inanactivity(e.g.,Geen,1968;Josephson,1988;Rule&Percival,Researchersha - PDF document

inanactivity(e.g.,Geen,1968;Josephson,1988;Rule&Percival,Researchershaveoperationalizedsituationsinanumberofways.Someconditionsarefairlyneutral,suchasthosethatinvolveaconfederatewhosilentlyobservesaparticipantcomplet-ingapuzzle(Carver&Glass,1978)oraconfederatewhoprovidesarelativelyneutralevaluationofaparticipant(Caprara&Renzi,1981).Otherneutralconditionsinvolveaparticipantreceivingpositivefeedbackfromanexperimenter(Caprara&Renzi,1981)oraconfederate(Bushman&Baumeister,1998).Finally,inafewconditionsarecomparedwithconditions.Participantsinthelatterconditionsreceivelowlevelsofanoxiousstimulus,and,assuch,theseconditionsaremeanttobecomparativelyneutral.Thus,theconditionsthatwewereabletocodeasneutralwereactuallyquitevariedinvalence.Nevertheless,becausetheseconditionsweremeanttobecomparedwithprovokingconditions,wetreatedthemasneutralconditions.Althoughweacknowledgethatnotalloftheconditionsinthestudiesaretrulyneutral,inthesectionsthatfollowweusethetermasameansforsimplifyingourterminology.TheoriesofAggressiveBehaviorAvarietyofcontemporarytheoriesarerelevantforunderstand-ingassociationsbetweenpersonalityandaggressivebehavior(e.g.,Anderson,Anderson,&Deuser,1996;Anderson&Bushman,2002;Berkowitz,1993;Caprara,Regalia,&Bandura,2002;Coie&Dodge,2000;Crick&Dodge,1994;Geen,1990;Huesmann,1998).ProminentamongtheseareCrickandDodge’s(1994)andHuesmann’s(1998)developmentalmodelsofantisocialandag-gressivebehavioraswellasAndersonandBushman’s(2002)generalaggressionmodel,allofwhichhavebeeninfluencedbyBerkowitiz’s(1984)andBandura’s(1973)earliertheorizing.ItisimportanttonotethatAndersonandBushman’s(2002)modelpersonfactorsaspredictorsofaggression,withtheac-knowledgmentthat“certaintraitspredisposeindividualstohighlevelsofaggression”(p.35).Nevertheless,noneofthesetheoriesprovidesaframeworkforunderstandinghowandwhyparticularpersonalityvariablespredictaggressivebehavior.However,theo-ristssuggestasetofunderlyingvariablesthatarelikelytobethemechanismsthroughwhichpersonalityvariablesinfluenceaggres-sivebehavior.Thesevariablesincludecognitiveprocessing,neg-ativeaffect,self-regulation,andsocial-informationprocessing.Peoplewhoareparticularlylikelytoengageinaggressivebehaviorhavemoreelaborateandreadilyaccessibleaggression-relatedcognitions(Anderson&Bushman,2002;Berkowitz,1983,1993;Huesmann,1988).Dodge(2002)underscoredthatchildrenacquiretheseaggressivecognitionsthroughearlyexperiencesandsocialization.AndersonandBushman(2002)suggestedthatthedevelopmentofaggression-relatedknowledgestructurescanshapeanindividual’spersonalityand,thus,influencethelikelihoodthattheindividualwillengageinaggressivebehavior.Onecouldarguethatanindividual’spersonalitymayfurtherbiasthewaysheorsheinterpretsinformation,which,inturn,mayguideaggressiveCognitionandnegativeaffect,oranger,are“inextricablylinked,”Huesmann(1998,p.98)argued.Inasimilarvein,Ander-sonetal.(1996)suggestedthatindividualdifferencesandsitua-tionalvariablesmayinteractby“traversingcognitiveandaffectivepathwaysbelievedtoinfluencethelikelihoodofaggression”(p.367).Negativeaffectresultsfromthecognitiveevaluationthatanexternalstimulusisprovoking(Huesmann,1998).Itislikelythatpersonalityvariablesthataremarkedbythepropensitytoexperi-encenegativeaffectortoperceivesituationsasprovokingwillbeassociatedwithhigherlevelsofaggressivebehavior.Mosttheoriesofaggressionlargelyignoretherolethatself-regulationplaysinaggressivebehavior.Recently,however,Ca-praraetal.(2002)pointedtotheimportanceofself-regulationinunderstandingaggressivebehavior;theirresearchshowsthatalackofself-regulatoryefficacyisassociatedwithincreasedvio-lence.Regulationfailureischaracterizedbythetendencytoactimpulsivelyorthetendencytoreacttosituationswithoutsufficientthoughtaboutfutureconsequences(e.g.,Barratt,1994).Personswhohavedifficultywithself-regulationareunlikelytobeabletoinhibiturgestobehaveaggressively.Assuch,personalityvariablescharacterizedbyself-regulationfailurearelikelytobeassociatedwithgreaterlevelsofaggressivebehavior.Intheirtheoreticalmodel,Dodgeandcolleagues(e.g.,Coie&Dodge,2000;Crick&Dodge,1994)havearticulatedtherolethatongoingsocial-informationprocessingplaysindirectingaggres-sivebehavior.AccordingtoDodge(2002),social-informationprocessingnotonlyincludescognitiveprocessing(e.g.,attention,perception,andmentalrepresentation)andaffectiveexperiencesbutalsoinvolves“thesettingofgoalsforrespondingwithinthesocialsituation,accessingofoneormorepossiblebehavioralresponses,evaluatingtheaccessedbehavioralresponses,andse-lectingoneforenactment,andthentranslatingadesiretoperformanactionintobehavior”(p.225).Forexample,DodgeandCoie(1987;Dodge,Lochman,Harnish,Bates,&Pettit,1997)haveshownthathostileattributionbiasisonesocial-informationpro-cessingmechanismthatisparticularlypredictiveofsometypesofDodgeandCoie(1987;Dodgeetal.,1997)theorizedthatdistinctwayspeopleprocesssocialinformationanddifferencesinpeople’ssalientsocialgoalsmutuallyinfluencethelikelihoodthattheywillengageinoneoftwotypesofaggression.Theauthorsdrewadistinctionbetweenreactiveandproactiveaggressionstyles,whichareessentiallysynonymouswithhostileandinstru-mentalaggressionsubclassifications(Berkowitz,1993;Hartup,1974).DodgeandCoiedefinedreactiveaggressionashostileresponsestoperceivedthreatorprovocation.Theyexplainedthat“perceptionsofthreatandexperiencesofangerthereactively Fortheneutralconditions,weexaminedthemeanpersonality-differenceeffectsizesaccordingtothetypeof“relativelyneutral”condi-tionusedasacomparisonforaprovocationcondition(trulyneutral,lowprovocation,orpositivecomparisonwithaprovokingcondition).Theresultsofthisanalysisshowedthatthemagnitudesofthemeanpersonality-differenceeffectsizeswereunaffectedbythetypeofneutralcomparisonAlthoughZelliandDodge(1999)conceptualizedhostileattributionsasapersonality-likecharacteristic,nostudiesexaminingtherelationbetweenhostileattributionbiasandaggressivebehaviorwereincludedinthecurrentmeta-analysisbecausethesetypicallyuseself-reportorscenarioandrole-playingmeasuresofaggressivebehavior.Also,theindividualdifferencesofhostilemasculinity,hypermasculinity,andgenderrolemas-culinity(Anderson,1997;Datlow,1999;Kogut,Langley,&O’Neal,1992;Lohr,1996;LoPresto&Deluty,1988;Malamuth,1988;Norris,1999)wereexcludedfromanalysisbecausetheseconstructsonlyfocusonmale-on-femaleaggression.PERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR participantswhoendorsedmoreAggressiveHostilityfactorswerecharacterizedbylowserotoninactivity.Takentogether,thisevidencesuggeststhataggression-relatedconstructsmaybedividedintotwomainfactors.TheangryfactorofaggressionappearstobepositivelyrelatedtoNeuroticism;thistypeofaggressionmaybesimilartoreactiveaggression.Assuch,Neuroticismmaybeparticularlylikelytobepositivelyrelatedtoaggressivebehavioronlyunderprovocation.Becauseithasbeenlinkedtocold-bloodedaggression,whichisnotnecessarilypre-cipitatedbyprovocation,Antagonism(i.e.,lowAgreeableness)maybepositivelyassociatedwithaggressivebehaviorunderneu-tralconditionsaswellasprovocationconditions.Despitethesuggestivefindingsfromthisresearchonpersonal-itydimensions,thestudiesarelimitedinacoupleofways.First,nonehasexaminedwhetherthesepersonalitydimensionspredictdifferentpatternsofactualaggressivebehavior.Instead,thestud-ieshaveprimarilyreliedonself-reportsofaggression.Indeed,afewstudieshavereportedcorrelationsbetweenparticipants’re-sponsestoassessmentsofthedimensionsofpersonalityandtheirresponsestotheBussandPerry(1992)AggressionQuestionnaire,thelatterofwhichisconsideredanassessmentofapersonalityvariable,namely,traitaggressiveness.Oneinterpretationofthesecorrelationalfindingsisthat,atbest,theydemonstrateconstructvalidityfortheAgreeablenessandNeuroticismdimensions.Also,inthesestudies,researchershavefailedtocompareaggressivebehaviorunderprovokingsituationsandneutralsituations.Indoingso,theyhaveneglectedtotakeintoaccountthewayspersonalitydimensionsmayinteractwithprovocationtopredictaggressivebehavior.PersonalityVariablesintheCurrentMeta-AnalysisResearchonthefive-factormodelsuggeststhattheAgreeable-nessandNeuroticismdimensionsmaypredictdifferentpatternsofaggressivebehaviorunderneutralandprovokingconditions.Aswehavenoted,however,little,ifany,empiricalliteratureconfirmsrelationsbetweenpersonalitydimensionsandpatternsofaggres-sivebehavior.Becausestudiesinourmeta-analysishaveassessedpersonalityvariables,manipulatedprovocationlevels,andmea-suredaggressivebehavior,themeta-analysishasthepotentialtoisolatepatternsofrelationsbetweenpersonalityvariablesandaggressivebehavior.Asmentionedpreviously,thepersonalityvariablesavailableforthemeta-analysisweredissipation–rumination,emotionalsusceptibility,impulsivity,narcissism,traitaggressiveness,traitanger,traitirritability,andTypeApersonal-Aswedescribeinthefollowingsubsections,empiricalresearchrevealsthatmostofthesepersonalityvariablesarerelatedtotheAgreeablenessandNeuroticismdimensions.Table1in-cludesthemeasuresofthepersonalityvariablesusedinthestudiesincludedinthecurrentmeta-analysis,theauthorsofthemeasures,thenumberofitems,andthetypeofscaleusedaswellasanexampleitemfromoneofthemeasures.Table2displaysthevariablesandtheircorrelations,whichwederivedfromthepub-lishedliterature.Unfortunately,wewereunabletoobtaineverypossiblecorrelation—inparticular,manyfordissipation–ruminationandTypeApersonalitywereunavailable.Ingeneral,thesecorrelationsrevealhighandpositiveassociationsamongsomeofthevariables.Othercorrelations,however,reveallowandsometimesnegativeinterrelations.Acknowledgingtheoverlapamongsomeofthesevariables,inourprimaryanalyses,weanalyzeandexplaintheresultsforeachpersonalityvariablesep-arately.Weadoptedthisapproachbecausewewantedtodeterminewhethereachvariablewasassociatedwithaggressivebehavioronlyinresponsetoprovocationoracrossconditions.TraitAggressivenessBussandPerry(1992;alsoseeAnderson&Bushman,2001;Berkowitz,1993)definedtraitaggressivenessasapropensitytoengageinphysicalandverbalaggression,toholdhostilecogni-tions,andtoexpressanger.Tiedens(2001)theorizedthatthetendencyforthosehighintraitaggressivenesstomakehostileattributionsmayincreaseangerandcreateaviciouscycleofhostilityandnegativeaffect.Afewstudies(Caprara,Barbaranelli,&Zimbardo,1996;R.Martinetal.,2000;Ruiz,Smith,&Rhode-walt,2001)revealednegativecorrelationsbetweentraitaggres-sivenessandtheAgreeablenessdimension.Capraraetal.(1996)andRuizetal.(2001)reportedpositivecorrelationsbetweenNeuroticismandtraitaggressiveness,butMartinetal.reportedasmallyetsignificantnegativerelationbetweenthetwoconstructs.Empiricalresearchshowsthat,comparedwithindividualswhoarelowintraitaggressiveness,thosehighintraitaggressivenessengageinhigherlevelsofaggressivebehaviorunderbothneutralandprovokingconditions(e.g.,Bushman,1995;Giancola&Zeichner,1995a,1995b;Hammock&Richardson,1992;Knott,1970;Larsen,Coleman,Forbes,&Johnson,1972;Scheier,Buss,&Buss,1978;Wingrove&Bond,1998;Zeichner,Frey,Parrott,&Butryn,1999).Forexample,inonestudy(Bushman,1995),par-ticipantsplayedacompetitivereactiontimegamewithafictitiousopponent;participantsreceivednoxiousnoiseafterlosingatrialandwereabletoadministernoxiousnoiseafterwinningatrial.Theresultsshowedthat,regardlessofprovocationlevel,partici-pantswhowerehighintraitaggressivenessadministeredhigherlevelsofnoxiousnoisethanthosewhowerelowintraitaggres-siveness.Despitethisgeneralpattern,afewstudieshavenotrevealedareliablerelationbetweentraitaggressivenessandag-gressivebehaviorundereitherprovokingornonprovokingcondi-tions(Bailey&Taylor,1991;Shondrick,1996).Nevertheless,themajorityofthefindingssuggestthatpeoplewhoarehighintraitaggressivenessbehavemoreaggressivelythanthosewhoarelowintraitaggressivenessandthatthisdifferenceisobservedevenwhenconditionsarerelativelyneutral.TraitIrritabilityThedefinitionofincludesbeingangrier,ingeneral,andtakingoffensetotheslightestprovocationaswellasthe Forsensationseeking,therewereonlytwoavailableeffectsizes(Cheong&Nagoshi,1999;Giancola&Zeichner,1995b),andeachwasderivedfromaprovocationcondition(noneforneutralconditions).Theaverageeffectsizesderivedfromthefixedandtherandomanalysesforsensationseekingwerepositive(2;fixedmean0.37;randommean0.33),buttheconfidenceintervals(CIs;fixedCI0.06,0.79;randomCI0.69,1.35)suggestedthatneitherwasgreaterthanzero.Thesefindingssuggestthattheassociationbetweensensationseekingandaggressivebehaviormaybeweak.Becausewewereinterestedincompar-ingthepersonality-differenceeffectsizeforneutralandprovokingcondi-tionsinthefollowinganalysissections,wedidnotincludesensationseekinginanyfurtheranalyses.PERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR propensitytobeoffensiveintheuseofaggressivebehavior(Ca-prara,1982;Caprara,Renzi,etal.,1986;Caprara,Renzi,Alcini,D’Imperio,&Travaglia,1983).Conceptually(Caprara&Renzi,1981)andempirically(M.D.Wood,1996),traitirritabilityisrelatedtotraitaggressiveness.InBussandDurkee’s(1957)initialassessmentoftraitaggressiveness(i.e.,Buss–DurkeeHostilityInventory),theyincludedanirritabilitysubscaleinthemeasure.BussandPerry’s(1992)updatedmeasureoftraitaggressiveness(AggressionQuestionnaire)omitstheirritabilitysubscale,butitneverthelessretainssomeoftheitemsfromthesubscale.Intheirresearchonaggressivebehavior,CapraraandRenzi(1981;Ca-praraetal.,1983,1987)examinedtraitirritabilityseparatelyfromtraitaggressiveness.BecauseBussandDurkee(1957;Buss&Perry,1992)treatedtraitirritabilityasaconstructwithintraitaggressionbutCapraraetal.distinguishedthetwo,itwasunclearwhethertheresultsfortraitaggressivenessandtraitirritabilityshouldbecombined.Inouranalysis,wefirstexaminedthemseparatelybutfoundthattheirrespectiveresultswereessentiallythesame;therefore,wecombinetheminsubsequentanalyses.Atleastonestudy(Caprara,Barbaranelli,&Zimbardo,1996)hasexaminedtherelationbetweentraitirritabilityandarevised,briefmeasureofthefivepersonalitydimensions.CorrelationalanalysessuggestthatirritabilityisnegativelyrelatedtoAgreeablenessandpositivelyrelatedtoNeuroticism.Consistentwiththefindingsfortraitaggressiveness,studyre-sults(Andersonetal.,2004;Lindsay,1999;Renzi,Caprara,Crudele,Galante,&Giannone,1984)fortraitirritabilityshowthat,evenwhensituationsarerelativelyneutral,individualswhoarehighintraitirritabilityengageinhigherlevelsofaggressivebehaviorthanthosewhoarelowintraitirritability.Forexample,inRenzietal.’s(1984)study,participantswereselectedonthebasisoftheirirritabilityscores,andtheexperimenterprovidedeitherpositiveordisparagingfeedback.Next,participantsinter-actedonacooperativetaskwithaconfederate.Theparticipantindicatedtheconfederate’sincorrectanswerswithwhatwasos-tensiblyelectricshock.Amaineffectoftraitirritabilitywasfoundonlevelsofshock,butnointeractionbetweentraitirritabilityandprovocationwasrevealed.Nevertheless,severalstudieshaveshownaninteractionbetweenirritabilityandlevelofprovocationonaggressivebehavior(Caprara&Renzi,1981;Capraraetal.,1983;Caprara,Renzi,Amolini,D’Imperio,&Travaglia,1984;Caprara,Barbaranelli,&Comrey,1992).Inthesestudies,highlevelsofirritabilitywereassociatedwithgreaterlevelsofaggres-sivebehaviorunderbothneutralandprovokingconditions,butthemagnitudeofthisassociationwaslargerundertheprovokingconditions.Insummary,theliteraturerevealsapositiverelationbetweentraitirritabilityandaggressivebehaviorunderbothneu-tralandprovokingconditions.TraitAngerTraitangerhasbeendefinedasthetendencyforsomeindivid-ualstofeelangermoreintensely,moreoften,andforalongerperiodoftimethanothers(Deffenbacheretal.,1996).Also,peoplewhoarehighintraitangerarepredisposedtowardrespondingangrilywhentheyareunfairlycriticized,treatedunjustly,ortreatedbadly(vanGoozen,Frijda,&vandePoll,1994;Spiel-berger,Jacobs,etal.,1983).vanGoozen,Frijda,andvandePoll(1994)notedthatpeoplewhoarehighintraitangerfocusonatargetthattheyseeasblameworthyandacttocorrecttheprovok-ingaction;theycandothisineitherconstructive(e.g.,assertive)ordestructive(e.g.,aggressive)ways.TraitangerispositivelycorrelatedwithNeuroticismandAntagonism(i.e.,lowAgreeable-ness),butthecorrelationwithNeuroticismappearstobelarger(Capraraetal.,1996;R.Martinetal.,2000).Asistruewiththeconstructofirritability,theconstructoftraitangeroverlapswithtraitaggressiveness.Inparticular,BussandPerry’s(1992)AggressionQuestionnaireincludesanAngersub-scaleasoneofitscomponents.Thus,thereislikelytobeconcep-tualaswellascontentoverlapbetweentraitaggressivenessandtraitanger.Forourcurrentpurposes,wetreatedthesealbeitcorrelatedconstructsasseparateindexes.Webelievedthat,whereastraitaggressivenessmayincludeapropensitytoexperi-enceanger,traitangeritselfisnotsynonymouswithtraitTable2CorrelationsAmongthePersonalityVariablesIncludedintheMeta-Analysis Variable123456781.Agg—2.Irr.30*–.77*—3.Anger.16*–.48*.57*—4.TypeA.01–.52*.02–.48*.08*–.42*—5.DisRum.25*–.38*.63*.30*6.EmoSus.42*–.68*.59*.43*.30*—7.Imp.07*–.39*.32*–.38*.01–.22*.16*–.52*.20*—8.Narc.09*–.50*.16*.13*–.37*.48*–.50*.07*—Thefollowingstudiesareincludedinthiscorrelationtable:Bartholowetal.