OntheDevelopmentofHarmoniousandObsessivePassion:TheRoleofAutonomySuppo - PDF document

OntheDevelopmentofHarmoniousandObsessivePassion:TheRoleofAutonomySupport,ActivitySpecialization,andIdentiÞcationWiththeActivityveA.Mageau,RobertJ.Vallerand,JulieCharest,Sarah-JeanneSalvy,NathalieLacaille, Inrecentyears,researchinpositivepsychology(Seligman&Csiks-zentmihalyi,2000)hasunderscoredtheimportanceofidentifyingthefactorsandprocessesthatmakelifeworthliving.Aspartofthisoverarchingperspective,Vallerandandhiscolleagues(2003)haveproposedthattheconceptofpassiontowardactivitiesmaycontrib-utetooptimalfunctioningandnurtureajoyofliving.Yetresearchonpassionhasshowntheexistenceoftwoformsofpassion,ahar-moniouspassionthatisassociatedwithadaptiveoutcomesandamoreobsessivepassionthatisassociatedwithlessadaptiveandattimesmaladaptiveconsequences(seeRatelle,Vallerand,Mageau,Rousseau,&Provencher,2004;Sevesque,Laliberte,Pelle-tier,Blanchard,&Vallerand,2003;Vallerandetal.,2003).Althoughemergingresearchhaslookedattheconsequencesofpassion,noresearchtodatehasdealtwiththedevelopmentofpassion.Thepurposeofthepresentresearchwastoaddressthisissue.TheConceptofPassionOnPassionTowardActivitiesAlthoughtheconceptofpassionhasgeneratedmuchinterestinphilosophy(seeRony,1990,forareview),ithasreceivedlittleem-piricalattentioninpsychology.Moststudiesonpassionpertaintoromanticrelationships(e.g.,Hat“eld&Walster,1978)orfocusonconceptsakintopassionsuchaspersonalinterests(e.g.,Krapp,2002),seriousplay(e.g.,Rathunde&Csikszentmihalyi,1993),vitalengagement(e.g.,Nakamura&Csikszentmihalyi,2003),personalexpressiveness(Waterman,1990,1993,2004),orundividedactivity(Dewey,1913).Despitetheirdifferentnames,allthoseconceptshaveincommonanengagementtowardanactivity,anemotionalcom-ponentattachedtotheactivity,andsomekindofvaluingoftheac-tivity.However,theseconstructsarelimitedinthattheyonlyapplytopositivetypesofsustainedengagement.Yetphilosophersandre-searchonintenseinvolvement(e.g.,Bonebright,Clay,&Ankenm-ann,2000;Dube,Kairouz,&Jodoin,1997;Glasser,1976;Sacks&Sachs,1981)suggestthatbothaproactiveandreactiveformofpas-sion(oractivityengagement)canbeobserved.Ontheonehand,Aristotle(seeFeertchak,1996)andReneDescartes(1649)arguedthatpassionsarepositiveaslongasreasonunderliesbehaviors.Inthisperspective,passionisperceivedasanecessaryingredientforhigherachievements(Hegel,ascitedinMageau,Vallerand,Charest,etal. Feertchak,1996)andasgeneratingpotentiallyimportantadaptiveandbene“cialoutcomes(seePaturet,2001).Ontheotherhand,others(Lagache,1936;Spinoza,1953)havearguedthatpassionscanleadtonegativeoutcomeswhenpeoplelosecontrolovertheiractivityen-gagement.Empiricalresearchonintenseinvolvementalsoidenti“esproactiveandreactiveformsofactivityengagement.Forexample,SacksandSachs(1981)proposethatpeoplecanbecomeaddictedtotheirexercisetotheextentthattheyneglectotherlifedomains(Rudy&Estock,1990)andexperiencesignsofdeprivationwhentheydonotengageintheiractivity.YetGlasser(1976)alsoproposesapositiveformofaddiction,wherepeopleexperiencefeelingsofeuphoriaand”owwhendoingtheiractivity(e.g.,running).Distinctionsbetweencommitmentandovercommitment(Dubeetal.,1997)andworken-thusiastsandworkaholics(Bonebrightetal.,2000)areotherexamplesofproactiveandreactiveformsofactivityengagement.Althoughimportant,thesedifferentlinesofresearchonintensein-volvementdonotconstitutetheoriesonpassion.Indeed,thisresearchislargelyatheoretical,and,consequently,itdoesnotaddressthepsy-chologicalprocessesunderlyingactivityengagement.Recently,Valle-randandhiscolleagues(2003)proposedadualisticmodelofpassion,wheretwotypesofpassion,harmoniousandobsessive,arede“ned.Thismodelisbasedonpreviousde“nitionsofpositiveformsofsus-tainedengagement(Dewey,1913;Krapp,2002;Rathunde&Csikszentmihalyi,1993;Waterman,1990,1993,2004)butalsoincor-poratesamorereactiveformofactivityengagement.Assuch,itbetterre”ectspreviousphilosophicalworkonpassionandpastresearchonintenseinvolvementthanformermodels.VallerandandhiscolleaguesrelyonSelf-DeterminationTheory(Deci&Ryan,1985,2000)todis-tinguishbetweenaharmoniousandanobsessivetypeofpassionandproposethatthetypeofinternalization(i.e.,autonomousvs.con-trolled)associatedwithpeoplesactivityengagementin”uencesthetypeofpassionthatwillemerge.Vallerandandhiscolleagues(2003)modelalsooffersavalidatedmeasuretoassessthetwotypesofpas-sions.Thismodelwasthuschosenasthetheoreticalframeworkofchoicetostudypassion.TheDualisticModelofPassionVallerandandhiscolleagues(2003;Vallerand&Houlfort,2003;Vallerand&Miquelon,2007)de“nepassionasastronginclinationOntheDevelopmentofPassion towardanactivitythatindividualslike(orevenlove),thattheyvalue,andinwhichtheyinvesttimeandenergy.Thedimensionofvalueisespeciallyimportantfortheconceptofpassionbecauseitseparatespassionateactivitiesfromotherinterestingbutunimpor-tantactivitiestowardwhichpeopleareintrinsicallymotivated.InlinewithSelf-DeterminationTheory(Deci&Ryan,1985,2000),Vallerandandhiscolleagues(2003)furtherproposethatpeoplecanvaluetheiractivityforautonomousorcontrolledreasons,whichre-sultsintwodistincttypesofpassion.Self-DeterminationTheory(Deci&Ryan,1985,2000)positsthatpeoplearenaturallyinclinedtoassimilateandintegrateexternalbehavioralregulationsinordertoexperienceself-determination.However,fortheinternal-izationprocesstofunctionoptimally,peopleneedtobeinsocialenvironmentsthatnurtureinnateneedsforautonomy,competence,andrelatedness.Incontrolling,overdemanding,andrejectingenvi-ronments,theintegrationprocessishinderedanddefensiveorself-protectiveprocesseswilloccur.Inlinewiththisdialecticalview,Vallerandandhiscolleagues(2003)proposethat,dependingonthecontextinwhichtheinternalizationofbehavioralregulationsoccurs,peoplewillvaluetheiractivityformoreautonomousreasons(becauseoftheinherentsatisfactionitbrings)ormorecontrolledreasons(becauseitservesimportantcompensatoryfunctions),whichinturnshouldgenerateamoreharmoniousorobsessivepassion,Harmoniouspassionisthushypothesizedtoresultfromanau-tonomousinternalizationofbehavioralregulations.Thistypeofin-ternalizationproducesastrong,butcontrollable,desiretoengageintheactivity,whichengendersasenseofvolitionandpersonalen-dorsementaboutpursuingtheactivity.Theimportanceoftheac-tivityisthusfreelyacceptedwithoutanycontingenciesattachedtoit.Theactivityiscoherentwithotheraspectsoftheselfandisfullyintegratedinthepersonsauthenticself(Deci&Ryan,2000).AsSelf-DeterminationTheory(Deci&Ryan,1985,2000)suggests,suchinternalizationshouldoccurinenvironmentswhereindividualsfeelautonomous,connected,andcompetent.Peoplewithaharmoniouspassionareautonomousintheirregulations,andtheypartakeintheirpassionateactivityaswellasinotheractivitieswithanopen-nessthatisconducivetopositiveexperiences(Hodgins&Knee,2002).Thepersoncanthusengageintheactivityina”exibleandmindfulmanner.Mageau,Vallerand,Charest,etal. Ontheotherhand,obsessivepassionishypothesizedtoresultfromalackofful“llmentofintrinsicneedsthatleadstotheinter-nalizationofintra-orinterpersonalpressuresorboth.Thesepres-surescancomefromcontingenciesthatareattachedtotheactivity(e.g.,feelingsofsocialacceptanceorself-esteem),fromanuncon-trollablesenseofexcitementderivedfromactivityengagement,orfrompressuringsocialenvironments.HodginsandKnee(2002)pro-posedthat,innonsupportiveenvironments,peoplewilldevelopanego-investedorfalseself(Ryan,1995)thatisbasedonexternalcon-tingencies.Suchego-investedself-structuresinturnservetodefendandprotectthepersonsself-worth.InlinewithHodginsandKneespropositions,Vallerandandhiscolleagues(2003)suggestthatinthecaseofanobsessivepassion,engagingintheactivitybecomesasubstituteforself-worth.Assuch,theactivityiseagerlyandenthu-siasticallypursuedbecauseofthereassuranceandego-af“rmingmomentsitprovides.However,asHodginsandKneesuggest,whenanactivityisegoinvesteditwillberigidlypursued,resultinginlessthanoptimalfunctioning.Whenanactivityisvaluedbecauseitservesdefensiveandprotectivefunctions,itisnoteasilyputaside.Peoplewithanobsessivepassionthusfeelcompelledtoengageintheiractivity.TheirpassionmustrunitscourseasitcomestocontrolResearchonPassionResearchonpassionprovidessupportfortheDualisticModelofPassionaswellasforthereliabilityandvalidityofthePassionScale(Vallerandetal.,2003).ThePassionScalehastwocomponents:athree-itemscalethatdistinguishesbetweenpassionateandnonpas-sionateindividualsandtwoseven-itemsubscalesthatassesstherel-ativeimportanceofharmoniousandobsessivepassion.Thethree-itemscaleformsonefactorandmeasuresthethreecomponentsofthede“nitionofpassion,namely,activityvaluation,lovefortheac-tivity,andtimespentontheactivity.Themiddleoftheresponsescalecanbeusedasthecutoffpointtodistinguishbetweenpassion-ateandnonpassionateindividuals(Vallerand&Houlfort,2003).TheDualisticModelofPassion(Vallerandetal.,2003)proposesthatifoneofthesecriteriaismissing,thepersonsrelationshipwiththeactivityshouldnotbeconsideredpassionate.Forexample,ifapersonlikesanactivity,investstimeinit,butdoesnotvalueit,theOntheDevelopmentofPassion