacursorhorizontallyonthevisualscale(index=left,middlefinger=right).Foreachtrial,thecursorwasplacedinthemiddleofthescalesothateverytrialineveryconditionrequiredmovingthecursoralongthescalebypressingandholdingdowneitheroftwokeys,therebycontrollingforthemotoroutputinvolvedintheratingprocessacrossallconditions.Subjectswereprovidedwithseveraltrainingtrialspriortothescanningsessionsinordertolearntousetheratingscaleandperformthetaskaccurately,andwithintheallottedtime.Thevisualanaloguescalesweresubsequentlydividedinto100equalintervalsforanalyses.Ablankscreenof3swasinsertedbetweeneachblockoftrials.Theorderofconditionswasrandomizedwithinahalfsession(i.e.,eachconditionwaspresentedoncebeforeanywererepeated).Nopicturewaspresentedmorethanoncethroughoutthewholeexperiment.Afterthescanningsessions,participantsweredebriefedabouthowtheyfeltduringtheexperiment,andaskedspecificquestionsconcerningwhatstrategytheyusedduringthetask.Theywerealsogivenascale(0–10)torate,ingeneral,theirownsensitivitytopain.DataacquisitionandanalysesMRIdatawereacquiredona3-Thead-onlySiemensMagnetomAllegraSystemequippedwithastandardquadratureheadcoil.Changesinbloodoxygenationlevel-dependent(BOLD)T2*weightedMRsignalweremeasuredusingagradientecho-planarimaging(EPI)sequence(repetitiontimeTR=2000ms,echotimeTE=30ms,FoV=192mm,flipangle80,6464matrix,32slices/slab,slicethickness4.5mm,nogap,voxelsize=3.04.5mm)Foreachscan,atotalof183EPIvolumeimageswereacquiredalongtheAC-PCplane.StructuralMRimageswereacquiredwithaMPRAGEsequence(TR=2500,TE=4.38,fov=256mm,flipangle=8,256256matrix,160slices/slab,slicethickness=1mm,nogap).ImageprocessingwascarriedoutusingSPM2(WellcomeDepartmentofImagingNeuroscience,London,UK),implementedinMATLAB6.1(MathworksInc.Sherborn,MA).ImageswererealignedandnormalizedusingstandardSPMprocedures.Thenormalizedimagesof22mmweresmoothedbyaFWHM6Gaussiankernel.Afirstfixedlevelofanalysiswascomputedsubject-wiseusingthegenerallinearmodelwithhemodynamicresponsefunctionmodeledasaboxcarfunctionwhoselengthcoveredthefoursuccessivepicturesofthesametype.First-levelcontrastswereintroducedinsecond-levelrandom-effectanalysistoallowforpopulationinferences.Maineffectswerecomputedusingone-sampletests,includingallsubjectsforeachofthecontrastsofinterest,whichyieldedastatisticalparametricmapofthestatistic(SPM),subsequentlytransformedtotheunitnormaldistribution(SPMZ).Avoxel-levelthresholdof0.0001uncorrectedformultiplecomparisons(=4.99),andacluster-levelspatialextentthresholdof0.05corrected,wereusedtoidentifypain-relatedregionsbasedonapriorihypotheses.Giventhatthisstudydidnotincludeanactualpaincondition,aregionofinterestanalysiswasconductedbytakingintoaccountpreviousneuroimagingstudiesthathaveexaminedbothselfpainexperienceandpainperceptioninothers.Specifically,regionsofinterestfortheanteriorcingulatecortexandanteriorinsulawerebasedonthestereotaxiccoordinatesfromSingeretal.