(2005);Bermanetal.(1998);Booth-KewleyandFriedman(1987);BussandDurkee(1957);BussandPerry(1992);Byrne(1996);Caprara,Barbaranelli,andComrey(1992);Capraraetal.(1996);Caprara,Manzi,andPerugin(1992);CapraraandPastorelli(1993);Capraraetal.(1994);deFloresandValdes(1986);Dill(1999);Emmons(1981);Frantz(1986);Fukunishietal(1996);Furnham(1984);Garcia-Leonetal.(2002);HartandJoubert(1996);Heaven(1989);Joiremanetal.(2003);KokkonenandPulkkinen(1999);McCannetal.(1987);Myrtek(1995);Netteretal.(1998);RaskinandTerry(1988);RhodewaltandMorf(1995);Smith(1984);Stanfordetal.(1995);Watkinsetal.(1992);Willsetal.(1994);Wood(1996);YuenandKuiper(1991).Aggtraitaggression;Irrirritability;Angertraitanger;TypeATypeApersonality;DisRumdissipation–rumination;EmoSusemotionsusceptibility;Impimpulsivity;NarcTolerancetowardviolencewasusedasaproxyfortraitaggressiveness.Self-ruminationwasusedasaproxyfordissipation–rumination.instabilitywasusedasaproxyforemotionsusceptibility.Egocentrismwasusedasaproxyfornarcissism.PERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR TypeAPersonalitySimilartonarcissism,theTypeApersonalityprofileischarac-terizedbyfeelingsofinadequacywithregardtoself-worth(i.e.,self-esteem;Price,1982).ThesefearsofinadequacyoftenresultintheneedforTypeAindividualstoprovethemselvesthroughpersonalaccomplishments(Glass,1977;Mutaneretal.,1989;Price,1982).Whenconfrontedwithathreatorchallengetoeithertheircontrolortheircompetence,TypeAindividualsbecomeangry,irritated,andimpatient(Brunson&Matthews,1981;Glass,Snyder,&Hollis,1974;R.A.Martin,Kuiper,&Westra,1989;Rosenman,1978).Thus,higherlevelsofTypeApersonalityappeartobeassociatedwithagreatervulnerabilitytothreatstoself-competenceandapropensitytoexperienceangerinthepres-enceofthreat(Strubeetal.,1984).Usingafive-factorinventory,Morrison(1997)foundthataTypeAmeasurewasnegativelycorrelatedwithAgreeablenessbutfoundnocorrelationwithNeu-roticism.However,Byrne(1996)aswellasSibilia,Picozzi,andNardi(1995)reportedthatTypeApersonalitywaspositivelycorrelatedwithNeuroticism.Intheliterature,individualswhodonotmanifesttheTypeAbehavioralpatternarereferredtoasTypeBindividuals(Booth-Kewley&Friedman,1987;Mutaneretal.,1989).Although,conceptually,TypeAindividualswhoareprovokedshouldengageinmoreaggressivebehaviorthanTypeBindividuals,theavailableempiricalfindingsaremixed.TwostudieshavefoundthatTypeAindividualswerenomoreaggressiveintheirbehaviorthanTypeBindividualsunderprovokingconditions(Baron,Russell,&Arms,1985;Holmes&Will,1985).AnotherstudyshowedthatTypeAindividualsbehavedmoreaggressivelythanTypeBindividualsunderbothneutralandprovokingconditions(Check&Dyck,1986).However,moststudies(Carver&Glass,1978;Llorente,Bernardo,deFlores,&Valdes,1985;Muntaneretal.,1989;Strubeetal.,1984)haverevealedthatTypeAindividualsengageinhigherlevelsofaggressivebehaviorthanTypeBindividualsunderprovokingconditionsbutnotunderneutralconditions.Dissipationandruminationareconsideredoppositeendsofacontinuum;dissipatorstendtogetoverfeelingsofangerandhostilityrapidlyfollowingprovocation,butruminatorstendtomaintainandexacerbatetheirfeelingsofangerandhostilityforprolongedperiodsoftime(Caprara,1986).Thetendencytorumi-natereferstotherehearsingofexperiencesofprovocationandthoughtsofretaliation.Becauseoftheirtendencytoperseverateoverprovokingincidents,highruminators(lowdissipators)shouldbemorelikelythanhighdissipators(lowruminators)tobehaveaggressivelyfollowingprovocation.Capraraetal.(1996)reportedanegativecorrelationbetweenrumination(hostilerumination)andAgreeablenessandapositivecorrelationbetweenruminationandNeuroticism(alsoseeCaprara,Barbaranelli,&Comrey,1992).Consistentwiththedefinitionofdissipation–rumination,Ca-prara,Coluzzi,etal.(1985)revealedthat,inresponsetoinsult,highruminators(lowdissipators)administeredsignificantlyhigherlevelsofshocktoaconfederatethandidhighdissipators(lowruminators).However,intheabsenceofprovocation,theauthorsfoundlittledifferenceinthelevelsofshockadministeredbybothtypesofindividuals.Similarly,CollinsandBell(1997)showedaninteractionofdissipation–ruminationandlevelsofprovocation.Thatis,highruminatorsintheprovocationconditionadministeredagreaternumberofshockstoanopponentthandidhighdissipa-tors,andtheauthorsfoundthereverseintheneutralcondition.Incontrast,onestudy(Capraraetal.,1987)hasshownthathighruminators(lowdissipators)generallybehavedmoreaggressivelythanhighdissipators,regardlessofprovocationlevel.Althoughtheseresultsarenotentirelyconsistent,theoretically,thedissipation–ruminationvariableshouldinteractwithprovocationtoinduceaggressivebehavior.Intheliterature(e.g.,Barratt,1994;Eysenck,Pearson,Easting,&Allsop,1985;Parker&Bagby,1997),isdefinedastheextenttowhichindividualsareunabletocontroltheirthoughtsandbehaviors.Therelativeinabilitytocontrolone’sbehavioristhoughttostemfromdeficitsintheself-regulationofaffect,motivation,andarousalaswellasinworkingmemoryandhigherordercognitivefunctionsthatordinarilygiverisetohindsight,forethought,anticipatorybehavior,andgoal-directedaction(Bark-ley,1997).Barratt(1994)suggestedthathighlyimpulsiveindivid-ualsarecharacterizedbya“hair-triggertemper”(p.71)andbythelackofself-controlthattheyneedtorefrainfromaggressivebehaviorafterbeingprovoked.McCraeandCosta(1985)reportedthatimpulsivityispositivelycorrelatedwithNeuroticismbutun-correlatedwithAgreeableness.Shafer(2001)andNetteretal.(1998)corroboratedthepositivecorrelationbetweenimpulsivityandNeuroticism(butseeAluja,Garcia,&Garcia,2002).Studies(e.g.,Hynan&Grush,1986;Netteretal.,1998)haverevealedarelationbetweenimpulsivityandaggressivebehavior,particularlyunderconditionsofprovocation.Netteretal.(1998),forexample,hadparticipantsperformajointtaskwithaconfed-erateandeitherprovokedparticipantswithfrustrationandinsultsordidnot.Theresultsshowedthatintheprovocationcondition,individualswhowerehighinimpulsivityadministeredmorein-tenseelectricalshockstotheconfederatethandidindividualswhowerelowinimpulsivity.However,underneutralconditions,theresultsshowedthatlevelofimpulsivitydidnotreliablyinfluenceaggressivebehavior.Itappearsthat,inthefaceofprovocation,emotionalvolatilityandpoorself-regulationmayinducegreaterlevelsofaggressivebehavioramonghighlyimpulsiveindividuals.ThePresentStudyInthepresentstudy,wemeta-analyticallyexaminetheassoci-ationsbetweenpersonalityvariablesandaggressivebehavior,sep-aratelyunderprovokingandrelativelyneutralconditions.Ourmeta-analysishasthecapacitytorefinetheoriesofaggressioninseveralways.First,becauseatleastsomepersonalityvariablesarelikelypredictorsofaggressivebehavior,acomprehensivequanti-tativereviewwilldocumenttheneedfortheoriesofaggressiontoarticulatehowandwhypersonalityvariablesinfluenceobservablebehavior.Second,themeta-analysisislikelytosuggestthatthe-oriesmustconsiderwhetherpersonalityvariablesarelikelytointeractwithlevelsofprovocationintheirinfluencesonaggressivebehavior.Third,themeta-analyticfindingsmaypointtoadditionalvariablesthatmoderatethelinkbetweenpersonalityantecedentsandaggressivebehavior.Forourquantitativereview,wecalculatedaneffect-sizeesti-matethatcomparestheaggressivebehaviorofindividualswhoPERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR Bettencourtdiscussedcodingdiscrepanciesforthistestsettofurtherclarifythevariabledefinitionsandcodingmethods.Afterthisinitialtraining,thetwocodersseparatelycodedeachofthevariablesforallofthestudies.Whenadiscrepancywasfound,bothcodersindependentlyre-viewedthestudyagainandmadeadecisionaboutwhethertheywouldretaintheircodeormodifyit.Finally,interraterreliabilities(percentageofagreement)werecalculated;theyrangedfrom.87to1.00.RemainingdiscrepancieswerediscussedandresolvedinconferencewithB.AnnApositivemeanpersonality-differenceeffectsizeindicatedthatparticipantswhoscoredhighonagivenpersonalityvariableweremorebehaviorallyaggressivethanthosewhoscoredlowonthesamevariable.Thesepersonality-differenceeffectsizesarere-portedinTable3.Inaddition,Table3containsthecodedvaluesforthelevelofprovocation(neutralorprovocation),typeofpersonalityvariable,typeofprovocation,genderofparticipant,targetofaggression,typeofaggression,optiontoaggress,studydesign,andsourceofeffectsize.Becausethepresenceofextremevaluesinadatasetcanseriouslydistorttheoutcomesofanalyses(Tabachnick&Fidell,1989;Wilcox,1995),includingmeta-analyses(seeBettencourt&Miller,1996,andCooper,Charlton,Valentine,&Muhlenbruck,2000,foradiscussionoftheissue),thedistributionsoftheeffectsizeestimateswereexamined.Twoeffectsizes(6.04and5.06)inthedistributionwereidentifiedasextremeoutliers(throughtheinterquartilerangeprocedure;SASInstitute,1985).Extremevaluesinadistributionmaybeeitherexcludedormod-ified(Tabachnick&Fidell,1989).Toretainallofthestudiesintheanalyses,wechangedthevaluesofthesetwoextremeeffectsizestoequalthevalueoftherespectivenextclosesteffectsizeinthedistribution(Tabachnick&Fidell,1989;Wilcox,1995).AsshowninthetoppanelofTable4,themeaneffectsizeandtheCIforthemodifieddatawereessentiallyidenticaltothosefortheunmodi-fieddata.Weusedthemodifieddatainallsubsequentanalyses.Weightingtheeffectsizeofeachstudybyitssamplesizeisrecommended(e.g.,Hedges&Olkin,1985).Themean-weightedeffectsizeassignsgreaterweighttostudieswithlargersamplesizesontheassumptionthattheireffectsaremorereliable.Ninety-fivepercentCIswerecomputedforeachmean-weightedeffectsizewithineachclassofacodedvariable.Between-classstatistics)andwithin-classhomogeneitystatistics()wereusedtoanalyzetheeffectsizeestimates(Hedges&Olkin,1985).Thebetween-classstatisticisanalogoustoanstatisticandindicatesthemagnitudeofthevarianceamongtheaverageeffectsizes.Thewithin-classhomogeneitystatisticsintheweightedanalysesindi-catesignificantheterogeneityamongtheeffectsizesinthesample.Weconductedbothfixed-effectsanalysesandrandom-effectsanalyses(seeCooper&Hedges,1994).Infixed-effectsmodels,meta-analystsareabletomakeinferencesabouttheeffectsizesonlyinregardtothesetofreviewedstudies(Hedges&Vevea,1998).Forthistypeofanalysis,theeffectsizeestimatesthepopulationeffectwiththeerrorcalculatedfromtherandomsam-plingofparticipantswithinthestudies.Inrandom-effectsmodels,meta-analystsareabletomakeinferencesthatgeneralizebeyondthesetofreviewedstudiestoabroaderpopulationofstudies.Forthistypeofanalysis,itisassumedthatthevariabilityamongeffectsizesemergesfrombothparticipant-levelsamplingerrorandran-domdifferencesamongstudies(Hedges&Vevea,1998).Therandom-effectsmodelassumesthatstudieshavebeenrandomlysampledfromapopulationofstudiesthatcouldhavebeencon-ducted.Cooperetal.(2000)pointedoutthatitisoftenunclearwhetherafixed-effectsanalysisorarandom-effectsanalysisismostappropriateforagivenmeta-analysis.Itisoftenbesttoperformbothtypesofanalysesbecause,asOverton(1998)hasshown,random-effectsmodelscanoverestimateerrorvariance.Similarly,WangandBushman(1999)pointedoutthattheriskofTypeIIerrorisgreaterwithrandom-effectsmodelsandrecom-mendedconductingbothfixed-effectsandrandom-effectsInthepresentstudy,weconductedbothtypesofanalyses.Nevertheless,becauseoftherelativelysmallnumberofstudiesavailableintheliterature,wefocusontheresultsfromthefixed-effectsanalysis.Thatis,giventhatthesmallnumberofeffectsizesrenderedrelativelylowpowerandthatrandom-effectsanalysescanbeoverlyconservativeinsuchcases,wefocusonthefixed-effectsanalysestointerprettheeffectsizesavailableintheliter-ature.Therefore,theconclusionswedrawmaynotbegeneraliz-abletoabroaderpopulationofstudies.SummaryofEffectSizeEstimatesAsshowninthefirsttworowsofTable4,anoverallanalysisofthemeanpersonality-differenceeffectsize,whichignoredlevelofprovocation(i.e.,neutralvs.provocation),revealedtheexpectedinfluenceofpersonalityvariablesonaggressivebehavior.Ingen-eral,thosewhoscoredhighonthepersonalitymeasuresincludedinthemeta-analysisbehavedmoreaggressivelythanthosewhoscoredlowonthesemeasures.Theanalysisalsoindicatedsignif-icantwithin-classheterogeneityforthismeaneffectsize,revealingasubstantialamountofunexplainedvarianceintheoverallmeaneffectsize,fixed.001;randomEffectofProvocationonPersonalityDifferencesinAggressiveBehaviorOnegoalofthecurrentmeta-analysiswastodeterminewhetherthemagnitudesofpersonality-differenceeffectsizesweredifferentunderneutralandprovokingconditions.AsshowninthelowerpanelofTable4,analysesthatignoredthespecifictypeofper-sonalityvariablebutcomparedneutralwithprovocationconditionsrevealedthatthemeanpersonality-differenceeffectsizeunderneutralconditionswasreliablysmallerthanthatunderprovocationconditions,fixed.01;random.05.Thisfindingsuggeststhatpersonalityvariablesandthelevelofprovocation(neutralvs.provoking)interacttoinfluenceaggressivebehavior.Nevertheless,therespective95%CIsshowedthatthemeaneffectsizesassociatedwithboththeneutralandtheprovokingconditionsweregreaterthanzero.Thislatterresultsuggeststhat,comparedwithparticipantswhoscoredlow,thosewhoscoredhighonthepersonalityvariablesbehavedmoreag-gressivelyunderbothneutralandprovokingconditions.TypeofPersonalityVariableOnthebasisofourqualitativereviewoftheliterature,wepredictedthatsomepersonalityvariableswouldbepositivelyassociatedwithaggressivebehavioronlyunderprovocationcon-PERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR Table3( GenderofTypeofofeffectGiancola(2002)1030.71262613121423Giancola&Zeichner(1995a)300.56581113121223Giancola&Zeichner(1995a)300.60542113121223Giancola&Zeichner(1995b)790.40422612121423Giancola&Zeichner(1995b)790.91642112121423Giancola&Zeichner(1995b)790.25962212121423Hammock&Richardson(1992)1946.03701113121122Hammock&Richardson(1992)1940.55862113121122Haskell(2002)480.19561722122122Haskell(2002)480.60492722122122Holmes&Will(1985)170.84701322122112Holmes&Will(1985)200.34312322122112Hynan&Grush(1986)200.44991822121112Hynan&Grush(1986)200.06602822121112Josephson(1988)3960.87122142222421Knott(1970)181.05582112122412Lagerspetz&Engblom(1979)570.51001103022511Lagerspetz&Engblom(1979)571.20601103012511Larsenetal.(1972)580.42371103021522Leibowitz(1968)380.61541102021522Lindsay(1999)970.60231223222122Lindsay(1999)880.46372223222122Llorenteetal.(1985)600.33501332121112Llorenteetal.(1985)600.66332332121112Matthews&Angulo(1980)600.52112343222411Muntaneretal.(1989)610.24911302021112Muntaneretal.(1989)600.41502332121112Netteretal.(1998)100.98181812121112Netteretal.(1998)100.89382812121112Parrott&Zeichner(2001)170.82292212121412Parrott&Zeichner(2002)250.64521612122212,3Parrott&Zeichner(2002)250.25912612122212,3Pihletal.(1997)290.29661612121213Pihletal.(1997)291.24322612121213Renzietal.(1984)500.12831223221112Renzietal.(1984)0.09642223221112Renzietal.(1984)500.12551523221112Renzietal.(1984)500.14802523221112Scheieretal.(1978)630.71401103021522Shembergetal.(1968)451.35131103021511Shondrick(1996)200.38611112121212Shondrick(1996)200.02982112121212Strubeetal.(1984)210.53921332232112Strubeetal.(1984)220.81612332232112Twenge&Campbell311.18472723122422Twenge&Campbell0.33121723222122Twenge&Campbell390.90622723222122tablecontinuesPERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR tions,fixed.001;randomAlso,asshowninthelowerpanelofTable5,fordissipation–rumination,themeanpersonality-differenceeffectsizeintheneutralcategorywassmallandequivalenttozero.Incontrast,highrumina-torsbehavedmoreaggressivelythandidlowruminators(i.e.,highdissipators)underprovocation.Theanalysisthatcomparedthepersonality-differenceeffectsizesintheneutralandprovocationcon-ditionsrevealedthatthesemeaneffectsizesweredifferent,fixed.001;randomTheresultsforemotionalsusceptibilityweresimilartothosefortraitanger,TypeApersonality,anddissipation–rumination,buttheeffectswereweaker.Theresultsforemotionalsusceptibilityrevealedthatthemeaneffectsizeunderprovokingconditionswasmarginallylargerthanthatunderneutralconditions,fixed.10;randomTheanalysisshowedthatthemeaneffectsizefortheneutralconditionwassmall,buttheCIsuggesteditwassomewhatlargerthanzero.Therespectiveeffectsizefortheprovocationconditionwasrelativelylargeanddiffer-entfromzero.Withafewexceptions,thefindingsfornarcissismandimpul-sivitywereconsistentwiththeinteractionpatternbetweenthepersonalityvariablesandlevelofprovocation.AsshownatthebottomofTable5,fortheneutralcategories,themeaneffectsizesTable4MeanPersonality-DifferenceEffectSizes