activitycanbeconsideredasbeingintrinsicallymotivating,butitisnotapassionyet.Thesecondpartofthescaleassessesthephenom-enologicalexperienceofthetwotypesofpassion.Theharmoniouspassionsubscalefocusesontheopennesstoexperiencethatpeoplefeelwhentheyengageintheiractivity,whereastheobsessivepassionsubscaleassessesthefeelingofobligationandobsessionassociatedwiththeactivity.Exploratoryfactoranalysesshowedthattheitemsofthetwopassionsubscalescanbeseparatedintwodistinctfactors.Thistwo-factorsolutionwasthencon“rmedusingcon“rmatoryfactoranalyses(Vallerandetal.,2003).Theconvergentvalidityofthepassionscalewasalsocon“rmed.Althoughqualitativelydifferent,boththeharmoniousandobsessivesubscalesmeasureamoregeneralconceptofpassion,becausetheyarebothpositivelyandequallycorrelatedwithlikingoftheactivity,timespentontheactivity,activityvaluation,andperceptionoftheactivityasbeingpassionate(Vallerandetal.,2003;Vallerand&Houlfort,2003).Theyarealsoassociatedwithdifferentregulatorystyles(Vallerandetal.,2006).Peoplewhohavemoreautonomousbehavioralregulationsalsoexperienceamoreharmonioustypeofpassion,whereaspeoplewhoregulatetheirbehaviorsmostlyinre-sponsetoexternalpromptsorintrojectstendtohaveamoreobses-sivepassion.Theseresultssupportthepropositionthatharmoniousandobsessivepassionresultfromanautonomousandcontrolledinternalizationofbehavioralregulations,respectively.Finally,thetwotypesofpassionareassociatedwithdifferentaffec-tive,cognitive,andbehavioraloutcomesthataretheoreticallycom-patiblewiththeirconceptualde“nition.Forexample,obsessivepassionisrelatedtoamorerigidandcon”ictingformoftaskpartic-ipationthatpreventsthepersonfromfullyfocusingonthetaskathand,interfereswiththeexperienceofpositiveaffectand”ow,andfacilitatesnegativeaffectduringactivityengagement(Vallerandetal.,2003).Peoplewithamoreobsessivepassionexperiencemorecon-”ictswithotherlifeactivities(Sevesqueetal.,2003),andtheytendtoengageintheiractivityevenwhencircumstancessurroundingactivityengagementwouldsuggestotherwise(Rip,Fortin,&Vallerand,2006;Rousseau,Vallerand,Ratelle,Mageau,&Provencher,2002;Vallerandetal.,2003,Studies3and4).Obsessivepassionisalsopositivelycorrelatedwithruminativethoughtsabouttheactivity(Ratelleetal.,2004)andlowerlevelsofpositiveaffect(Mageau&Vallerand,2007)whenpeoplearepreventedfromengaginginit.Mageau,Vallerand,Charest,etal. Incontrast,theaffective,cognitive,andbehavioralcorrelatesofharmoniouspassionre”ectthemoreautonomousbehavioralregu-lationsunderlyingthistypeofpassion.Positiverelationshipshavebeenfoundbetweenharmoniouspassionandmeasuresof”owandpositiveaffectexperiencedduringactivityengagement(Mageau,Vallerand,Rousseau,Ratelle,&Provencher,2005;Vallerandetal.,2003,Study1;Vallerandetal.,2006)aswellasgeneralsubjec-tivewell-being(Vallerandetal.,2003,Study2;Vallerandetal.,2007).Inaddition,harmoniouspassionisunrelatedtonegativeout-comessuchasrumination(Ratelleetal.,2004),rigidpersistence(Ripetal.,2006;Vallerandetal.,2003,Study3)orcon”ictswithotherlifedomains(Sevesqueetal.,2003).Instead,peoplewithahar-moniouspassiondisplaya”exibleformofactivityengagement,wheretheypersistinthepassionateactivityonlyifpositivereturnsareexpected.Ifconditionsbecomepermanentlynegative,behavioralinvolvementstops.Overall,researchonpassionsupportsthedualisticmodelofpas-sion(Vallerandetal.,2003).ThisresearchisalsocoherentwithSelf-DeterminationTheoryresearch(Deci&Ryan,1985,2000),whichshowstheimportanceofautonomousregulationsforpeoplespsy-chologicalwell-being.Asisthecaseforautonomousformsofself-regulations(e.g.,self-determinedmotivation,Deci&Ryan,1985,2000;self-concordance,Sheldon&Elliot,1998,1999;intrinsicgoals,Kasser&Ryan,2001),harmoniouspassionisrelatedtoheightenedlevelsofwell-beingandoptimalfunctioning.Althoughpassionissimilartotheseothermotivationalconstructsinthattheyallre”ectautonomousfunctioning,harmoniouspassiondiffersfromthemboththeoreticallyandempirically.Indeed,harmoniouspassionre-ferstoactivitiesthataredeeplyvaluedbytheindividualandassuchitshouldhaveamoreprofoundimpactonpeopleslivesthanmo-tivationorgoals.Empiricalevidencecon“rmsthatwhenself-deter-minedmotivationandharmoniouspassionareassessedtowardagivenactivity(i.e.,playingfootball),harmoniouspassionpredictspositiveandnegativeaffectsaboveandbeyondwhatispredictedbymotivation(Vallerandetal.,2003,Study2).Inaddition,althoughresearchhasjustbeguntounravelthedifferencesbetweenmotiva-tion,self-concordance,intrinsicgoals,andharmoniouspassion,available“ndingsonpassion(Vallerandetal.,2003,Study1)sug-gestthatpassioninvolvesidentityprocessesthatarenotnecessarilypresentinothermotivationalconstructs.Forexample,peoplewithOntheDevelopmentofPassion highlevelsofharmoniousorobsessivepassiontendtoperceivetheactivityasself-de“ning,whereaspeoplehighlymotivatedtowardanactivitydonotnecessarilyidentifywiththeactivity.Identitypro-cesseswillbefurtheraddressedwhenwediscussthedevelopmentofFinally,researchonpassionisimportantbecauseitgeneratesnewandexcitingresearchavenuesbyencouragingresearcherstostudypeoplesexperiencesinmoreextremecontexts.Forexample,whereasSelf-DeterminationTheoryresearchshowsthatautonomousmoti-vations(Deci&Ryan,2000)andgoals(Houser-Marko&Sheldon,2006)arerelatedtopersistencetowardanactivitywhenpeoplefacechallengesorboredom,researchonpassionshowsthatwhencir-cumstancessurroundingactivityengagementmakeitilladvisedorcounterproductivetopursuetheactivity,autonomouslyregulatedpassionwillnotberelatedtopersistence.Conversely,itisobsessivepassion,whichinvolvesamorecontrolledformofregulation,thatwillpredicthazardousorunhealthypersistence(Ripetal.,2006;Vallerandetal.,2003,Study3).Theseresultsarenotcontradictorybecausethecontextinwhichpersistenceismeasuredgreatlydiffersacrossthesedifferentstudies.Infact,Self-DeterminationTheorywouldpredictthatmoreautonomouspeoplearebetterequippedtoevaluatethecircumstancessurroundingtheiractivityengagementinmoreextremecontextsandtointerrupttheirparticipationwhenap-propriate.Conversely,peoplemotivatedthroughintrojectsorexter-naldemandsshouldpersistintheactivityaslongastheexternalcontingenciesthatmotivatethemarepresent,regardlessofcontexts.Theseresultsneverthlessshowthatstudyingpassioncanofferanewperspectiveonagivenphenomenonanditsrelationshipwithauton-omouslyregulatedconstructs.OntheDevelopmentofPassionTheDualisticModelofPassion(Vallerandetal.,2003)positsthattherearefouressentialelementstothede“nitionofpassion:activityvaluation;likingoftheactivity;timespentontheactivity,whicharecommontobothtypesofpassion;andtheinternalizationprocess(autonomousvs.controlled),whichdistinguishesbetweenharmoni-ousandobsessivepassion.Vallerandandhiscolleagues(2003)pro-posethatpersonalandcontextualfactorsthatin”uencetheseelementsshouldcontributetothedevelopmentofpassion.WhereasMageau,Vallerand,Charest,etal. anincreasingamountofresearchhaslookedatthedifferentialim-pactofharmoniousandobsessivepassiononvariousoutcomes,verylittleresearchhastestedthevariablesinvolvedinthedevelopmentofpassion.Thepresentpaperfocusesonthreepersonalandcontextualvariables(i.e.,identi“cationwiththeactivity,participantsandpar-entspreferencesforactivityspecialization,andparentsvaluationoftheactivity)thatshouldin”uenceparticipantspassionthroughtheirhypothesizedimpactonactivityvaluationandtimespentontheac-tivity.Furthermore,buildingonSelf-DeterminationTheoryresearch(Deci&Ryan,1985,2000)andmorespeci“callyonGrolnickandhercolleagues(Grolnick&Ryan,1989;Grolnick,Ryan,&Deci,1991)workontheimpactofautonomysupportontheinternalizationprocess,weexaminewhetherthequalityoftheinterpersonalcontext(i.e.,autonomy-supportivevs.controlling)distinguishesthetwotypesofpassion.Identi“cationwiththeactivity.Vallerandandhiscolleagues(2003)suggestthatapersonsidenti“cationwiththeactivity,throughitsimpactonactivityvaluation,isoneofthekeyprocessesthroughwhichaninterestingactivitybecomesapassion.Whenanactivitygreatlyresonateswithapersonssenseofself,thatpersonbeginstothinkofhimselforherselfintermsofthisactivity(Aron,Aron,&Smollan,1992).Itishypothesizedthatindividualswhocometoseeanactivityascontributingtotheiridentity(Schlenker,1985),orashavingthepotentialtodosoeitherintheshortorlongterm,arelikelytovaluetheactivity,spendtimeengaginginit,andthusbecomemorepassionatetowardit.Identityreferstoanindividualsrelevantfeatures,characteristics,andexperiences;howtheseareinterrelated;aswellasthesocialandself-regulationfunctionsthatsuchfeaturesserve(Schlenker,1985).Identityimagesarehypothesizedtoaffectthepersonsthoughts,affect,andbehaviorsasafunctionoftheircen-trality,importance,andsalience(e.g.,Cantor&Mischel,1979;Greenwald,1980;Markus,1977).Passionsareassumedtocomefromimportantidentityimages,andassuchtheybecomepartofwhopeopleare.Peoplewithapassionforswimmingorforwritingdonotmerelyswimorwrite.Theyareswimmersorwriters.Inhisworkonpersonalexpressiveness,aconceptsimilartohar-moniouspassion,Waterman(1990,p.47)alsoproposedthatpersonallyexpressiveactivitiesinvolveafeelingofspecial“tormeshingwithanactivityandanimpressionthatthis[activity]isOntheDevelopmentofPassion