(2004)inthePain–NopaininOtherscontrast(ACC:[0,27,33],[3,12,42];AnteriorInsula:[33,21,9],[39,12,3],[36,12,0]).Forthe Fig.1.Samplepicturesofhandsandfeetinpainful(Pain)andneutral(No-Pain)conditions.Notethatonlyrightlimbswerepresented,andforeachpastimulus,acorrespondingneutral(No-Pain)onewasprovided.P.L.Jacksonetal./NeuroImage24(2005)771–779 therighthemisphereforthehand(56,62,14)intheposteriortemporalcortexcorrespondingtotheMTregion,onlyinthecontrastsinvolvingpainfulstimuliversusnon-painfulones.Ourstudyinvestigatedthehemodynamicresponseduringtheperceptionofpaininothers,whichisawaytoaddresstheprocessinvolvedinempathy(DecetyandJackson,inpress;HodgesandWegner,1997;Ickes,2003).Here,weconsiderperceptionofpaininothersasasocialstimulusthattriggersaspecificmental(affective)stateintheperceiverfromwhichempathicprocessingmaystem.Notethatourintentionwasnottoinvestigateself-pain-processingassuch,ratherwewereinterestedinthehemodynamicchangesstemmingfromthesightofothersinpotentiallypainfulsituations.Theresultsdemonstratethatwatch-ingotherindividualsinpain-inducingsituationstriggersaspecificpartofaneuralnetworkknowntobeinvolvedinself-pain Fig.3.Clustersofbilateralactivationfoundinthelateraloccipito-temporalcortexcorrespondingtothebodyselectiveregionEBA;seeDowningetal.,2001ResultsfromthecontrastbetweenalltheconditionsdepictingrighthandsandrightfeetandthebaselineconditionaresuperimposedontheMNItempl Fig.2.(A)AnteriorinsularcortexAIC,thalamusandposteriorpartoftheanteriorcingulateACCactivationduringtheobservationandassessmentofsomeoneelseinpainfulsituationscontrastedwithneutral(No-Pain)situations.ResultsaresuperimposedontheMNIMRItemplate.(B)ACCclustersuperimposedontoasagittalsectionandscatterplotshowingthepositivecorrelationbetweentheindexedratingsandthelevelofactivityinthisregion=14,=20,=44.P.L.Jacksonetal./NeuroImage24(2005)771–779 Ourfirsthypothesisthatperceptionofhandsandfeetinpainfulsituationswouldbeassociatedwithspecificchangesinthesomatosensorycortices(SI–SII)wasnotconfirmed.Althoughtheabsenceofsignificanthemodynamicchangeinthesecorticalregionscouldberelatedtothespecificityofourdesign,itremainsconsistentwithtworecentstudiesthatexaminedbothpaininselfandinothers(Morrisonetal.,inpress;Singeretal.,2004).ThesestudiesdidnotreportanyactivationofSIorSIIinconditionsofpaininothers,eventhoughactivationsintheseregionswereobservedwhenthesamesubjectreceivedactualpain.Moreover,theroleoftheprimarysomatosensorycortexinpainperceptionisstilldebated,andseveralstudiesdidnotreportitscontribution(seeTable1ofBushnelletal.,1999;Peyronetal.,2000).Inaddition,thesomatosensorycortexismostoftenassociatedwithsensoryaspectsofpainratherthantheaffectiveaspects(etal.,1999;Craig,2003),andtheformeraspectislesslikelytobepresentsincetherewasnoactualnociceptivestimulation.Finally,onepossibleexplanationforthelackofinvolvementofSIandSIIcouldbethattheintensityordepthoftheinducedprocesswasnotsufficienttoprimethewholesensory-affectivepaincontinuum.Infact,involvementofMIinmotorimageryhasbeenfoundinconsistentlyintheliteratureprobablyforsimilarreasons`zesandDecety,2001;Jacksonetal.,2001).Thus,itispossiblethatanexperimentthatusesmoreshockingormoreintensestimuliwouldleadtoSIand/orSIIactivationduringobservationofpaininothers,butsuchadesignwouldalsotapintootherrelatedprocessessuchasdiscomfortandpersonalThemainfindingofthisstudyshowingactivationintheACCandintheanteriorinsuladuringtheperceptionandassessmentofsomeoneelse’spainisconsistentwithpreviousimagingstudiesofpainprocessingthathavedemonstratedtheirroleintheaffectiveaspectofpainprocessing(Coghilletal.,1999;Hofbaueretal.,2001;Ploghausetal.,1999;Rainvilleetal.,1997;Sawamotoetal.,2000),aswellaswithrecentfMRIstudiesofempathyforpain(Morrisonetal.,inpress;Singeretal.,2004).Infact,thepeaksofactivationintheACC[(8,26,40)and(10,18,44)]andanteriorinsula[(32,18,6)and(42,14,4)]forthePain–Nopaincontrastinthisstudyareveryclose,within1cm,tothosereportedbySingeretal.(2004)inthePain–NopainOtherscontrast[ACC:(0,27,33)and(3,12,42);anteriorinsula:(39,3)and(36,12,3)].Theseregionsareconsideredaskeycorticalareasinvolvedinregulatingthesubjectivefeelingsofpain-relatedunpleasantnessinhumans(Bushetal.,2000;Rain-ville,2002).Eventhoughthesubjectsinthisstudywereaskedtoratethelevelofpainintensityaftereachstimulus,theyhadtoextractthisvalueintheabsenceofitsrelatedsensationintheself.Interestingly,post-scanninginterviewsandquestionnairesindicatethatthesubjectsimaginedthelevelofpainthesituationwouldproducetotheotherperson,whichdrawsonaffectiveandevencognitive/evaluativeprocesses(Bushetal.,2000).FurthersupportfortheroleofACCintheaffectivedimensionofpainalsocomesfromarecentfMRIstudythatdemonstratedactivationofthisregion([8,16,44],[10,26,28])whenparticipantslistentoJapanesepain-evokingwordsascomparedtononsensesyllablesOsakaetal.,2004Furthermore,thestrongcorrelationbetweentheratingsandthelevelofactivitywithintheposteriorACC(seeFig.2supportsthepivotalroleofthisregionininterrelatingatten-tionalandevaluativefunctionsassociatedwithpain-evokingsituations(Price,2000).Ourresultssuggestthatsuchamechanismisalsoinvolvedintheevaluationofpaininothers,andsupporttheinterestingdiscoverybyHutchisonetal.(1999)whoidentifiedneuronsintheACCofneurologicalpatientsthatrespondedbothtopainfulstimulationandtotheanticipationortheobservationofthesamestimulationappliedtoanotherAnalternateinterpretationwouldbethattheperceptionandassessmentofpaininothersleadstoanunspecificstateofarousalsuchaspersonaldistressandanxiety(Critchley,2004;Eisenberg,).Insuchacase,however,changesinactivityshouldbeobservednotonlyintheACCandanteriorinsulabutalsoinemotion-relatedsystems,notablytheamygdala.Indeed,anumberofstudiesofnegativeemotionssuggestthatdistressisrelatedtoactivityintheamygdala(e.g.,Irwinetal.,1996;see2002;PosnerandRothbart,1998forreviews).Interestingly,arecentreviewhasarguedthattheamygdalacould,however,playaroleinpersistentpain(Neugebaueretal.,2004).Noneofthesecomponents(distressandpersistentpain)wereelicitedbyourAnothercomplementaryinterpretationofourresultsisthatwatchingpainfulstimuliinsuchdailylivingcontextspromptsanticipatorymechanisms.Severalneuroimagingstudieshaveindeeddemonstratedthatanticipationofpainfulstimulibeingadministeredtotheselfincreasesthehemodynamicsignalinpain-relatedneuralregions(Peyronetal.,1999;Ploghausetal.,1999;Porroetal.,2002,2003;Sawamotoetal.,2000However,inourstudy,participantswerenotinflictedpainnorweretheyledtobelievethattheycouldreceiveanociceptivestimulusduringthecourseofourexperiment.Nevertheless,onecannotexcludethatsuchamechanismisinvolvedbecauseitmaybearguedthatwatchingpaininotherspromptsanticipationofpaininoneself.Thesetwointerpretationsarenotmutuallyexclusiveinthelightofthesharedrepresentationmodel,consideringthatanticipatorymechanismsarecrucialforone’sownsurvival.Otherresultsalsosuggestthatthefeelingofpainisnotrestrictedtoitsphysicalsensation,butoccurswithintheindividualasaresultofobservinganother’semotionalstate.Thisresultfitswellwithrecentfindingsthatthereisaneuralrealizationoftheideathatsocialrelationshipscansometimesbe.ThislatteraspectwasdemonstratedinanfMRIstudyshowingthattheneuralcircuitinvolvedinpainprocessing,includingtheanteriorinsulaandtheACC,wasactivatedwhentheparticipantsweresociallyexcludedfromanon-linecomputergame(Eisenbergeretal.,2003).Interestingly,theACCwasmoreactiveduringexclusionanditsactivitycorrelatedpositivelywithself-reporteddistress.Theauthorsarguedthatsocialpainisanalogousinitsneurocognitivefunctiontophysicalpain.ContrarytothestudybySingeretal.(2004),wedidnotfindanysignificantcorrelationbetweentheempathyquestionnaireandthehemodynamicchanges.Moreover,nocorrelationwasfoundbetweenself-reportofpainsensitivityandpainintensityratings.Theseresultsmaynotbethatsurprisingconsideringthatselfmeasuresofempathyarepoorpredictorsofactualempathicbehavior(DavisandKraus,1997RepresentationofbodypartsNospecificactivationwasdetectedinassociationwiththevisualperceptionofhandsandfeetinthesomatosensoryandP.L.Jacksonetal./NeuroImage24(2005)771–779 motor/premotorcortex.ThisdoesnotsupportthesomatotopicpredictionthatwasmadebasedonanfMRIstudythatshowedinvolvementofdifferentialpremotorandparietalsomatosensoryareaswhensubjectsobservedobject-relatedactionsmadewithdifferenteffectorsincludinghandandfoot(Buccinoetal.,2001However,theneuronsexhibitingmirrorpropertieshavebeenmainlydiscoveredinmonkeysandhumansduringobservationofgoaldirectedactions,andnotduringnon-directedactionswhenwatchingstaticpictures(Rizzolattietal.,2001),aswereusedinthecurrentstudy.Inaddition,ourstimulidepictedactionsforwhichthesubjectswereactedupon,notacting.Thismayrepresentanimportantfunctionaldifferenceinthewaymirrorneuronsaretriggered,and,ifso,itconstrainstheirinvolvementinmanyeverydayempathicsituations.ActivationofareaMTisconsistentwithitsinvolvementinimpliedorimaginedvisualmotion(Stevensetal.,2000).AnfMRIstudybyKourtziandKanwisher(2000)foundstrongeractivationinMTduringviewingofstaticphotographswithimpliedmotion(e.g.,abasketballplayerabouttoshoottheball)comparedtoviewingphotographswithoutimpliedmotion(e.g.,apersonsittinginachair).Itispossiblethatthepainfulphotographsinthisexperimentimplymotiontotheobserverbecauseeachpainfuleventislikelycausedbythemotionofthebodytowardanobjectortheopposite(e.g.,doorclosingonaItiswellrecognizedthatvisualstimuliareprocessedinspecializedcorticalareas(Allisonetal.,2000),andmorespecifically,thereisaregionintheoccipito-temporalcortexthatrespondsselectivelytoimagesofhumanbodiesandbodypartsDowningetal.,2001).