CategoryofeffectsizesFixedRandom95%CIMean95%CINonmodified1020.540.48,0.600.660.57,0.75Modified1020.500.45,0.550.570.44,0.69Neutralconditions430.370.29,0.450.380.19,0.57Provokingconditions590.600.53,0.670.720.55,0.89Personality-differenceeffectsizesthatarepositiveindicatethathighscoresonthepersonalityvariableswereassociatedwithhigherlevelsofaggressivebehavior.Confidenceintervals(CIs)thatdonotincludezeroindicatethattheeffectsizeisofsignificantmagnitude.numberofeffectsizesinthecategory.Table5MeanPersonality-DifferenceEffectSizesasaFunctionofPersonalityVariable CategoryandconditionFixedRandom95%CI95%CITraitaggressivenessNeutral110.850.68,1.020.850.47,1.23Provoking160.710.57,0.840.740.42,1.06TraitirritabilityNeutral60.640.46,0.830.750.26,1.24Provoking100.790.62,0.961.170.77,1.57TraitangerNeutral30.090.17,0.370.060.46,0.99Provoking80.500.32,0.670.560.12,0.99TypeApersonalityNeutral100.29,0.140.020.39,0.44Provoking120.470.29,0.660.480.11,0.86Neutral21.14,0.141.52,0.48Provoking21.060.40,1.731.080.06,2.10Neutral40.240.03,0.460.300.29,0.89Provoking50.500.29,0.710.620.07,1.17Neutral50.110.11,0.320.040.50,0.58Provoking60.530.33,0.740.590.10,1.09Neutral20.000.74,0.730.130.95,1.21Provoking50.490.13,0.860.630.02,1.28Personality-differenceeffectsizesthatarepositiveindicatethathighscoresonthepersonalityvariableswereassociatedwithhigherlevelsofaggressivebehavior.Confidenceintervals(CIs)thatdonotincludezeroindicatethattheeffectsizeisofsignificantmagnitude.numberofeffectsizesinthecategory.PERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR differentfromtheobject,asopposedtowhentheprovocateurandtheobjectofaggressionwerethesameperson,fixed.001;randomTheseoutcomesmaysuggestthatindividualswhoscoredlowontheaggression-pronevariablesbehavedlessaggressivelywhentheobjectoftheaggressionwasnotthesourceofprovocationbutthatthosewhoscoredhighonthesevariablesbehavedmoreaggressivelyregardlessofwhethertheindividualwasthesourceofprovocation.Also,weanalyzedtheresultsaccordingtothetypeofprovoca-tiontowhichtheparticipantswereexposed.Thethreetypesincludedphysicalprovocation,verbalprovocation,andfrustration.Therespectivemeanpersonality-differenceeffectsizesfortheprovocationconditionsareshowninTable9.Forallthreetypesofprovocation,theresultsindicatedthathigherscoresontheperson-alityvariableswereassociatedwithhigherlevelsofaggressivebehaviorinboththeprovocation-sensitiveandtheaggression-Table6MeanPersonality-DifferenceEffectSizesbyTypeofAggression FixedRandom95%CIMean95%CIProvocationsensitiveProvoking30.770.30,1.230.660.01,1.32AggressionproneProvoking10.370.45,1.200.371.05,1.79ProvocationsensitiveNeutral240.070.04,0.180.030.27,0.32Provoking340.500.40,0.590.510.26,0.75AggressionproneNeutral170.750.62,0.870.800.50,1.10Provoking230.720.62,0.830.870.61,1.13Thephysicalaggressioncategoryincludedmeasuressuchashittingorshoving,intensityofnoxiousnoise,intensityofelectricshock,durationofnoxiousnoise,durationofelectricshock,numberofnoxiousnoiseblasts,andnumberofelectricshocks,andtheverbalaggressioncategoryincludedmeasuresofnegativeevaluativefeedbackordisparagingcommentsdirectedattheconfederate.numberofeffectsizesinthecategory;CIconfidenceinterval.Table7MeanPersonality-DifferenceEffectSizesbyGenderofParticipant CategoryofaggressionFixedRandom95%CIMean95%CINeutralconditionsProvocationsensitiveMale160.20,0.160.040.28,0.38Bothfemaleandmale60.040.14,0.220.51,0.47AggressionproneMale30.410.04,0.850.370.41,1.15Bothfemaleandmale170.630.52,0.740.740.45,1.03ProvokingconditionsProvocationsensitiveFemale20.590.05,1.130.600.33,1.54Male210.500.35,0.650.550.26,0.84Bothfemaleandmale100.520.37,0.660.600.22,0.98AggressionproneMale110.720.53,0.910.750.35,1.16Bothfemaleandmale190.680.58,0.790.900.63,1.18Note.knumberofeffectsizesinthecategory;CIconfidenceinterval.PERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR bility,narcissism,andimpulsivity.Forthemeta-analysis,wecal-culatedeffectsizesoftheassociationsbetweenthesepersonalityvariablesandaggressivebehavior(i.e.,personality-differenceef-fectsizes),separatelyunderprovocationandneutralconditions.Overall,ourmeta-analyticfindingsareconsistentwiththefind-ingsfromourqualitativereview.Themeanpersonality-differenceeffectsizesrevealedthatafewofthepersonalityvariableswerepositivelyassociatedwithaggressivebehaviorunderbothprovo-cationandneutralconditionsbutthatotherpersonalityvariableswerepositivelyassociatedwithaggressivebehavioronlywhentheaggressorhadbeenprovoked.Inparticular,traitaggressivenessandtraitirritability(whichwerehighlyandpositivelycorrelated)wereassociatedwithgreateraggressivebehavioracrosscondi-tions.Thatpersonswhoscorehighontraitaggressivenessdirectgreaterlevelsofaggressivebehaviortowardothersevenwhensituationsarerelativelyneutralmaysuggestthattheyhavethecapacitytoengageinacold-bloodedstyleofaggressivebehavior.Bycontrast,traitanger,TypeApersonality,dissipation–rumination,emotionalsusceptibility,narcissism,and,forthemostpart,impulsivitywereassociatedwithgreateraggressivebehavioronlyunderprovocationconditions.Thisfindingmaysuggestthatthosewhoscorehighontheselatterpersonalityvariableshaveaparticularpropensitytoexhibitahot-bloodedstyleofaggressiveOnthebasisofthetwopatternsofaggressivebehaviorrevealedinthemeta-analysis,weconductedaseriesofadditionalexplor-atoryanalyses.Todoso,wedividedtheeffectsizesintotwocategories;thefirstcategoryincludedthepersonalityvariablesthatseemedtobeindicativeofindividualspronetoaggressivebehav-ioracrosssituations(i.e.,aggressionprone),andthesecondcate-goryincludedthosewhoseemedtoreactaggressivelyonlyunderconditionsofprovocation(i.e.,provocationsensitive).Usingthesecategories,wewereabletofurtheranalyzethepersonality-differenceeffectsizesbythetypeofaggressivebehavior,partic-ipants’gender,thetypeofprovocation,thesourceofprovocation,andtheavailableoptionsforrespondingotherthanaggressiveThemeta-analyticfindingsshowthatboththeprovocation-sensitiveandtheaggression-pronecategorieswerepositivelyas-sociatedwithaggressivebehavior,regardlessofthetypeofprov-ocation(i.e.,physicalprovocation,verbalprovocation,andfrustration).Anunanticipatedfindingwasthat,fortheaggressive-pronecategory,thisassociationbetweenthepersonalityvariablesandaggressivebehaviorwaslargerunderfrustrationthanunderphysicalprovocation.Itmaybethatthemagnitudeofthepersonality-differenceeffectsizewaslargerbecausethosewhoscoredlowonthesepersonalityvariables(i.e.,traitaggressiveness)wereunlikelytoengageinaggressivebehaviorwhenthesituationwasmerelyfrustrating,whichtherebyyieldedalargerpersonality-differenceeffectsizeforthistypeofprovocation.Nevertheless,thesefindingsshouldbeconsideredtentative,becauseonlytwoeffectsizeswereavailableinthiscategory.Also,wecategorizedthemeaneffectsizesintermsofthesourceofprovocationvis-a`-vistheobjectoftheaggressivebehavior.Fortheprovocation-sensitivecategory,theresultsshowedthatwhethertherecipientoftheaggressivebehaviorhadbeenthesourceofprovocationorsomeoneelsehadbeenthesourcehadlittleinflu-enceontheassociationbetweenthepersonalityvariablesandaggressivebehavior.Bycomparison,fortheaggression-pronecategory,themagnitudeofthepersonality-differenceeffectsizewaslargerwhentherecipientwasnotthesourceofprovocation.Itmaybethatthosewhoscoredlowonthepersonalityvariablesintheaggression-pronecategorywerenotparticularlylikelytobeaggressivetowardanotherpersonwhohadnotprovokedthembutthatthosewhoscoredhighonpersonalityvariablesintheaggression-pronecategorywerehighlyaggressive,regardlessofwhoprovokedthem.Inaddition,weexaminedwhetherthemagnitudesofthepersonality-differenceeffectsizesweremoderatedbywhethertheparticipantswereprovidedaresponseoptionotherthanthatofaggressivebehavior(i.e.,anonaggressiveresponse).Whenanoptionotherthanaggressivebehaviorwasavailabletoparticipants,aggressivebehaviorwasleastappropriateinneutralconditions.Althoughthemeaneffectsizewasderivedfromonlythreereports,Table10MeanPersonality-DifferenceEffectSizesbyOptiontoAggress CategoryandFixedRandom95%CIMean95%CIForcedtoaggressProvocationsensitiveNeutral150.070.07,0.220.080.25,0.41Provoking190.520.39,0.650.570.28,0.87AggressionproneNeutral140.890.74,1.040.880.55,1.22Provoking150.770.63,0.910.940.61,1.27FreetoaggressProvocationsensitiveNeutral110.070.09,0.230.070.32,0.45Provoking190.500.37,0.630.580.29,0.88AggressionproneNeutral30.440.21,0.670.510.17,1.18Provoking110.700.55,0.850.870.49,1.25Note.knumberofeffectsizesinthecategory;CIconfidenceinterval.PERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR replicatedintherandom-effectsanalyses.Theoretically,thislackofcorrespondencemaysuggestthattheoutcomesofthemeta-analysisaretrueforthespecificsampleofstudiesweincludedbutnottrueforthepotentialpopulationofstudiesthatcouldbeconducted.Practically,however,thisproblemmaypointtothefactthatrandom-effectsmodelsoverestimateerrorvariance(Overton,1998;Wang&Bushman,1999).Also,therewassomevariabilityinthequalityoftheincludedstudies.Althoughsomemeta-analystsexcludestudiesonthebasisofjudgmentsthatthestudiesareoflesserscientificquality,wewereunabletodothisbecauseoftherelativefewstudiesavailablefortheanalysis.Toassessthequalityofstudiesinourdataset,wederivedthejournalimpactfactorsfromthesciencecitationcriteriaforjournalquality.Wefoundthattherangeinimpactfactorsforthejournalsthatpublishedthestudiesweincludedinthemeta-analysisrangedbetween0.277and3.862;however,themeanoftheseimpactfactorswas1.96,andonlythreeofthepublicationshadimpactfactorslowerthan1.00.Thus,althoughsomeofthestudieswereunpublishedorwerepublishedinlowerqualityjour-nals,theimpactfactorssuggestthatthemajorityofthestudiesincludedinourmeta-analyticreviewwereofrelativelyhighItisimportanttonote,however,thatfindingsofthemeta-analysismaybegeneralizableonlytoinstancesofphysicalag-gression.Themeta-analysisuncoveredthefactthatthevastma-jorityofstudiesintheavailablesampleofstudieshadmeasuredphysicalaggression,largelyignoringothertypesofaggressivebehavior(e.g.,verbalaggression,indirectaggression).Itshouldbenotedthattheseaggressivebehaviorsaretypicallydirectedatconfederatesandincludearelativelylimitedsetofoperationaliza-tionsofphysicalaggression,suchasostensibleshocksornoiseblasts.Clearly,futureresearchshouldexplorewhetherthepositiveassociationsrevealedinthemeta-analysisgeneralizetoothertypesofaggressivebehavior.Becausewewantedtoavoidbiasesassociatedwithself-reportsofaggressivebehavior,welimitedtheincludedstudiestothosethatmeasuredactualaggressivebehavior.However,somefeaturesofstudiesthatmeasureaggressivebehaviormaylimitthegener-alizabilityofthefindings.Forexample,thesestudiestypicallyincludeonlyalimitedsetofoperationalizationsofaggression(e.g.,ostensibleshock,intensityofnoiseblasts),aggressivebehaviorbetweenrelativestrangers,limitedopportunitiestoretaliateagainsttheaggressor,andfewopportunitiesforresponsesotherthanaggressivebehavior.Nevertheless,researchers(Andersonetal.,1999;Berkowitz&Donnerstein,1982)haverevealedthatthefindingsofexperimentalstudiesofaggressivebehaviorhaveex-ternalvalidityandarecomparabletothefindingsofstudiesusingothermethodologies.Giventheconfluenceofevidence,itseemslikelythatspecificpersonalityvariablespredictthetendencyeithertoengageinaggressivebehavioracrossavarietyofsituations(i.e.,aggressiveprone)ortoengageinaggressivebehaviorprimarilyinresponsetoprovocation(i.e.,provocationsensitive).Nevertheless,theconnec-tionswedrawamongthepersonalitydimensionsspecifiedbythefive-factormodel,thepersonalityvariablesincludedinourmeta-analysis,anddifferentpatternsofaggressionlargelyremaintheo-retical.Thus,ourreviewpointstotheneedforfurtherresearchthatexaminestherelationsbetweenpersonalitydimensionsandag-gressivebehavior.Withinthedevelopmentalliteratureonaggres-sion,itmightbeusefultostudychildren’slevelsofAgreeablenessandNeuroticism.Also,anyresearchusingdimensionsfromthefive-factormodelshouldincludedirectmeasuresofaggressivebehavioraswellasdeterminetheassociationsbetweentheper-sonalitydimensionsandaggressivebehavior,separatelyundernonprovokingandprovokingconditions.Moreover,researcherswhostudyspecificpersonalityvariablesmayneedtoprovideanunderstandingofthewaysthesevariablesarefittedintotherubricofthefive-factormodel.Usingthismodelofpersonalitymaybringmoreconceptualclaritytotheplethoraofpersonalityvariablesthathavebeenconsideredintheliteratureonaggressivebehavior.Perhapsmostimportant,generaltheoriesshouldincludeperson-alityasacentralvariableinmodelsofaggressionforanumberofreasons.Ourmeta-analysisdocumentsthatpersonalityvariablesarepositivelyassociatedwithaggressivebehavior,atleastundersomecircumstances.Moreover,themeta-analysisshowsthatsomeofthepersonalityvariablesinteractedwithsituationalprovocationintheirinfluencesonaggressivebehavior.Giventhesefindings,theoreticalmodelsofaggressionarelikelytohavemoreexplan-atorypoweriftheyarticulatethewayspersonalityshouldbeassociatedwithaggressivebehavior.Indeed,theoreticalmodelsofaggressionthatconsidertheroleofpersonalitymaybeabletomoreclearlyarticulatewhichvariablesunderlielinksbetweenpersonalityandaggressivebehavior.Onthebasisofourreviewofcontemporarytheoriesonaggression,wespeculatethatdifferencesincognitiveprocesses,social-informationprocessing,levelsofnegativeaffect,anddifficultieswithself-regulationmightbemechanismsthatexplainassociationsbetweenpersonalityandaggression.Itremainsunclear,however,whetheralloronlysomeoftheseprocessvariablesexplainrelationsbetweenspecificper-sonalityvariablesordimensionsandaggressivebehavior.Althoughagrowingbodyofliteraturesuggeststhat,tounder-standaggressivebehavior,researchersmuststudythedistinctinfluencesofpersonality,muchresearchontheseissuesisneeded.Becauseproblemswithaggressionandviolencecontinuetoplaguepeople’sinterpersonallives,theirintergroupinteractions,andso-cietyingeneral,itisincumbentonsocialscientiststodevelopabetterunderstandingofthecomplexdynamicsamongpersonalityvariables,situationalvariables,andaggressivebehavior.Doingsonotonlywillenrichthefield’stheoreticalunderstandingofhumanaggressionbutalsopromisestorefinetherapeuticandpolicyinterventionsaimedatreducingaggressionandviolence.ReferencesmarkedwithanasteriskindicatestudiesincludedintheAllport,G.(1961).Patternandgrowthinpersonality.NewYork:Holt,Rinehart,&Winston.Altemeyer,B.(1981).Right-wingauthoritarianism.Winnipeg,Manitoba,Canada:UniversityofManitobaPress.Aluja,A.,Garcia,O.,&Garcia,L.F.(2002).AcomparativestudyofZuckerman’sthreestructuralmodelsforpersonalitythroughtheNEO-PI-R,ZKPQ-III-R,EPQ-RSandGoldberg’s50-bipolaradjectives.sonalityandIndividualDifferences,33,Anderson,C.A.(1997).Effectsofviolentmoviesandtraithostilityonhostilefeelingsandaggressivethoughts.AggressiveBehavior,23,PERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR Buss,A.H.(1963).Physicalaggressioninrelationtodifferentfrustrations.JournalofAbnormalandSocialPsychology,67,Buss,A.H.(1966).Instrumentalityofaggression,feedback,andfrustrationasdeterminantsofphysicalaggression.JournalofPersonalityandSocialPsychology,3,Buss,A.H.,&Durkee,A.(1957).Aninventoryforassessingdifferentkindsofhostility.JournalofConsultingPsychology,21,Buss,A.H.,&Perry,M.P.(1992).TheAggressionQuestionnaire.ofPersonalityandSocialPsychology,63,452–459.Buss,A.H.,&Plomin,R.(1975).AtemperamenttheoryofpersonalityNewYork:Wiley.Byrne,D.G.(1996).TypeAbehaviour,anxietyandneuroticism:Recon-ceptualizingthepathophysiologicalpathsandboundariesofcoronary-pronebehaviour.StressMedicine,12,*Caprara,G.V.(1982).Acomparisonofthefrustration-aggressionandemotionalsusceptibilityhypotheses.AggressiveBehavior,8,234–236.Caprara,G.V.(1986).Indicatorsofaggressiveness:TheDissipation-RuminationScale.PersonalityandIndividualDifferences,7,Caprara,G.V.,Barbaranelli,C.,&Comrey,A.L.(1992).Apersonologicalapproachtothestudyofaggression.PersonalityandIndividualDiffer-ences,13,77–84.Caprara,G.V.,Barbaranelli,C.,&Zimbardo,P.G.(1996).Understandingthecomplexityofhumanaggression:Affective,cognitive,andsocialdimensionsofindividualdifferencesinpropensitytowardaggression.EuropeanJournalofPersonality,10(2),133–155.Caprara,G.V.,Cinanni,V.,D’Imperio,G.,Passerini,S.,Renzi,P.,&Travaglia,G.(1958).Indicatorsofimpulsiveaggression:Presentstatusofresearchonirritabilityandemotionalsusceptibilityscales.andIndividualDifferences,6,665–674.*Caprara,G.V.,Coluzzi,M.,Mazzotti,E.,Renzi,P.,&Zelli,A.(1985).Effectofinsultanddissipation-ruminationondelayedaggressionandNeurologiaePsichiatria,46,130–139.Caprara,G.V.,Gargaro,T.,Pastorelli,C.,Prezza,M.,Renzi,P.,&Zelli,A.(1987).Individualdifferencesandmeasuresofaggressioninlabora-torystudies.PersonalityandIndividualDifferences,8,885–893.Caprara,G.V.,Manzi,J.,&Perugini,M.