whatthepersonwasmeanttodo.Watermanfurthersuggeststhatpersonallyexpressiveactivitiesareactivitiesthatareconsistentwithapersonstrueself.Addingtothislineofwork,weproposethatapersoncanalsostronglyidentifywithanactivityfornon-self-deter-minedreasons,suchasexternalcontingenciesordefensivemecha-nisms(Hodgins&Knee,2002),resultinginanother,moreobsessivetypeofpassion.Researchonpassionconductedwithcollegestu-dentscon“rmsthatbothtypesofpassionarepositivelycorrelatedwithameasureofactivityinclusionintheself(e.g.,Aronetal.,1992),althoughthisrelationshipisstrongerforobsessivepassion(Vallerandetal.,2003).Thepresentpaperaimsatreplicatingthisrelationshipwithmorediversesamples.Participantsandparentspreferenceforactivityspecializationandparentsactivityvaluation.Anotherfactorthatshouldin”uencethevaluationoftheactivityispeoplesvaluesregardingtheimportanceofspecializinginagivenactivity.Althoughthishypothesisismoreexploratory,itmakessensetothinkthatpeoplewhostronglybelieveinvaluesofdedicationandexcellenceshouldbemorelikelytofocusononespeci“cactivity.Suchfocusedattentionshouldincreasetimespentontheactivityaswellastheactivitysvalue,thusin”uencingpassion.Inaddition,thesocialenvironmentshouldalsoplayanimportantroleinleadingchildrentovalueagivenactivityandspendtimeengaginginit(Eccles&Wig“eld,2002).Ac-cordingtoBloom(1985a),parentsinvolvementinchildrensparticipationiscrucialtothedevelopmentofexpert-levelperformersbecauseparentscreateacontextwherechildrencanengageindeliberatepractice,which,ininteractionwithchildrensinitialtalent,leadstohigherlevelsofperformance.Otherresearchershavealsounderscoredtheimportanceofparentalinvolvementatthebe-ginningofchildrensparticipation(Snyder&Spreitzer,1973;Spreitzer&Snyder,1976)andinmaintainingactivityinvolvement(Brown,Frankel,&Fennell,1989).Inthepresentpaper,wearguethatsigni“-cantotherswhounderscorethevalueoftheactivityandencouragespecializationinthisactivityshouldin”uencechildrensvaluationoftheactivity,increasetimespentontheactivity,andthusfacilitatethedevelopmentofpassiontowardtheactivity.Autonomysupport.Althoughparentalinvolvementhasmostoftenbeendiscussedasaunidimensionalconstruct,parentsandadultsinMageau,Vallerand,Charest,etal. generalhavebeenknowntocreatedifferentcontextsofactivityen-gagementthroughthequalityoftheirinvolvement(Grolnick,Ben-jet,Kurowski,&Apostoleris,1997).Whereassomehighlyinvolvedparentsmaybeperceivedbytheirchildrenasbeingoverinvolved,otherequallyinvolvedparentsareviewedasbeinginvolvedjustenough(Stein,Raedeke,&Glenn,1999).Parentsoverinvolvementhasbeenlinkedtoathletesstressandburnout(Brustad,1988;Go-uld,Tuffey,Udry,&Loehr,1997;Scanlan,Carpenter,Lobel,&Si-mons,1993;Steinetal.,1999).Yet,asSteinandhiscolleagues(1999)andothers(e.g.,Weiss&Hayashi,1995)havesuggested,overin-volvedparentsdonotnecessarilydevotemoretimetotheirchildrenthanotherparents;itisthequalityoftheirinvolvementthatdiffers.Focusingonthequalityofinvolvement,GrolnickandRyan(1987,1989)proposedthathighlyinvolvedparents(orcoaches)differfromoverinvolvedoneswithrespecttotheamountofautonomysupport(vs.control)theyconvey.Adultscaneithersupportorthwartchildrensperceptionofautonomy,whichinturnenhancesorim-pedeschildrensinternalizationprocessandgeneralpsychologicalTheautonomy-supportivestylereferstoadultswhoconsiderchil-drentobeseparateindividualshavingtherightandtheneedtoex-presstheirfeelingsandpreferences.Autonomy-supportiveadultsplacevalueonself-initiationandencouragechoiceandparticipationindecisionmaking.Theyalsoacknowledgethechildsfeelingsandexplainthereasonsbehindtheirrulesanddemands(seeMageau&Vallerand,2003,forareviewofautonomy-supportivebehaviors).Conversely,acontrollingstylereferstoadultswhoplacevalueoncontrol,applyingpressurestomakechildrenthink,feel,andbehaveinspeci“cways.Decadesofresearchnowsupportthebene“cialin-”uenceofautonomysupportonchildrensdevelopment.Forexam-ple,autonomysupportfromparents,teachers,andcoacheshasbeenrepeatedlyrelatedtoenhancedinterest(e.g.,Grolnick&Ryan,1987),intrinsicmotivation(e.g.,Black&Deci,2000),cre-ativity(Koestner,Ryan,Bernieri,&Holt,1984),perceivedschoolcompetence(e.g.,Grolnick&Ryan,1989),andschoolperformance(e.g.,Boggiano,Flink,Shields,Seelbach,&Barrett,1993),whereascontrollingbehaviorswereshowntoimpedethesepositiveoutcomes.Ofparticularinteresttothepresentpaper,autonomysupporthasbeenrepeatedlyrelatedtomoreautonomousformsofinternalizationofbehavioralregulations(e.g.,Black&Deci,2000;Grolnick&Ryan,OntheDevelopmentofPassion 1989;Grolnicketal.,1991).Becauseautonomousandcontrolledreg-ulationsarehypothesizedtodifferentiatebetweenthetwotypesofpassion,thesocialcontext,throughitsimpactontheinternalizationprocess,shouldalsoin”uencethetypeofpassionthatwilldevelop.Speci“cally,autonomy-supportiveadultsshouldfosterharmoniouspassionsbycreatingacontextofexplorationandchoice,wherechil-drencanfreelyendorsetheimportanceoftheactivity.Conversely,controllingadultsarehypothesizedtosetthestageforthedevelopmentofobsessivepassionsbyimposingadditionalpressuresandcreatingself-activitycontingencies.ThePresentResearchBloom(1985a,1985b)suggestedthatactivityengagementcanbedi-videdintothreephases,thenovice,intermediate,andexpertphases,andthatmotivationandpersistenceineachofthesephasesarein-”uencedbydifferentvariables.Wehypothesizedthatthedetermi-nants(orcorrelates)ofpassionmayalsodifferacrossthesestages.WethussoughttopredictpassioninsamplesrepresentingeachofBlooms(1985b)phasesofactivityengagement.Blooms“rstphaseischaracterizedbythepersonsintroductiontotheactivityandin-volvesmostlyexplorationandplay.Thesecondphasebeginswiththestartofinstructionintheactivityandconsistsofanextendedperiodofdeliberatepractice.Thissecondleveltypicallytakesplaceafterseveralmonthsofactivityinvolvement.Finally,thethirdphasestartswhenthepersonpursuestheactivityonafull-timebasisanddemonstratesexpert-levelperformance.Studies1and2ofthepres-entpaperinvolveexpertandintermediateperformers,respectively,andStudy3focusesonchildrenwhohadneverengagedintheac-tivitypriortothestudy(noviceperformers).Usingthe“rstpartofthepassionscale,whichmeasurestheoperationalde“nitionofpas-sion(anactivitythatoneloves,“ndsimportant,andengagesinonaregularbasis;Vallerandetal.,2003),wedistinguishedparticipantswhowerepassionatefromthosewhowerenotateachofthethreephases.Participantsweredividedusingthemiddleoftheresponsescaleasthecutoffpoint,suchthatthenumberofpassionateversusnonpassionatepeoplecouldbecomparedacrossstudies.Itwashypothesizedthatmostperformersattheexpertandintermediatelevels(Studies1and2)wouldbepassionatebecauseofthehighin-Mageau,Vallerand,Charest,etal. volvementitentails,whereasthereversetrendwasexpectedinthenovicesample(Study3).Moreharmoniouslypassionateandobsessivelypassionateindi-vidualswereidenti“edusingtheirrelativestandardizedscoresonthetwopassionsubscales.Thisstrategywasusedbecausemostpassion-ateindividualsstronglyendorsethepassionateitemsfrombothsubscales,makingitimpossibletodistinguishpureharmoniouslypassionatepeople(thosewhodidnotalsoendorseitemsfromtheobsessivesubscale)frompureobsessivelypassionpeoplewithoutlosingthemajorityofthesampletoamixedgroup.Groupswerethusformedbycreatingtwoequallylargepassionategroupsineachsample,thosewhoweremoreharmoniousandthosewhomoreobsessive.Usingpersonal/contextualvariablesanddiscriminantfunctionanalyses,wethendifferentiatednonpassionateindividualsfrompas-sionateonesaswellasdistinguishedbetweenmoreharmoniousandmoreobsessivepassionatepeople.Speci“cally,weinvestigatedpeo-plesidenti“cationwiththeactivity(Studies1and3),theirprefer-enceforactivityspecialization(Studies2and3),andtheirsigni“cantotherspreferencesforactivityspecialization(Studies2and3),val-uationoftheactivity(Study3),andautonomysupport(allthreestudies)aspotentialdeterminantsofpassioningeneralandharmo-niousandobsessivepassioninparticular.STUDY1ThepurposeofStudy1wastotestsomeoftheabovehypothesesaboutthedeterminants(orcorrelates)ofpassioninanexpertsam-ple.FollowingBlooms(1985b)propositions,peoplewereconsid-eredtobeintheexpertphaseiftheyhadbeenperformingtheiractivityforseveralyears,iftheydevotedaconsiderableamountof1.Usingagroupapproachwasessentialtoexaminedifferencesbetweennon-passionate,harmoniouslypassionate,andobsessivelypassionateindividuals.Inthepast,researchonpassionhasfocusedoncorrelatesofthetwopassionsub-scales.Althoughvalid,thisapproachinvestigatesdifferencesbetweenpeoplewhohavehighscoresononetypeofpassion(harmoniousorobsessive)andthosewhodonot.Moreharmoniouslypassionatepeoplearethusneveractuallycomparedtomoreobsessivelypassionateones.Bypredictinggroupmembership,thepresentresearchoffersanewwaytolookatpossibledifferencesbetweennonpassionate,harmoniouslypassionate,andobsessivelypassionateindividuals.OntheDevelopmentofPassion