Examinationoftheconditionsinvolvingbodypartsversusthebaselineconditionrevealedactivationofseveralclustersintheoccipitallobeinbothhemispheres,aswellasthemedialprefrontalandlateralobitofrontalcortex.Notably,abilateralactivatedfocuswasfoundintheoccipito-temporalregion([56,70,4],[56,68,2]).Thisfitsverywellwiththefindingthatsomeneuronsintheposteriortemporalcortexrespondselectivelytothevisualappearanceofthebody.Forexample,electrophysiologicalrecordingsbyJellemaetal.(2002)haveidentifiedneuronsinthesuperiortemporalsulcusofthemonkeybrainthatdischargeselectivelyatthesightofthebody.Recently,Downingetal.(2001)haveextendedthisfindingbydiscoveringabody-selectiveregioninthelateraloccipito-temporalcortex,whichproducedasignificantlystrongerresponsewhensubjectsviewedstillphotographsofhumanbodiesandbodypartsthanwhentheyviewedvariousinanimateobjects.AssuggestedbyDowningetal.(2001),thisregionmightnotbeexclusivetovisualstimulibutcouldrelaygeneralamodalsemanticknowledgeaboutthebody(Chaminade,Meltzoff,andDecety,2004).Moreover,thisregion(intheposteriorSTS)hasreciprocalconnectionwiththeamygdalaandorbitofrontalcortex,andispartofacircuitinvolvedintheelaborationoftheaffectiveaspectsofsocialbehavior(Adolphs,2003;PuceandPerrett,Oneoftheevolutionarybenefitsofsharedneuralrepresen-tationsforselfandotheristhattheycanbeusedtolearnfromandtounderstandothers.Theobservationofpositiveexperi-encesinothersmayhaveareinforcingvalue.Conversely,throughwatchingnegativeconsequencesofotherpeople’sbehavior,individualslearntoavoidsituationsthatarepoten-tiallyhazardousandlikelytoinjurethemselves,withouthavingtoexperiencethem.Here,weinvestigatedtheneuralresponseelicitedbytheassessmentofpainfulsituationsexperiencedbyothersasameansofexploringthisimportantaspectofinterpersonalbehavior.Ourresultsdemonstratethattheanteriorcingulateandanteriorinsulacortices,regionsoftenreportedasbeingpartofthepainaffectivesystem,arerecruitedwhenwatchingsomeoneelse’spain.Thesefindingsofferoneplausibleexplanationofhowoneisaffectedbyanotherperson’sstateandfeelings.ThisresearchwassupportedbytheInstituteforLearningandBrainSciences,andagrantfromNIH(HD-22514)toANM,aswellasaresearchfellowshipfromtheCanadianInstitutesofHealthResearchawardedtoDr.PhilipL.Jackson.ThefMRIexperimentwasconductedattheLewisCenterforNeuroimaging,UniversityofOregon,Eugene(OR),wherethestaffatthisfacilitycontributedtorunningthisexperimentsmoothly.CorrespondenceshouldbeaddressedtoProf.JeanDecety(decety@u.washington.edu).Adolphs,R.,2003.Cognitiveneuroscienceofhumansocialbehavior.Nat.Rev.,Neurosci.4,165–178.Adolphs,R.,Damasio,H.,Tranel,D.,Cooper,G.,Damasio,A.R.,2000.Aroleforsomatosensorycorticesinthevisualrecognitionofemotionasrevealedbythree-dimensionallesionmapping.J.Neurosci.20,2683–2690.Allison,T.,Puce,A.,McCarty,G.,2000.Socialperceptionfromvisualcues:roleoftheSTSregion.TICS4,267–278.Barsalou,L.W.,Niedenthal,P.M.,Barbey,A.,Ruppert,J.,2003.Socialembodiment.In:Ross,B.H.