(1992).Investigatingguiltinrelationtoemotionalityandaggression.PersonalityandIndividualDifferences,13,519–532.Caprara,G.V.,Passerini,S.,Pastorelli,C.,Renzi,P.,&Zelli,A.(1986).Instigatingandmeasuringinterpersonalaggressionandhostility:Amethodologicalcontribution.AggressiveBehavior,12,Caprara,G.V.,&Pastorelli,C.(1993).Earlyemotionalinstability,proso-cialbehaviour,andaggression:Somemethodologicalaspects.JournalofPersonality,7(1),19–36.Caprara,G.V.,Perugini,M.,&Barbaranelli,C.(1994).Studiesofindi-vidualdifferencesinaggression.InM.Potegal&J.F.Knutson(Eds.),Thedynamicsofaggression:Biologicalandsocialprocessesindyadsandgroups(pp.123–153).Hillsdale,NJ:Erlbaum.Caprara,G.V.,Regalia,C.,&Bandura,A.(2002).Longitudinalimpactofperceivedself-regulatoryefficacyonviolentconduct.EuropeanPsy-chologist,7,63–69.*Caprara,G.V.,&Renzi,P.(1981).Thefrustration-aggressionhypothesisvs.irritability.RecherchesdePsychologieSociale,3,75–80.*Caprara,G.V.,Renzi,P.,Alcini,P.,D’Imperio,G.,&Travaglia,G.(1983).Instigationtoaggressandescalationofaggressionexaminedfromapersonologicalperspective:Theroleofirritabilityandofemo-tionalsusceptibility.AggressiveBehavior,9,*Caprara,G.V.,Renzi,P.,Amolini,P.,D’Imperio,G.,&Travaglia,G.(1984).Theelicitingcuevalueofaggressiveslidesreconsideredinapersonologicalperspective:Theweaponseffectandirritability.peanJournalofSocialPsychology,14,*Caprara,G.V.,Renzi,P.,D’Aguello,D.,D’Imperio,G.,Rielli,I.,&Travaglia,G.(1986).Interpolatingphysicalexercisebetweeninstigationtoaggressandaggression:Theroleofirritabilityandemotionalsuscep-AggressiveBehavior,12,Carlson,M.,Marcus-Newhall,A.,&Miller,N.(1989).Evidenceforageneralconstructofaggression.PersonalityandSocialPsychologyBulletin,15,Carlson,M.,Marcus-Newhall,A.,&Miller,N.(1990).Effectsofsitua-tionalaggressivecues:Aquantitativereview.JournalofPersonalityandSocialPsychology,58,622–633.Carlson,M.,&Miller,N.(1988).Badexperiencesandaggression.ologyandSocialResearch,72,*Carver,C.S.,&Glass,D.C.(1978).Coronary-pronebehaviorpatternandinterpersonalaggression.JournalofPersonalityandSocialPsychol-ogy,36,Carver,C.S.,&Scheier,M.F.(2000).Perspectivesonpersonality.Boston:Allyn&Bacon.Cattell,R.B.,Beloff,H.,&Coan,R.W.(1958).Psychologicaltestreviews:IPATHighSchoolPersonalityQuestionnaire(H.S.P.Q.).JournalofConsultingPsychology,22,*Check,J.V.P.,&Dyck,D.G.(1986).HostileaggressionandTypeAPersonalityandIndividualDifferences,7,819–828.*Cheong,J.,&Nagoshi,C.T.(1999).Effectsofsensationseeking,instructionset,andalcohol/placeboadministrationonaggressivebehav-Alcohol,17,81–86.Cherek,D.R.,Schnapp,W.,Moeller,F.G.,&Dougherty,D.M.(1996).Laboratorymeasuresofaggressiverespondinginmaleparoleeswithviolentandnonviolenthistories.AggressiveBehavior,22,Coie,J.D.,&Dodge,K.A.(2000).Aggressionandantisocialbehavior.InW.Damon(SeriesEd.)&N.Eisenberg(Vol.Ed.),Handbookofchildpsychology:Vol.3.Social,emotional,andpersonalitydevelopmented.,pp.779–862).NewYork:Wiley.*Collins,K.,&Bell,R.(1997).Personalityandaggression:Thedissipation-ruminationscale.PersonalityandIndividualDifferences,22,Cooper,H.M.(1989).Integratingresearch:Aguideforliteraturereviews(2nded.).NewburyPark,CA:Sage.Cooper,H.,Charlton,K.,Valentine,J.C.,&Muhlenbruck,L.(2000).Makingthemostofsummerschool:Ameta-analyticandnarrativeMonographsonChildDevelopment,65(1),v–118.Cooper,H.,&Hedges,L.V.(Eds.).(1994).ThehandbookofresearchNewYork:RussellSageFoundation.Costa,P.T.,&McCrae,R.R.(1992).RevisedNEO-PersonalityInventory(NEO-PI-R)andNEOFive-FactorInventory(FFI)manual.Odessa,FL:PsychologicalAssessmentResources.Costa,P.T.,McCrae,R.R.,&Dembroski,T.M.(1989).Agreeablenessversusantagonism:ExplicationofapotentialriskfactorforCHD.InA.W.Siegman&T.M.Dembroski(Eds.),Insearchofcoronaryprone(pp.41–63).Hillsdale,NJ:Erlbaum.Crick,N.R.,&Dodge,K.A.(1994).Areviewandreformulationofsocialinformation-processingmechanismsinchildren’ssocialadjustment.PsychologicalBulletin,115,74–101.Datlow,C.M.(1999).Theeffectsofdiazepamandhostilitytowardwomenonphysicalaggressioninmen.DissertationAbstractsInternational:SectionB:TheSciences&Engineering,60(1-B),0363,AbstractNo.1999–95014–067.Deffenbacher,J.L.,Oetting,E.R.,Thwaites,G.A.,Lynch,R.S.,Baker,D.A.,Stark,R.S.,etal.(1996).State–traitangertheoryandtheutilityoftheTraitAngerScale.JournalofCounselingPsychology,43,deFlores,T.,&Valdes,M.(1986).BehaviourpatternA:Reward,fightorPersonalityandIndividualDifferences,7,319–326.Dengerink,H.A.(1971).Anxiety,aggression,andphysiologicalarousal.JournalofExperimentalResearchinPersonality,5,Dengerink,H.A.,O’Leary,M.R.,&Kasner,K.H.(1975).Individualdifferencesinaggressiveresponsestoattack:Internal-externallocusofcontrolandfielddependence-independence.JournalofResearchinPersonality,9,Dill,K.E.(1999).ViolentvideogameandtraitaggressioneffectsonPERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR Hedges,L.V.,&Olkin,I.(1985).Statisticalmethodsformeta-analysis.NewYork:AcademicPress.Hedges,L.V.,&Vevea,J.L.(1998).Fixed-andrandom-effectsmodelsinPsychologicalMethods,3,486–504.Hennig,J.,Reuter,M.,Netter,P.,Burk,C.,&Landt,O.(2005).TwotypesofaggressionaredifferentiallyrelatedtoserotonergicactivityandtheA779CTPHpolymorphism.BehavioralNeuroscience,119,16–25.Hoaken,P.N.S.,Shaughnessy,V.K.,&Pihl,R.O.(2003).Executivecognitionfunctioningandaggression:Isitanissueofimpulsivity?AggressiveBehavior,29,*Holmes,D.S.,&Will,M.J.(1985).Expressionofinterpersonalaggres-sionbyangeredandnonangeredpersonswithTypeAandTypeBbehaviorpatterns.JournalofPersonalityandSocialPsychology,48,Huesmann,L.R.(1988).Aninformation-processingmodelforthedevel-opmentofaggression.AggressiveBehavior,14,Huesmann,L.R.(1998).Theroleofsocialinformationprocessingandcognitiveschemainacquisitionandmaintenanceofhabitualaggressivebehavior.InR.G.Geen&E.Donnerstein(Eds.),Humanaggression:Theories,research,andimplicationsforpolicy(pp.73–109).NewYork:AcademicPress.*Hynan,D.J.,&Grush,J.E.(1986).Effectsofimpulsivity,depression,provocationandtimeonaggressivebehavior.JournalofResearchinPersonality,20,158–171.Irwin,A.R.,&Gross,A.M.(1995).Cognitivetempo,violentvideogames,andaggressivebehaviorinyoungboys.JournalofFamilyViolence,10,Ito,T.A.,Miller,N.,&Pollock,V.E.(1996).Alcoholandaggression:Ameta-analysisonthemoderatingeffectsofinhibitorycues,triggeringevents,andself-focusedattention.PsychologicalBulletin,120,60–82.Jenkins,C.D.,Zyzanski,S.J.,&Rosenman,R.H.(1979).JenkinsActivityNewYork:PsychologicalCorporation.Jensen-Campbell,L.A.,Gleason,K.A.,Adams,R.,&Malcolm,K.T.(2003).Interpersonalconflict,Agreeableness,andpersonalitydevelop-JournalofPersonality,71,1059–1085.Jensen-Campbell,L.A.,&Graziano,W.G.(2001).Agreeablenessasamoderatorofinterpersonalconflict.JournalofPersonality,69,Johnson,B.T.(1989).DSTAT:Softwareforthemeta-analyticreviewofresearchliteratures.Hillsdale,NJ:Erlbaum.Johnson,B.T.,&Eagly,A.H.(2000).Quantitativesynthesisofsocialpsychologicalresearch.InH.T.Reis&C.M.Judd(Eds.),Handbookofresearchmethodsinsocialandpersonalitypsychology(pp.496–528).NewYork:CambridgeUniversityPress.Joireman,J.,Anderson,J.,&Strathman,A.(2003).Theaggressionpara-dox:Understandingthelinksamongaggression,sensationseeking,andtheconsiderationoffutureconsequences.JournalofPersonalityandSocialPsychology,84,*Josephson,W.L.(1988).Televisionviolenceandchildren’saggression:Testingthepriming,socialscript,anddisinhibitionpredictions.ofPersonalityandSocialPsychology,53,882–890.Kernberg,O.(1975).Borderlineconditionsandpathologicalnarcissism.NewYork:JasonAronson.Kernis,M.H.,Grannemann,B.D.,&Barclay,L.C.(1989).Stabilityandlevelofself-esteemaspredictorsofangerarousalandhostility.ofPersonalityandSocialPsychology,56,*Knott,P.D.(1970).Afurthermethodologicalstudyofthemeasurementofinterpersonalaggression.PsychologicalReports,26,807–809.Knott,P.D.,Lasater,L.,&Shuman,R.(1974).Aggression-guiltandconditionabilityforaggressiveness.JournalofPersonality,42,Kogut,D.,Langley,T.,&O’Neal,E.C.(1992).Genderrolemasculinityandangryaggressioninwomen.SexRoles,26,Kokkonen,M.,&Pulkkinen,L.(1999).Emotionregulationstrategiesinrelationtopersonalitycharacteristicsindicatinglowandhighself-controlofemotions.PersonalityandIndividualDifferences,27,Kolvin,I.,Miller,F.J.,Scott,D.M.,Gatzanis,S.R.M.,&Fleeting,M.Continuitiesofdeprivation?Aldershot,Hampshire,England:*Langerspetz,K.M.,&Engblom,P.