timetoit,andiftheydisplayedhigh-levelperformance.Muchre-searchsuggeststhatpeoplewhoreachtheexpertlevelhaveonav-erageof10yearsofdeliberatepracticeintheactivity(e.g.,Ericsson&Charness,1994).Suchwasthecaseoftheparticipantsinthepresentstudy.Wehypothesizedthatmostifnotallparticipantswouldbepassionatefortheiractivityandthatharmoniouslypas-sionateexpertscouldbedistinguishedfromobsessivelypassionateonesonthebasisofthelevelofautonomysupporttheyexperiencedintheirsocialenvironmentandthesenseofidentitytheyderivedfromtheirpassionateactivity.Peoplessocialenvironmentincludedcoaches(ormusicteachers)aswellasparentsbecausebothsourcesofin”uencehavebeenshowntoaffectpeoplesactivityengagement(e.g.,Ommundsen&Vaglum,1991).Becauseautonomy-supportivecontextisassociatedwithamoreautonomousinternalizationpro-cess,itwaspredictedthatexpertswithaharmoniouspassionshouldhaveamoreautonomy-supportiveenvironmentthanexpertswithanobsessivepassion.Furthermore,althoughidenti“cationwiththeac-tivityshouldin”uencebothtypesofpassionthroughitsimpactonpeoplesactivityvaluationandtimespentontheactivity,itwashy-pothesizedthatidenti“cationwiththeactivitymaydifferentiatebe-tweenharmoniousandobsessivepassionattheexpertlevel.Peoplewhode“nethemselvessolelyononeactivitymightbemoreatrisktobeoverlydependentontheiractivityforself-de“nition,thusdevel-opingamoreobsessivetypeofpassion.Thishypothesisisinlinewithpastresearchthatshowsthatobsessivepassionismorestronglyrelatedtoidenti“cationmeasuresthanharmoniouspassion(Valle-randetal.,2003,Study1)MethodParticipantsandProcedureThesamplewascomposedof84musiciansand145athletes(i.e.,116swimmersand29skiers)withameanageof19.6years(range12to38yearsold).These144womenand85mendisplayedfull-timecommitmenttowardtheiractivityandanexpert-levelperformance.Mostathletespracticedtheirsportatthenationallevelandmostmusicianswereplayingamusicalinstrumentatthebachelordegreelevelinahighlyselectivefacultyofmusic.Theyhadbeencompetingorpubliclyperformingforanaverageof9yearsand8months(range1yearand6monthsto26years)withonly3participantsreportingunder3yearsofexperience.Onaverage,participantswerepracticing20.9hoursaweek.Mageau,Vallerand,Charest,etal. Participantswerecontactedatsummertrainingcamps,wheretheycom-pleteda20-minsurvey.Forswimmers,thedatacollectiontookplaceatanationalcompetitionduringthesummer.Musicianswereparticipantsataninternationalsummermusiccamp,andskierscompletedthesurveysduringtheirtrainingcamp.Allparticipantssignedaninformedconsentformandreceived$10(Canadian)uponreturnofthequestionnaire.MeasuresParticipantscompletedmeasuresofpassion,senseofidentityderivedfromtheactivity,andperceivedautonomysupportfromsocialsurround-ings.Demographicvariablessuchasageandgenderwerealsomeasured,alongwithcompetitiveorpublicperformanceexperience,weeklyprac-tice,andlevelofactivityengagement(e.g.,nationallevel).HarmoniousandobsessivepassionsweremeasuredusingthePassionScale(Vallerandetal.,2003).Participantswere“rstaskedtothinkabouttheiractivityandthentocompletethetwosectionsofthePassionScaleaccordingtoit.The“rstcomponentassessestheextenttowhichpeoplehaveapassionfortheactivity.Thelevelofpassionismeasuredwiththemeanofthethreecri-terionitemsdealingwiththede“nitionofpassion.Speci“cally,partici-pantsareaskedtoreporttheextenttowhichtheyvaluetheactivity,devotetimeandenergytoit,andloveit.Thesethreeitemswereinter-correlatedinthepresentstudy(.71;seeTable1).Thesecondcom-ponentofthePassionScaleassessesharmoniousandobsessivepassionswithtwoseven-itemsubscales.AsampleitemforharmoniouspassionisMyactivityisinharmonywiththeotheractivitiesinmylife,andasampleitemforobsessivepassionisIhaveatoughtimecontrollingmyneedtodothisactivity.ThePassionScalehasbeenshowntobestruc-turallysound,theoreticallyvalid,andreliable(Vallerandetal.,2003).Inthepresentstudy,thetwotypesofpassionshowedadequatereliability,asindicatedbyCronbachsalphasof.77and.82forharmoniousandob-sessivepassions,respectively.Thesenseofidentityderivedfromtheactiv-wasmeasuredusingtwoitems.The“rstitemconsistedofpresentingparticipantswithsevenpairsofoverlappingVenn-likediagrams(Aronetal.,1992),whereonecirclerepresentedtheperson(ME)andtheothertheactivity(ACTIVITY).Eachpairofcirclesdifferedinthepercentageofoverlaptheyshared.Whereasthe“rstpairdidnotoverlap,theseventhpairrepresentedclosetototaloverlap.Participantswereaskedtoselectthediagramthatbestdescribedthedegreetowhichtheactivitywasacentralpartofwhotheywere.Thisitemhadbeensuccess-fullyusedinpreviousresearch(Vallerandetal.,2003,Study1).Thesec-2.AllthreestudieswereconductedinFrench.OntheDevelopmentofPassion Table1Study1:MeansandStandardDeviationsfortheTwoPassionGroupsandCorrelationsBetweenAllVariablesVariablesMeans()Total(HarmoniouslyPassionatePassionateCorrelations1.2.3.4.5.6.7.8.9.10.1.Harmoniouspassion5.99(0.61)5.37(0.94).77„.53.05.08.04.322.Obsessivepassion4.06(1.13)5.03(1.12).82„.43.03.14.04.503.Lovefortheactivity6.34(0.94)6.12(1.11)„„.63.01.10.06.134.Activityvaluation6.54(0.82)6.31(0.83)„„.41.02.08.04.255.Timeinvestment6.26(1.03)6.34(1.04)„„.05.12.096.Competitiveorpublicperformanceexperience(years)10.0(4.2)9.3(3.7)„„.08.38.08.107.Practice(hoursper20.5(7.7)21.3(7.5)„„.25.01.138.Level(3nationallevel)3.2(0.76)3.1(0.81)„„9.Senseofidentityderivedfromtheactivity(standardized).06(0.89)0.15(0.81).67„10.Perceivedautonomysupportfromsocialnetwork5.64(0.79)5.28(0.84).63 onditemaskedparticipantstoevaluate,inpercentage,thedegreetowhichtheactivityde“nedwhotheywere.Boththediagramandpercent-ageitemswerecorrelated(.51).Theywerethusstandardizedandav-eragedtocreateasinglescoreofidenti“cationwiththeactivity.Perceivedautonomysupportwasmeasuredwitha“ve-itemscale.Thismeasureevaluatedwhetherparentsandcoacheswereperceivedassupportingparticipantsautonomyregardingtheiractivity.AsampleitemisInmyrelationshipwithmyparents,Ihaveasayinwhathappensandcanvoicemyopinionsregardingmyactivity.Participantswereaskedtoratetheextenttowhicheachitemwastrueforthemusinga7-pointLikert-typescale,whichrangedfrom1(notatalltrue)to7(verytrue.63).ResultsandDiscussionInlinewithpreviousresearch(Vallerand&Houlfort,2003),atwo-stepprocedurewasusedtocreatethreedifferentgroups:nonpas-sionate,harmoniouslypassionate,andobsessivelypassionate.First,thethreecriteriaofpassion(i.e.,activityvaluation,timeinvestment,andlovefortheactivity;Vallerandetal.,2003)wereusedtodiffer-entiatebetweenthenonpassionateandpassionatepeople.ualswhosemeanscoreonthethreepassioncriteriawassituatedatmidpointoraboveontheresponsescalewereclassi“edaspassion-ate.Aswouldbeexpected,resultsshowedthat100%ofthepresentsampleofexpertperformerswaspassionateabouttheiractivity.Second,harmoniouslypassionatepeopleweredifferentiatedfromtheobsessivelypassionateonesbycomparingstandardizedscoresonthetwopassionsubscales.Peoplewereclassi“edinthepassiongroupcorrespondingtotheirhigheststandardizedscoreonthetwosub-scales(seeVallerand&Houlfort,2003).Thescoreswerestandard-izedbecause,althoughmostpeoplescoredhigherontheharmonioussubscalethanontheobsessivesubscale,somepeoplescorednotice-ablyhigheronobsessivepassionthanotherpeople.Standardizingthescoresallowedustocapturethesebetween-subjectdifferencesandclassifypeopleintheirappropriatepassiongroup(seeKoestner&Zuckerman,1994,forasimilarprocedure).Theharmoniousand3.Correlationscon“rmedthatbothharmoniousandobsessivepassionarestronglyandsigni“cantlyrelatedtothethreepassioncriteriaofactivityvalua-tion,timeinvestment,andlovefortheactivityinallthreestudies(seeTables1,3,and5).4.Dividingparticipantsintofourgroupsusingthemedianontheharmoniousandobsessivesubscaleswouldyieldthefollowinggroupsizes:relativelynonpas-OntheDevelopmentofPassion