(Ed.),ThePsychologyofLearningandMotivation,vol.43.AcademicPress,SanDiego,CA,pp.43–92.Batson,C.D.,1991.TheAltruismQuestion:TowardaSocial-PsychologicalAnswer.Erlbaum,Hillsdale,NJ.Batson,C.D.,1997.Self-othermergingandtheempathyaltruismhypothesis:replytoNeubergetal.J.Pers.Soc.Psychol.73,517–522.Brothers,L.,1989.Abiologicalperspectiveonempathy.Am.J.Psych.146,10–19.Buccino,G.,Binkofski,F.,Fink,G.R.,Fadiga,L.,Fogassi,L.,Gallese,V.,Seitz,R.J.,Zilles,K.,Rizzolatti,G.,Freund,H.J.,2001.Actionobservationactivatespremotorandparietalareasinasomatotopicmanner:anfMRIstudy.Eur.J.Neurosci.13,400–404.Bush,G.,Luu,P.,Posner,M.I.,2000.Cognitiveandemotionalinfluencesinanteriorcingulatecortex.TICS4,215–222.Bushnell,M.C.,Duncan,G.H.,Hofbauer,R.K.,Ha,B.,Chen,J.-I.,Carrier,B.,1999.Painperception:istherearoleforprimarysomatosensorycortex?Proc.Natl.Acad.Sci.96,7705–7709.Carr,L.,Iacoboni,M.,Dubeau,M.C.,Mazziotta,J.C.,Lenzi,G.L.,2003.Neuralmechanismsofempathyinhumans:arelayfromneuralsystemsforimitationtolimbicareas.Proc.Natl.Acad.Sci.100,5497–5502.Chaminade,T.,Meltzoff,A.N.,Decety,J.,2004.AnfMRIstudyofimitation:actionrepresentationandbodyschema.Neuropsychologia(inpress).Coghill,R.C.,Sang,C.N.,Maisog,J.M.,Iadarola,M.,1999.Painintensityprocessingwithinthehumanbrain:abilateral,distributedmechanism.J.Neurophysiol.82,1934–1943.Craig,A.D.,2003.Anewviewofpainasahomeostaticemotion.TINS26,303–307.P.L.Jacksonetal./NeuroImage24(2005)771–779 Puce,A.,Perrett,D.,2003.Electrophysiologyandbrainimagingofbiologicalmotion.Philos.Trans.R.Soc.Lond.,B358,435–445.Rainville,P.,2002.Brainmechanismsofpainaffectandpainmodulation.Curr.Opin.Neurobiol.12,195–204.Rainville,P.,Duncan,G.H.,Price,D.D.,Carrier,B.,Bushnell,M.C.,1997.Painaffectencodedinhumananteriorcingulatebutnotinsomato-sensorycortex.Science277,968–971.Rizzolatti,G.,Fogassi,L.,Gallese,V.,2001.Neurophysiologicalmecha-nismsunderlyingtheunderstandingandimitationofaction.Nat.Rev.Neurosci.2,661–670.Rochat,P.,Striano,T.,2000.Perceivedselfininfancy.InfantBehav.Dev.23,513–530.Sawamoto,N.,Honda,M.,Okado,T.,Hanakawa,T.,Kanda,M.,Fukuyama,H.,Konishi,J.,Shibasaki,H.,2000.Expectationofpainenhancesresponsestononpainfulsomatosensorystimulationintheanteriorcingulatecortexandparietaloperculum/posteriorinsula:anevent-relatedfunctionalmagneticresonanceimagingstudy.J.Neurosci.20,7438–7Singer,T.,Seymour,B.,O’Doherty,J.,Kaube,H.,Dolan,R.J.,Frith,C.D.,2004.Empathyforpaininvolvestheaffectivebutnotsensorycomponentsofpain.Science303,1157–1162.Stevens,J.A.,Fonlupt,P.,Shiffrar,M.,Decety,J.,2000.Newaspectsofmotionperception:selectiveneuralencodingofapparenthumanmovements.NeuroReport11,109–115.Treede,R.D.,Kenshalo,D.R.,Gracely,R.H.,Jones,A.K.P.,1999.Thecorticalrepresentationofpain.Pain79,105–111.Trevarthen,C.,1979.Communicationandcooperationinearlyinfancy.In:Bullowa,M.(Ed.),BeforeSpeech:TheBeginningofInterpersonalCommunication.CambridgeUniv.Press,London,pp.321–347.Wallbott,H.G.,1991.Recognitionofemotionfromfacialexpressionviaimitation?Someindirectevidenceforanoldtheory.Br.J.Soc.Psychol.30,207–219.Wicker,B.,Keysers,C.,Plailly,J.,Royet,J.P.,Gallese,V.,Rizzolatti,G.,2003.Bothofusdisgustedininsula:thecommonneuralbasisofseeingandfeelingdisgust.Neuron40,655–664.P.L.Jacksonetal./NeuroImage24(2005)771–779