(1979).ImmediatereactionstoTV-violencebyFinnishpre-schoolchildrenofdifferentpersonalityScandinavianJournalofPsychology,20,*Larsen,K.S.,Coleman,D.,Forbes,J.,&Johnson,R.(1972).Isthesubject’spersonalityortheexperimentalsituationabetterpredictorofasubject’swillingnesstoadministershocktoavictim?JournalofPer-sonalityandSocialPsychology,22,Lefkowitz,M.M.,Eron,L.D.,Walder,L.O.,&Huesmann,L.R.(1977).Growinguptobeviolent:AlongitudinalstudyofthedevelopmentofNewYork:PergamonPress.*Leibowitz,G.(1968).Comparisonofself-reportandbehavioraltech-niquesofassessingaggression.JournalofConsultingandClinicalPsychology,32,Leyens,J.,Herman,G.,&Dunand,M.(1982).Theinfluenceofanaudienceuponthereactionstofilmedviolence.EuropeanJournalofSocialPsychology,12,Lieberman,J.D.,Solomon,S.,Greenberg,J.,&McGregor,H.A.(1999).Ahotnewwaytomeasureaggression:Hotsauceallocation.Behavior,25,*Lindsay,J.J.(1999).Attributioninhumanaggression:ProductandUnpublisheddoctoraldissertation,UniversityofMissouri—*Llorente,M.,Bernardo,M.,deFlores,T.,&Valdes,M.(1985).TypeAbehaviorandBuss’sinstrumentalaggressionparadigm(BIAP).NervosaSuperior,27,106–109.Lohr,B.A.(1996).Aninvestigationofnonsexualaggressioninsexuallycoercivemen.Unpublisheddoctoraldissertation,UniversityofGeorgia.LoPresto,C.T.,&Deluty,R.H.(1988).Consistencyofaggressive,assertive,andsubmissivebehaviorinmaleadolescents.JournalofSocialPsychology,128,619–632.Lynman,D.R.,&Widiger,T.A.(2001).Usingthefive-factormodeltorepresentthepersonalitydisorders:Anexpertconsensusap-JournalofAbnormalPsychology,110,401–412.Malamuth,N.E.(1988).Predictinglaboratoryaggressionagainstfemaleandmaletargets:Implicationsforsexualaggression.JournalofRe-searchinPersonality,22,474–495.Martin,R.,Watson,D.,&Wan,C.K.(2000).Athree-factormodeloftraitanger:Dimensionsofaffect,behavior,andcognition.JournalofPer-sonality,68,869–895.Martin,R.A.,Kuiper,N.A.,&Westra,H.A.(1989).CognitiveandaffectivecomponentsoftheTypeAbehaviorpattern:Preliminaryevi-denceforaself-worthcontingencymodel.PersonalityandIndividualDifferences,10,Matthews,K.A.(1988).CoronaryheartdiseaseandTypeAbehaviors:UpdateonandalternativetotheBooth-KewleyandFriedman(1987)quantitativereview.PsychologicalBulletin,104,Matthews,K.A.,&Angulo,J.(1980).MeasurementoftheTypeAbehaviorpatterninchildren:Assessmentofchildren’scompetitiveness,impatience-anger,andaggression.ChildDevelopment,51,466–475.McCann,B.S.,Woolfolk,R.L.,Lehrer,P.M.,&Schwarcz,L.(1987).GenderdifferencesintherelationshipbetweenhostilityandtheTypeAbehaviorpattern.JournalofPersonalityAssessment,51,McCord,J.(1991).Familyrelationships,juveniledelinquency,andadultCriminology,29,397–417.McCrae,R.R.,&Costa,P.T.(1985).ComparisonofEPIandpsychoti-cismscaleswithmeasureofthefive-factormodelofpersonality.sonalityandIndividualDifferences,6,Miller,J.D.,Lynam,D.,&Leukefeld,C.(2003).Examiningantisocialbehaviorthroughthefive-factormodelofpersonality.AggressiveBe-havior,29,Morrison,K.A.(1997).Personalitycorrelatesofthefive-factormodelforasampleofbusinessowners/managers:AssociationswithscoresonPERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR vateself-consciousnessonaggressiveandprosocialresponses.andAppliedSocialPsychology,11,387–403.Stanford,M.S.,Greve,K.W.,&Dickens,T.J.(1995).Irritabilityandimpulsiveness:Relationshiptoself-reportedimpulsiveaggression.sonalityandIndividualDifferences,19,Stephens,M.W.,Nelson,D.,&Hudgens,G.A.(1974).Aggressivenessandperformanceinamini-systemcontext(TechnicalMemorandumNo.4–74).Adelphi,MD:U.S.ArmyHumanEngineeringLaboratory.Strayer,J.,&Roberts,W.(2004).Empathyandobservedangerandaggressioninfive-year-olds.SocialDevelopment,13,*Strube,M.J.,Turner,C.W.,Cerro,D.,Stevens,J.,&Hinchey,F.(1984).InterpersonalaggressionandtheTypeAcoronary-pronebehaviorpat-tern:Atheoreticaldistinctionandpracticalimplications.JournalofPersonalityandSocialPsychology,47,839–847.Suls,J.,Martin,R.,&David,J.P.(1998).Person-environmentfitanditslimits:Agreeableness,neuroticism,andemotionalreactivitytointerper-sonalconflict.PersonalityandSocialPsychologyBulletin,24,88–98.Tabachnick,B.G.,&Fidell,L.S.(1989).Usingmultivariatestatistics(2nded.).NewYork:Harper&Row.Taylor,S.P.(1967).Aggressivebehaviorandphysiologicalarousalasafunctionofprovocationandthetendencytoinhibitaggression.ofPersonality,35,Taylor,S.P.,&Hulsizer,M.R.(1998).Psychoactivedrugsandhumanaggression.InR.G.Geen&E.Donnerstein(Eds.),Humanaggression:Theories,research,andimplicationsforsocialpolicy(pp.139–165).SanDiego,CA:AcademicPress.Tiedens,L.Z.(2001).Theeffectofangeronthehostileinferencesofaggressiveandnonaggressivepeople:Specificemotions,cognitivepro-cessing,andchronicaccessibility.MotivationandEmotion,Trull,T.J.,&Widiger,T.A.(1997).StructuredInterviewfortheFive-FactorModelofPersonality(SIFFM):Professionalmanual.FL:PsychologicalAssessmentResources.*Twenge,J.M.,&Campbell,K.W.(2003).“Isn’titfuntogettherespectthatwe’regoingtodeserve?”Narcissism,socialrejection,andaggres-PersonalityandSocialPsychologyBulletin,29,vanGoozen,S.H.M.(1994).Maleandfemale:Effectsofsexhormonesonaggression,cognitionandsexualmotivation.Unpublisheddoctoraldis-sertation,UniversityofAmsterdam,Amsterdam,theNetherlands.*vanGoozen,S.H.M.,Frijda,N.H.,Kindt,M.,&vandePoll,N.E.(1994).Angerpronenessinwomen:DevelopmentandvalidationoftheAngerSituationQuestionnaire.AggressiveBehavior,20,79–100.vanGoozen,S.H.M.,Frijda,N.H.,&vandePoll,N.E.(1994).Angerandaggressioninwomen:InfluenceofsportschoiceandtestosteroneAggressiveBehavior,20,Verona,E.,Patrick,C.J.,&Lang,A.R.(2002).Adirectassessmentoftheroleofstateandtraitnegativeemotioninaggressivebehavior.ofAbnormalPsychology,111,249–258.Wang,M.C.,&Bushman,B.J.(1999).Integratingresultsthroughmeta-analyticreviewusingSASsoftware.Cary,NC:SASInstitute.Watkins,P.L.,Ward,C.H.,Southard,D.R.,&Fisher,E.B.(1992).TheTypeAbeliefsystem:Relationshipstohostility,socialsupport,andlifeBehavioralMedicine,18(1),27–32.Widiger,T.A.,Trull,T.J.,Clarkin,J.F.,Sanderson,C.,&Costa,P.T.Jr.(1994).Adescriptionofthepersonalitydisor-derswiththefive-factormodelofpersonality.InP.T.CostaJr.&T.A.Widiger(Eds.),Personalitydisordersandthefive-factormodelofper-(pp.41–56).Washington,DC:AmericanPsychologicalWiggins,J.A.,&Pincus,A.L.(1994).Personalitystructureandthestructureofpersonalitydisorders.InP.T.Costa&T.A.Widiger(Eds.),Personalitydisordersandthefive-factormodelofpersonality73–93).Washington,DC:AmericanPsychologicalAssociation.Wilcox,R.(1995).Statisticsforthesocialsciences.Orlando,FL:Aca-demicPress.Wilkinson,C.J.(1985).Effectsofdiazepam(Valium)andtraitanxietyonhumanphysicalaggressionandemotionalstate.JournalofBehavioralMedicine,8,Wills,T.A.,Vaccaro,D.,&McNamara,G.(1994).Noveltyseeking,risktaking,andrelatedconstructsaspredictorsofadolescentsubstanceuse:AnapplicationofCloninger’stheory.JournalofSubstanceAbuse,6*Wingrove,J.,&Bond,A.J.(1998).Angryreactiontoafailureonacooperativecomputergame:Theeffectoftraithostility,behaviouralinhibition,andbehaviouralactivation.AggressiveBehavior,24,*Winkel,M.,Novak,D.M.,&Hopson,H.(1987).Personalityfactors,subjectgender,andtheeffectsofaggressivevideogamesonaggressioninadolescents.JournalofResearchinPersonality,21,Wood,M.D.(1996).Alcoholandaggression:Anexaminationoftherolesofcognitivefunctioningandpersonality.Unpublisheddoctoraldisserta-tion,UniversityofMissouri—Columbia.Wood,W.,Wong,F.Y.,&Chachere,J.G.(1991).Effectsofmediaviolenceonviewers’aggressioninunconstrainedsocialinteraction.PsychologicalBulletin,109,Yuen,S.A.,&Kuiper,N.A.(1991).CognitiveandaffectivecomponentsoftheTypeAhostilitydimension.PersonalityandIndividualDiffer-ences,12,*Zeichner,A.,Frey,F.C.,&Parrott,D.J.(2003).Correlatesofdelayedphysicalaggressioninresponsetoprovocation.InternationalReviewofSocialPsychology,16,*Zeichner,A.,Frey,F.C.,Parrott,D.J.,&Butryn,M.F.(1999).Mea-surementoflaboratoryaggression:Anewresponse-choiceparadigm.PsychologicalReports,85,1229–1237.*Zeichner,A.,Parrott,D.J.,&Frey,F.C.(2003).Genderdifferencesinlaboratoryaggressionunderresponsechoiceconditions.AggressiveBe-havior,29,Zelli,A.,&Dodge,K.A.(1999).Personalitydevelopmentfromthebottomup.InD.Cervone&Y.Shoda(Eds.),Thecoherenceofpersonality:Social-cognitivebasesofconsistency,variability,andorganization94–126).NewYork:GuilfordPress.ReceivedDecember1,2004RevisionreceivedDecember17,2005AcceptedJanuary23,2006PERSONALITY,PROVOCATION,ANDAGGRESSIVEBEHAVIOR