obsessivepassiongroupsdidnotdifferasafunctionofage,gender,competitiveorpublicperformance,levelsofperformance,andnum-berofhoursofweeklypractice.Meansandstandarddeviationsforeachgroup,aswellascorrelationsbetweenallvariables,arepre-sentedinTable1.Finally,adiscriminantfunctionanalysiswasperformedinordertodistinguishbetweentheharmoniousandobsessivepassiongroups.Resultsrevealedonesigni“cantdiscriminantfunction,Ca-nonicalCorrelation.24,Wilksswhichshowedthatperceivedautonomysupportandasenseofiden-tityderivedfromtheactivityseparatedtheobsessivelypassionatefromtheharmoniouslypassionategroup.TheseresultsarepresentedinTable2.Correlationsbetweenthepredictorsandthediscriminantfunctionshowedthattheharmoniouslypassionateindividualsweremorelikelytoexperiencehigherlevelsofautonomysupportfromtheirsocialenvironment(.87)thanthoseintheobsessivepassiongroup.Conversely,themorepeoplederivedasenseofidentityfromtheiractivity(.51),themorelikelytheyweretohaveanobsessiveThepresentresultshighlightthreemajorpoints.First,theysuggestthatadultautonomysupportishigherforharmoniouslypassionateexpertsthanobsessivelypassionateones.Socialenviron-mentsthatprovideautonomysupportmaythusservetopromoteharmoniouspassionattheexpertlevel.Conversely,controllingadultswhodirectlypressureotherstopursueanactivityortoengageinitincertainwaysmaysetthestageforamoreobsessiveformofpassion.Thesecondimportant“ndingofthisstudyisthatsionategroup73,harmoniouslypassionategroup35,obsessivelypassionate42,andmixedgroup79.Ascanbeobserved,mostpassionatepeoplestronglyendorsetheitemsfrombothsubscalesandcannotbedifferentiated.Comparingtheharmoniouslypassionategrouptotheobsessivelypassionategroupwhileignoringthemixedgroupwouldthusresultinanimportantlossofpower.Furthermore,ifthisprocedurewereused,nonpassionateindividualswouldnolongerbenonpassionateinabsolutetermsbutrelativetootherpartic-ipantsinthesample.Consequently,thenumberofpassionateversusnonpassion-atepeoplewouldnolongerbecomparableacrosssamples.Forthesereasonswechosetodistinguishpassionateindividualsfromnonpassionateonesusingancriteria(theresponsescalemidpoint)andtostudythebetweenharmoniousandobsessiveindividualsusingstandardizedscoresonthetwopassionsubscales.Mageau,Vallerand,Charest,etal. Table2ResultsofDiscriminantFunctionAnalysisofPassionGroupsfortheThreeStudiesPredictorsStudy1Study2Study3CorrelationsofPredictorsWithDiscriminantFunctionsUnivariate(1,223)CorrelationsofPredictorsWithDiscriminantFunctionsUnivariate(1,121)CorrelationsofPredictorsWithDiscriminantFunctionsUnivariate(2,174)Function1Function2HarmoniousPassion()vs.ObsessivePassion(HarmoniousPassion)vs.ObsessivePassion(Passionate()vs.Nonpassionate(HarmoniousPassion()vs.Passion(Senseofidentityderivedfromtheactivity.513.55,.06„„.70.0518.53,Childspreferenceforactivityspecialization.413.27,.07.50.219.49,Parentspreferenceforactivityspecialization.729.98,.01.47.018.22,Perceivedvaluationoftheactivitybyparents„„„„.82.3926.34,Autonomysupportfromsigni“cantothers.8710.22,.01.545.66,.05.36.888.91,Canonical.24.37.55.24Eigenvalue.06.16.43.06.Correlationslowerthan.3werenotinterpreted. peoplewithanobsessivepassionseemtoderiveahighersenseofidentityfromtheiractivitythanpeoplewithaharmoniouspassion.Itmightbethatpeoplewithanobsessivepassionnolongerseetheiractivityengagementasachoicebutratherasawaytomaintaintheiridentity.This“ndingisconsistentwithpreviousresultswhereithasbeenshownthatalthoughbothtypesofpassionarerelatedtoonesidentity,thecorrelationisstrongerforobsessivepassion(Vallerandetal.,2003,Study1).Finally,thepresentresultsrevealthat100%ofStudy1ssampleofexpertperformerswaspassionate.Thissuggeststhatpassionmightplayanimportantroleinsustainingpeopleseffortsinreachingexpertlevels,therebyreinforcingtheimportanceofstudyingthedevelopmentofpassion.STUDY2TheresultsofStudy1showedthatautonomysupportwashigherfortheharmoniousthantheobsessivepassiongroup.However,itshouldbenotedthatautonomysupportwasassessedthroughaperceivedautonomysupportscalecompletedbytheparticipantsthemselvesandnotdirectlybytheadultsintheirsocialenvironment.Becauseofpo-tentialresponsebiasesinparticipantsresponses,the“rstpurposeofStudy2wastoreplicateStudy1s“ndingsusingaparentalautonomysupportscalecompletedbytheparentsthemselves.ThesecondpurposeofStudy2wastotesttheroleofautonomysupportinharmoniousandobsessivepassionwithparticipantswhocorrespondedtoBlooms(1985b)intermediatephaseofactivityen-gagement.Suchperformerstypicallyhavebeeninvolvedintheirac-tivityforseveralmonths(andevenafewyears),havereceivedformalinstructionontheiractivity,andhavestartedtoengageindeliberatepractice.InlinewithStudy1sresults,itwashypothesizedthattheseperformerswouldbehighlypassionate,thusallowingustofurtherstudythedistinctionbetweenharmoniousandobsessivepassion.Itwaspredictedthatchildrenwhoseparentsengagedinautonomy-supportivebehaviorswithrespecttothepassionateactivitywouldbemorelikelytoexperienceaharmoniousthananobsessivepassiontowardtheiractivity.The“nalpurposeofStudy2wastoexaminepreferenceforac-tivityspecialization,whichshouldin”uencepassionthroughitsim-pactonactivityvaluationandtimespentontheactivity.ItwasMageau,Vallerand,Charest,etal. expectedthatthemorechildrenvaluedactivityspecialization,themoretheyshoulddevelopapassiontowardtheiractivity.However,itwasexpectedthatonespreferenceforactivityspecializationmayalsodifferentiatebetweenharmoniousandobsessivepassion.In-deed,wehypothesizedthatchildrenwhotendtoneglectotherareasoftheirlivesinordertospecializeintheirpassionateactivitymightbemoreatriskofdevelopinganobsessivetypeofpassionbecausetheymightbecomeoverlydependentontheiractivityforself-de“-nition.Adultsshouldalsoplayanimportantroleinguidingchildrentowardactivitiesthattheythemselvesvalue.Itwasexpectedthatparentswhoencouragespecializationinagivenactivitywoulden-couragechildrentobecomepassionatefortheactivity.However,aschildrenprogressinpro“ciencyintheactivity,theirparentsen-couragementtofocusontheactivitymaybeperceivedasexternalpressurestoengageinthepassionateactivity,whichshouldleadtoamoreobsessivepassionbylinkingsocialapprovaltothechildsac-tivityengagement.MeasuresoftheactivityasbeingpartofonesidentitywerenotincludedinStudy2becauseseveralparticipantswouldhavebeentooyoungtounderstandthemeasure.ParticipantsandProcedureThesamplewascomposedof163childrenandteenagers(88girlsand75boys)withameanageof11years(range6to16yearsold),whohadbeeninvolvedinaparticularactivity(e.g.,learningsciences,dancing,ex-ploringnature)forsometimeandwhohadregisteredinasummercampspecializedintheiractivity.Whennotatcamp,childrengenerallyen-gagedintheiractivity1.9timesaweekandhadonaverageof3yearsand10monthsofexperiencewiththeiractivity.Thechildrenandtheirparentswerecontactedthroughthesummercampdirectors.Parentswhoagreedtohavetheirchildengageinthestudyreceivedaquestionnaireandaconsentformbymail.Childrenweremetattheirsummercampduringafree-timeperiod,wheretheywereaskedtocompleteaquestionnaire.Finally,atotalof111mothersand38fathersparticipatedinthestudy(meanage42years;range32to58yearsold).Thechildrensquestionnaire.Thechildrensquestionnairecontainedmeasuresofpassion,preferencesforactivityspecialization,previousex-periencewiththeactivity,frequencyofactivityengagement,age,andgender.ThePassionScale(Vallerandetal.,2003)describedinStudy1OntheDevelopmentofPassion wasadaptedforyoungerchildrenforthepurposeofthepresentstudy.Subscalesandtheanswerscalewereshortenedanditemsweresimpli“edtofacilitatecomprehension.Childrenwereaskedtothinkaboutandnametheactivitytheyweredoingatcamp(i.e.,learningsciences,danc-ing,orexploringnature).Theywerethenaskedtoindicatetheextenttowhicheachitemwastrueforthemregardingthisactivityonananswerscalerangingfrom1(nottrueatallforme)to4(verytrueforme).Aswasthecasefortheadultversionofthescale,thepresenceofpassionwasmeasuredbyaskingchildrentoratetheextenttowhichtheydevotedtimetotheactivity,founditimportant,andlovedit.Thesethreeitemscor-relatedwitheachother(.73;seeTable3).Harmoniousandobsessivepassionsweremeasuredwithfouritemseach.Asampleitemforharmo-niouspassionisThisactivitygoeswellwiththeotheractivitiesinmylife(e.g.,school,friends,family,etc.)(.60),andasampleitemforob-sessivepassionisSomethinginsideofmemakesmedothisactivityChildrenspreferencetospecializewasassessedusingathree-itemscalethataskedchildrentoreporttheextenttowhichtheybelieveditwasimportanttofullyfocusontheirfavoriteactivityevenifthatmeantputtingsomeotheractivitiesaside.ChildrenindicatedhowmucheachitemwastrueforthemusingthesameresponsescaleastheoneusedwiththePassionScale.AsampleitemisInmyfreetime,IdrathergiveupsomeactivitiesinordertohavemoretimetodomyfavoriteactivityTheparentsquestionnaire.Parents(i.e.,oneparentperchild)completedaquestionnairedealingwiththeirownbehaviorsandpreferences.Foreachscale,parentswereaskedtoratetheextenttowhichtheyagreedwitheachitemusinga7-pointLikert-typescale,whichrangedfrom1(donotagreeatall)to7(verystronglyagree).Demographicvariablessuchasageandgenderwerealsomeasured.Autonomy-supportiveparentalbehaviorswereassessedusingasix-itemscale.Autonomy-supportiveindividuals(1)provideasmuchchoiceaspossiblewithinspeci“climits;(2)providearationalefortasks,limits,andrules;and(3)inquireaboutandacknowl-edgeothersfeelings(Koestneretal.,1984).Inthepresentstudy,eachofthesebehaviorswasassessedwithtwoitems.SampleitemsareIoffermychildmanyopportunitiestomakehis(her)ownchoicesregardingthisactivity(providingchoice),Isometimesexplaintomychildtheuse-fulnessofthisactivity(providingarationale),andIsometimesopenlyconsidermychildsthoughtsandfeelingseventhoughtheydifferfrommine(acknowledgingtheotherpersonsfeelingsandperspective;.71).Parentalpreferenceforactivityspecializationwasmeasuredus-ingasix-itemscale.Speci“cally,thismeasureassessedtheextenttowhichparentswishedthattheirchildwouldchooseandfocusontheirpassion-Mageau,Vallerand,Charest,etal. Table3Study2:MeansandStandardDeviationsfortheTwoPassionGroupsandCorrelationsBetweenAllVariablesMeans(Total(1.2.3.4.5.6.7.8.9.1.Harmoniouspassion3.56(0.33)3.20(0.44).60„.30.04.132.Obsessivepassion1.42(0.40)2.13(0.58).66„.29.14.07.183.Lovefortheactivity3.77(0.43)3.64(0.51)„„.49.08.114.Activityvaluation3.16(0.80)3.25(0.77)„„.49.11.08.015.Timeinvestment2.99(0.87)2.96(0.76)„„.14.01.12.116.Frequencyofparticipation(timesper1.7(1.9)2.3(2.1)„„.03.16.057.Childspreferenceforactivityspecialization1.49(0.63)1.75(0.69).69„.128.Parentspreferenceforactivityspecialization2.46(0.83)2.91(0.82).71„.029.Parentalautonomy5.15(1.07)4.73(0.86).71„ ateactivityandputasideotherlessimportantactivities.AsampleitemisIbelieveitispreferablethatmychildsacri“cessomethingsinordertoexcelinoneactivity(.71).ResultsandDiscussionUsingthesametwo-stepprocedureasinStudy1,childrenwereclassi“edinoneofthreedifferentgroups:nonpassionate,harmoni-ouslypassionate,andobsessivelypassionate.Itwasexpectedthatmostchildrenwouldbepassionatetowardtheiractivitybecausetheyhadbeenselectedfortheirsustainedengagementintheiractivity.Resultsshowedthat145childrenwereconsideredpassionate(i.e.,73harmoniouslyand72obsessivelypassionate),andonlyafewwerenot(13;5participantshadmissingvaluesonthepassionSubsequentanalysesfocusedonthedistinctionbetweenharmoniouslyandobsessivelypassionatechildrenbecausenonpas-sionatepeoplewerelargelyunderrepresented(92%ofpassionatevs.8%ofnonpassionateparticipants).Thedataofthenonpassion-atechildrenwerethusdiscardedfromtheanalyses.Nodifferenceswerefoundbetweentheharmoniousandobsessivepassiongroupsinage,gender,andweeklyengagementinthepassionateactivity.Meansandstandarddeviationsforeachgroup,aswellascorrela-tionsbetweenallvariables,arepresentedinTable3.SimilarlytoStudy1,adiscriminantfunctionanalysiswasper-formedtodistinguishbetweenharmoniouslyandobsessivelypassion-atechildrenusingparentsreportofautonomy-supportivebehaviorsandparentsandchildrenspreferencesforactivityspecializationaspredictors.Resultsrevealedonesigni“cantdiscriminantfunction,Ca-nonicalCorrelation.37,Wilkss17.83,.001,whichshowedthatthepredictorscouldsigni“cantlyseparatethetwopassiongroups.ResultsfromthediscriminantfunctionanalysisarepresentedinTable2.Correlationsbetweenthepredictorsandthediscriminantfunctionshowedthatchildrenintheharmoniouspassiongrouphadparentswhoweremoresupportiveoftheirautonomy(.54)thanthoseintheobsessivepassiongroup.Conversely,childreninthe5.Dividingparticipantsintofourgroupsusingthemedianontheharmoniousandobsessivesubscaleswouldyieldthefollowinggroupsizes:relativelynonpas-sionategroup31,harmoniouslypassionategroup11,obsessivelypassionate31,andmixedgroup29(61participantsweresituatedatthemedianandcouldnotbeclassi“ed).Mageau,Vallerand,Charest,etal. obsessivepassiongroupandtheirparentsbothvaluedspecializinginoneparticularactivitymorethanchildrenintheharmoniouspassiongroupandtheirparents(childspreferenceforactivityspecializa-.41;parentsThepresentresultsnicelyreplicatethe“ndingsofStudy1,thistimeusingparentsownreportoftheirautonomy-supportivestyle.Itthusappearsthatchildrenwhoengageinanenjoyableactivityunderparentalautonomy-supportiveconditionsmaybemorelikelytoex-perienceaharmoniousrelativetoanobsessivepassiontowardtheactivity.Inaddition,althoughcorrelationalinnature,thepresent“ndingsextendthoseofStudy1inshowingthattoengageinactivityspecializationduringthesecondphaseoftalentdevelopmentmightalsopromoteobsessivepassion.Ofparticularinterestisthe“ndingthatbothparentsandchildrensmeasuresyieldedthesameresults.Itmayverywellbethatparentsbeliefsabouttheimportanceofspe-cializinginoneactivityareinternalizedbythechildren,whichthenfacilitatesamoreobsessivepassion.STUDY3Takentogether,resultsofStudies1and2suggestthatpassionmaybeimplicatedinsustainingactivityengagementforseveralyears.Indeed,thevastmajorityofindividualsinvolvedinaspeci“cactivityforasigni“cantperiodoftimedisplayedatleastamoderatelevelofpassion.Furthermore,identi“cationwiththeactivity(Study1),preferenceforactivityspecialization(Study2),andautonomysup-port(Studies1and2)distinguishedbetweenharmoniousandob-sessivepassion.ThepurposeofStudy3wastoreplicatetheseresultsbyassessingallthreevariablesinashort-termlongitudinalstudy.Inaddition,wewishedtostudytheconditionsthatfacilitatethedevel-opmentofpassion(vs.nopassion)attheinitialphaseofactivityengagement(Bloom,1985b).Toreachthisgoal,weaskedasampleofjuniorhighschoolstudentswhohadnopriorexperiencewithaspeci“cactivity(i.e.,playingamusicalinstrument)toparticipateinastudyinvolvingthreetimesofmeasurement.Byfollowingpartici-pantswhowereregisteredintheir“rstmusicclassoverthecourseoftheir“rstsemester,wesoughttopredictwhowoulddevelopapas-sionformusicattheendofthetermand,amongthosewhodid,predictthosewhowoulddevelopaharmoniousoranobsessivepas-OntheDevelopmentofPassion sion.Itwashypothesizedthatatthenovicestage,identi“cationwiththeactivity,parentsandchildrenspreferenceforactivityspecial-ization,andautonomysupportfromparentsandmusicteacherswouldbeconducivetothedevelopmentofapassiontowardmusicattheendoftheterm,5monthslater.Wealsoassessedtheextenttowhichparentsvaluetheirchildrensactivityasanadditionalpredic-torofpassion.AccordingtoBloom(1985b),parentschanneltheirchildrensinvolvementtowardactivitiesthattheythemselvesvalue,andwewishedtotestthisadditionalhypothesis.Furthermore,inlinewithStudies1and2,itwashypothesizedthathighautonomysupportfromcloseadults(parentsandmusicteach-ers)wouldbeconducivetothedevelopmentofamoreharmoniouspassion.Wealsopredictedthatvariablesthatfocusthechildononeparticularactivity(i.e.,activityspecializationandparentsactivityvaluation)shouldresultinmoreobsessivepassionbecausethesevariablescanbecomeadditionalpressurestoengageintheactivity.However,contrarytoStudy1,wewerenotsurewhetheridenti“ca-tionwiththeactivitywouldpredicttheoccurrenceofobsessivepas-sionbecauseitmightbetooearlytowitnessoveridenti“cationtowardtheactivityatthenovicelevel.MethodParticipantsandProcedureThesamplewascomposedof196juniorhighschoolstudents(96girls,96boys,and4participantswhodidnotspecifytheirgender;12years;11to15yearsold)whohadneverplayedamusicalinstrumentbefore.MusicteachersfromdifferenthighschoolsintheMontrealareawerecontactedandaskedpermissiontomeettheir“rst-yearstudents.Studentswereaskedtoparticipateinathree-waveshort-termlongitudi-nalstudy,wheretheirattitudestowardplayingamusicalinstrumentwouldbeexaminedovera5-monthperiod(lateAugusttolateJanuary).Thestudentswhoagreedtoparticipateobtainedasignedconsentformfromtheirparentsandcompleteda20-minquestionnairesinclass,oneatthebeginning,oneatmidpoint,anda“naloneattheendoftheMeasuresThe“rstquestionnairecontainedmeasuresofpersonalvariables(identi-“cationandpreferenceforactivityspecialization),whichwereassessedMageau,Vallerand,Charest,etal. duringthe“rstorsecondmusicclassofthesemester.Thesecondques-tionnairecomprisedmeasuresofinterpersonalvariables(perceivedau-tonomysupport,perceivedparentalvaluationoftheactivity,andperceivedparentalpreferenceforspecialization,whichweremeasured2monthslaterinthesemester).Thissecondquestionnairewasassessedla-terinordertoallowinteractionstooccurbetweenchildrenandtheirso-cialenvironment(i.e.,themusicteacherandparents)withrespecttothemusicalactivity.Thelastquestionnairewascompletedattheendofthesemesterandassessedstudentspassiontowardplayingamusicalinstru-ment.Foreachscale,studentswereaskedtoratetheextenttowhichtheyagreedwitheachitemusinga7-pointLikert-typescale,whichrangedfrom1(donotagreeatall)to7(verystronglyagreeTime1measures.Inthe“rstquestionnaire,studentswereaskedaboutthesenseofidentityderivedfromtheactivityandtheirpreferenceforactivityspecialization.Measuresofageandgenderwerealsoincludedalongwithoneitempertainingtothestudentspreviousexperiencewithamusicalinstrument.Thesenseofidentityderivedfromtheactivitymeasuredusingafour-itemscalethatassessedwhetherplayingamusicalinstrumentcouldbecomeanintegralpartofthestudentsidentity(Schlenker,1985).AsampleitemisIfIhadtodescribemyself,IwouldmentionthatIplayamusicalinstrument(.90).Thefour-itemscaleusedinStudy2wasusedtoassesstheteenagerspreferenceforactivityspecialization.73).Thisscalemeasurestheextenttowhichteenagersbelieveditwasimportanttoputsomeactivitiesasidetofocusononeparticularactivity.Time2measures.Twomonthslater,participantswereaskedtoevaluatetheirparentspreferenceforactivityspecialization,theimportanceofplayingamusicalinstrumentintheirparentseyes,andthelevelofau-tonomysupportprovidedbyadultsintheirsocialenvironment.Partic-ipantswerealsoaskedaboutthenumberofhourstheypracticedtheirinstrumenteachweek.Perceivedparentalpreferenceforspecializinginwasassessedusingasix-itemmeasurethatevaluatedtheextenttowhichparticipantsperceivedthattheirparentsvaluedtheirspecializationinmusicasopposedtotheirinvolvementinmanyactivities.Wewereinterestedinknowingwhetherteenagersbelievedthattheirparentspre-ferredtoseethemfocusonandimproveinmusicoriftheywouldratherseethemhavemorediversi“edleisure,wheretheywouldlearnfrommanydifferentactivities.AsampleitemisMyparentsencouragemetodevoteallmyfreetimetomymusicalinstrument(.61).Perceivedvaluationformusicbyonesparentswasassessedusingafour-itemscalethatmeasuredteenagersperceptionoftheimportanceoftheactivityintheirOntheDevelopmentofPassion parentseyes.AsampleitemisToplayamusicalinstrumentisveryimportantformyparents(Perceivedautonomysupportwasas-sessedusinganadaptedversionofPelletierandVallerands(1996)six-itemscale.Itemswereslightlymodi“edtotargetsigni“cantothers(i.e.,parents,musicteacher)autonomy-supportivebehaviorsregardingtheactivity.Thismeasureassessedtheextenttowhichparticipantsfeltfreetoorganizeandmanagetheirmusicalpracticesastheywishedasopposedtofeelingcoercedtopracticetheirmusicalinstrumentatspeci“ctimes.Threeitemstargetedbehaviorsthatconveychoice,andthreeotheritemsmeasuredcontrollingbehaviors.Theselastthreeitemswererecodedtoobtainatotalscoreofperceivedautonomysupport.AsampleitemisIngeneral,adultsinmysocialsurroundingsgivemethefreedomtopracticemymusicalinstrumentbymyself(.62).Time3measures.Attheendofthesemester(i.e.,5monthsafterthebeginningofthestudy),participantswereaskedtoreporttheirpassiontowardmusic.Participantsalsoreportedtheiramountofweeklypracticeoftheirinstrument.ThePassionScale(Vallerandetal.,2003)describedinStudy1wasused.Participantswereaskedtothinkabouttheirmusicalinstrumentandtocompletethescaleaccordingtoit.Bothpassionscalesshowedsatisfactoryreliability(harmoniouspassion,.92;obsessivepassion,.91).Thethreepassioncriteria(activityvaluation,timein-vestment,andlovefortheactivity)werealsocompleted(ResultsandDiscussionDescriptiveStatisticsUsingthesametwo-stepprocedureasinStudy1,peoplewereclas-si“edinoneofthreegroups:nonpassionate,harmoniouslypassion-ate,andobsessivelypassionate.Only36%ofparticipantsdevelopedapassionfortheirmusicalinstrumentoverthe5-monthperiod.Speci“cally,resultsshowedthat70peoplewereconsideredpassion-ateattheendofthesemester(i.e.,40harmoniouslyand30obses-sivelypassionatepeople),andamajoritywasnot(threegroupsdidnotdifferinageandtherewereanequalnumberofmenandwomenineachgroup.Meansandstandarddeviationsfor6.Dividingparticipantsintofourgroupsusingthemedianontheharmoniousandobsessivesubscaleswouldyieldthefollowinggroupsizes:relativelynonpas-sionategroup72,harmoniouslypassionategroup28,obsessivelypassionate23,andmixedgroupMageau,Vallerand,Charest,etal. eachgrouparepresentedinTable4,andthecorrelationsbetweenallvariablesarepresentedinTable5.Inthemiddleofthesemester(Time2),studentsreportedprac-ticingonaverageof90mineachweek.Althoughmanydidnotpracticeatall(37.8%),abouthalfofthestudents(46.5%)practicedfrom1to2hrperweek.Towardtheendoftheterm(Time3),themeanamountofweeklypracticehaddecreasedtoameanof62min.Thisdropinpracticelevelwasstatisticallysigni“cant,.01,butwasnotrelatedtoanddidnotinteractwithpassion.These“ndingsrevealthatharmoniously,obsessively,andnonpassionatestudentsexperiencedasimilarandsigni“cant Table4Study3:MeansandStandardDeviationsforEachGroupMeans(1.Harmoniouspassion2.43(1.06)4.87(1.02)4.99(1.29).922.Obsessivepassion1.57(0.65)2.26(0.87)4.29(1.28).913.Lovefortheactivity3.16(1.73)6.33(0.86)6.23(0.86)„4.Activityvaluation1.79(0.94)4.13(1.54)4.90(1.86)„5.Timeinvestment1.81(0.98)4.25(1.41)5.43(1.46)„6.Practice(hours/week)atTime20.91(1.53)2.51(4.03)2.48(2.75)„7.Practice(hours/week)atTime30.53(0.79)1.62(2.07)2.35(2.81)„8.Identi“cationwiththe2.05(1.31)3.15(1.43)3.77(1.59).909.Teenagerspreferenceforspecialization2.30(1.26)2.90(1.29)3.53(1.10).7310.Perceivedparentalpreferencefor2.61(1.13)3.16(1.06)3.70(1.22).6111.Perceivedvaluationofactivitybyparents1.75(0.94)2.71(1.62)3.81(1.68).8912.Perceivedautonomy5.35(1.07)6.10(0.90)5.52(0.95).62OntheDevelopmentofPassion Table5Study3:CorrelationsBetweenAllVariablesTotal(1.2.3.4.5.6.7.8.9.10.11.12.1.Harmoniouspassion„.722.Obsessivepassion„.503.Lovefortheactivity„.604.Activityvaluation„.715.Timeinvestment„.386.Practice(hours/week)atTime2„.777.Practice(hours/week)atTime3„.328.Identi“cationwiththeactivity„.569.Teenagerspreferencefor„.1810.Perceivedparentalpreferencefor„.4211.Perceivedvaluationofactivityby„.0612.Perceivedautonomysupport dropinpracticefromTime2toTime3.Thisdropinpracticetimemayhavebeenduetoadecreaseinthenoveltyoftheactivity.Pas-sionatepeoplepracticedtheirmusicalinstrumentmorethannon-passionatepeoplebothatTime2,(2,166).001,andatTime3,(2,178).001.Attheendofthesemester,non-passionatepeoplepracticedtheirinstrument30minperweekonaverage,whereaspassionatepeoplepracticedapproximately2hrperweek(seeTable4forexactnumbers).Theseresultsarenotsurpris-ing,giventhatpassionateindividualsareidenti“edinpartbyaskingpeoplehowmuchtimetheyspendontheiractivity.Ofinterestisthattherewasnosigni“cantdifferenceinpracticetimebetweenthehar-moniousandobsessivepassiongroups,whichseemstosuggestthatharmoniousandobsessivepassionleadtosimilaractivityengage-ment.This“ndinghasbeenconsistentlyfoundinthepresentthreestudiesandisconcordantwiththeassertionofVallerandandhiscolleagues(2003)totheeffectthatbothtypesofpassionarerelatedtotimeandenergyinvestmentinthepassionateactivity.OntheDevelopmentofaPassionAdiscriminantfunctionanalysiswasperformedtotestourhypothesesonthedevelopmentofpassion.Resultsrevealedtwosigni“cantdisc-riminantfunctions,CanonicalCorrelation.55,Wilkss72.17,.001;CanonicalCorrelation.24,Wilkss10.34,.05.Examinationofthediscriminantfunctionsateachgroupscentroidsshowedthatthe“rstdiscriminantfunctiondistin-guishedbetweennonpassionate(.47)andpassionategroups(obses-sivelypassionategroup1.18;harmoniouslypassionategroup.67).Theseconddiscriminantfunctionseparatedtheharmoniouslypassion-ategroup(.40)fromtheobsessivelypassionategroup(.44),withthenonpassionategroupsituatedinthemiddle(.04).ResultsfromthediscriminantfunctionanalysisarepresentedinTable2.Correlationsbetweenthepredictorsandthe“rstdiscriminantfunctionshowedthatearlyoninthetermpeoplewhoeventuallybecamepassionatetowardtheirmusicalinstrumentattheendofthetermdifferedfromnonpassionatepeopleinthattheyweremorelikelytoseethepossibilityofderivingasenseofidentityfromtheactivity(.70),theyvaluedactivityspecializationtoagreaterextent(.50),theirparentssharedthispreferenceforspecialization(.47),adultsintheirsocialsurroundingsweremoresupportiveoftheirOntheDevelopmentofPassion autonomyregardingtheiractivity(.36),and“nallytheirparentsvaluedmusictoalargerextent(.82).Thus,childrenwhoidenti“edwiththeactivity,preferredtospecializeinanactivity,andthosewhoseparentssupportedtheirautonomywhilevaluingtheactivityandactivityspecializationwerefoundtohavedevelopedapassionformusic5monthslater.DistinguishingHarmoniousFromObsessivePassionTheseconddiscriminantfunctionseparatedpeoplewhodevelopedanobsessivepassionfromthosewhodevelopedamoreharmonioustypeofpassion.Correlationsbetweenthepredictorsandthisseconddiscriminantfunctionshowedthatthequalityofsigni“cantothersinvolvementin”uencedthetypeofpassionthatwasexperienced.Childrenwhodevelopedaharmoniouspassionbytheendofthetermhadinteractedwithadultswhoweremoreautonomy-support-ive(.88)andwhovaluedtheactivityless(.39)thanchildrenwhodevelopedanobsessivepassion.Theseresultssuggestthatexcessivesocialpressurestoengageinanactivitymayfacilitateobsessivepas-sion.Indeed,controllingparentsandparentswhohighlyvaluetheactivityseemtofacilitateamoreobsessivetypeofpassion.Contrarytoexpectations,preferencesforactivityspecializationatthisstageofpassiondevelopmentdidnotdifferentiatebetweenthetwotypesofpassion.Itisprobablethatatearlystagesofactivityengagement,valuingactivityspecializationhelpsthechildfocusontheactivityanddoesnotimplythatheorsheneglectsotherlifeaspects,whichinturnfacilitatespassioningeneralbutnotobsessivepassioninpar-ticular.Finally,identi“cationwiththeactivitywasnotfoundtopredictthetypeofpassion(harmoniousvs.obsessive)thatdevelopedovertheterm.Thisresultsuggeststhatoveridenti“cationwiththeactivitymightonlyoccuratlaterstagesofpassiondevelopment.TheDevelopmentofPassion:ACross-SectionalLookTofurtherexploredevelopmentaldifferences,wecomparedournovicebutpassionatemusicians(70)ofStudy3withourexpertmusicians(85)ofStudy1.Thesetwogroupswerechosenbe-causebothcompletedthesameversionofthePassionScale,theywerepassionatetowardtheiractivity,andtheywereinvolvedwiththesameactivity,namely,playingamusicalinstrument.SeparateMageau,Vallerand,Charest,etal. ANOVAsshowedthat,althoughbothgroupswerepassionate,ex-pertmusiciansweremorepassionateingeneral(intermsofthepas-sioncriteria)towardtheiractivity(6.25)thanbeginners(1,153).001.Speci“cally,expertmusiciansvaluedtheiractivitytoagreaterextent(Experts,6.44;Begin-4.46)andreportedspendingmoretimeonit(Experts,6.04;Beginners,4.76)thannovicemusicians.However,nodifferencewasfoundbetweenthetwogroupsonhowmuchtheylovedtheactivity,.92.Expertmusiciansalsopracticedmuchmore(21.02hr/week)thanbeginners(1.91hr/week),(1,148).001.Theywereobviouslyolder(Experts,22.2;Beginners,(1,275).001,butdidnotdifferingender,.19.TheseresultsarepresentedinTable6.Theselastresultsunderscorethefactthataspassiondevelopsthroughyearsofinvolvementintheactivity,lovefortheactivitymaynotbethemoredynamicfactoratplay.Ratheractivityvaluationandtimeandenergyexpendituretowardtheactivityseemtobemoreimportanttomaintainpassionovertheyears.Finally,differencesbetweenthetwotypesofpassionwereinves-tigatedusinga22mixeddesignANOVA,wherethetwotypesofpassionwereenteredaswithin-subjectmeasuresandpeopleslevelofexperienceasbeginnersorexpertswasincludedasabetween-subject Table6MeanDifferencesBetweenBeginning(Study3)andExpertMusicians(Study1)Study3,Study1,5.335.69(1,153)Obsessivepassion3.524.82(1,153)Levelofpassion5.176.25(1,153)Activityvaluation4.466.44(1,153)Lovefortheactivity6.296.27(1,153)Timeinvestment4.766.04(1,153)Practice(hoursper1.9121.02(1,148)OntheDevelopmentofPassion factor.Resultsrevealedaninteractionbetweenthetypeofpassionandpeoplesphases(novicevs.expert),(1,153)Speci“cally,simpleeffectsshowedthatwhileexpertsweremorehar-monious((1,279).001,andmoreobsessive(1,279).001,thantheirbeginningcoun-terparts(HP,5.33;OP,3.52),thedifferencebetweenex-pertsandbeginnerswasmorepronouncedforobsessivethanharmoniouspassion.Theseresultssuggestthataspeoplebecomemoreheavilyinvolvedintheiractivity,bothtypesofpassionincrease,andthisincreaseisparticularlyimportantforobsessiveGENERALDISCUSSIONThepurposeofthepresentresearchwastostudytheprocesseslead-ingtothedevelopmentofpassioningeneralandharmoniousandobsessivepassionsinparticular.Inthreestudiesinvolvingnovice,intermediate,andexpertparticipants,wepredictedwhobecamepassionateornot(Study1)andwhethersuchapassionbecameharmoniousorobsessive(allthreestudies).AshypothesizedbytheDualisticModelofPassion(Vallerandetal.,2003),thepresent“nd-ingsunderscoretherelevanceofidenti“cationwiththeactivity,au-tonomysupport,preferencesforactivityspecialization,andparentsactivityvaluationforthedevelopmentofpassion.Additionalresultsalsoshowthatinteractingwithautonomy-supportiveadultsmayfa-cilitatethedevelopmentofharmoniouspassioninparticular.Incontrast,identi“cationwiththeactivity,preferencesforactivityspe-cializationandparentsactivityvaluationmayattimesbecomead-ditionalpressurestoengageintheactivityandfacilitateamoreobsessivepassion.Anumberoftheoreticalimplicationscanbedrawnfromtheseresults.OntheDevelopmentofPassionFirst,thepresentresearchconstitutesanimportantsteptowardunfoldingtheprocessesunderlyingthedevelopmentofpassion.TheDualisticModelofPassionpositsthatvariablesthatincreaseactivityvaluationandtimespentontheactivity,identi“cationwiththeactivity,andautonomy-supportivesocialenvironmentsarekeyvariablesforthedevelopmentofpassion.ResultsofMageau,Vallerand,Charest,etal. Study3supportedthesehypothesesusingashort-termlongitudinalPassionsarehypothesizedtobestrongindinationstowardactiv-itiesthatpeoplelikebutthattheyalsogreatlyvalueandinvesttimeandenergyin.Pastresearch(Vallerandetal.,2006;Vallerand&Houlfort,2003)hasshownthatthetwopassionsubscalesarepos-itivelyandequallyrelatedtoactivityvaluationandtimespentontheactivity.Thiswasalsothecaseinthepresentthreestudies.Variablesthatin”uenceactivityvaluationandtimespentontheactivityshouldthusbeimportantinthattheyshoulddeterminewhichactivitywillbecomepassionateandwhichwillnot.Inthepresentresearch,par-entsandparticipantspreferencesforactivityspecializationandparentsvaluationoftheactivitywerepresumedtoin”uenceactivityvaluationandtimespentontheactivity,andassuchtheywerehypothesizedtoin”uencepassion.Resultscon“rmthatthesevari-ablesdifferentiatebetweenpassionateandnonpassionatepeople,thussuggestingtheirimportanceforthedevelopmentofbothtypesofpassion.Anothervariablethatmayholdakeyroleinthedevelopmentofapassionistheidenti“cationwiththeactivity.EcclesandBarber(1999)havesuggestedthatintheteenageyearsactivitychoiceislikelytobothgrowoutofandreinforceanemergingsenseofidentity.Waterman(1990)alsoproposedthatpersonallyexpressiveactivities,aconceptsimilartothatofharmoniouspassions,areac-tivitiesthatresonatewiththepersonstrueself.Thepresent“ndingsareconsistentwiththesepropositions.ResultsfromStudy3showedthatinthevery“rststepsofactivityinvolvement,derivingasenseofidentityfromtheactivity(orseeingthatonecoulddosoeventually)leadstothedevelopmentofpassionmonthslater.These“ndingssuggestthatpeoplearelikelytodevelopapassionforanactivitythatextendsanaspectoftheiridentityorpertainstohowtheycanseethemselvesinthefuture.Thequalityofthesocialenvironmentisone“nalvariablethatdistinguishesbetweenpassionateandnonpassionatepeople.Aspastresearchhasshown(Deci&Ryan,1987;Grolnick&Ryan,1989;Koestneretal.,1984;Mageau&Vallerand,2003),adultswhosup-portchildrensautonomycreateoptimalconditionsofactivityen-gagementwherechildrenarefreetoexploretheactivity,displaycreativity,andexperiencepositiveaffectiveoutcomes.Althoughcor-relationalinnature,resultsfromStudy3suggestthatpassionisyetOntheDevelopmentofPassion anotherpositiveconsequenceofanautonomy-supportiveenvironment.Weproposethatincontrollingsettingspassionislesslikelytomateri-alizebecausealargeportionofchildrensattentionisdirectedtowardmeetingothersexpectationsandnotontheactivityitself.InlinewithCsikszentmihalyis(1975,1982)theoryon”owexperiences,childrenwhoaredistractedfromtheiractivitywouldbelesslikelytoexperience”owandotherpositiveconsequencesduringactivityengagement.Ifwetakeintoconsiderationthefactthatadults,andespeciallyparents,typ-icallyinitiatechildrentotheactivity,conveythatitistobevalued,andsuggestthatitmayrepresentadesirableaspectofonesidentity,onecanstartappreciatingtheimportantrolethatparentsplayinthedevelop-mentofpassioninchildren(seeEccles&Wig“eld,2002).Asecondtheoreticalimplicationpertainingtothedevelopmentofpassionisthatpassionseemstofollowadynamicmovementovertime,asitappearstobemoreprevalentaspeoplespendanincreas-ingamountoftimeontheactivity.Whendescriptiveanalysesofallthreestudieswerecompared,importantdifferenceswerefoundinthepercentageofpassionatepeopleacrossthethreesamples,whichcorrespondedtothethreelevelsofBloomstalentdevelopmentphases.Whereasonly36%ofbeginners(Study3)wereidenti“edaspassionate,92%ofchildrenofStudy2(whohadameanaverageof3yearsofexperience)and100%ofexperts(withapproximately10yearsofexperience)werefoundtobepassionatetowardtheiractivity.Thisincreaseinthepercentageofpassionatepeopleintheactivityasonemovesfromthenovicetotheexpertstagesmightre-sultfromtwoprocesses.First,thenonpassionatepeoplemaybeweededoutastimegoeson,leavingonlypassionatepeopleengaginginthatspeci“cactivity.Becausemanypassionateactivitiesarelei-sureactivities,ifsomepeopledonotenjoyanactivity,presumably,theywilltryto“ndanotherthatsuitsthembetter.Thereissupportforthis“rstprocessasthenumberofparticipantsinsports,forin-stance,decreasesaschildrengetolder(Roberts,1984).Thesecondprocessmaydependonthebene“tsofactivityengagement.Aspeo-pleinvesttimeandeffortinagivenactivity,theyarelikelytoex-periencepositiveoutcomesthatshouldfacilitatepassion.Furthermore,throughcognitivedissonance(Festinger,1957),onemightbecomeevenmoreinvolvedandpassionateafterhavingin-vestedimportantresourcesandtimeintheactivity.Longitudinalresearchisneededinordertotestthesehypothesesandmoreclearlychartthedevelopmentofpassionoveryearsofactivityinvolvement.Mageau,Vallerand,Charest,etal. A“naltheoreticalimplicationdealingwiththedevelopmentofapassionisthatpassionentailsauniqueperson…activityinterface(Vallerandetal.,2003).Indeed,resultsofStudy3showedthatonlyaminorityofchildreneventuallybecamepassionatetowardplayingamusicalinstrument.Speci“cally,outof196beginnersinStudy3,only70(36%)became(atleastmoderately)passionatetowardmusic4monthslater.Thus,beingpassionateshouldnotbeviewedasapersonalitytraitbutasaspecialrelationshiponedevelopswithaspeci“cactivity.Inlinewithotherswhoinvestigatedthedevelop-mentofstronginterests(Waterman,1990,1993,2004),wesuggestthatpeoplecannotbepassionatetowardallactivitiesbutthatamatchbetweenonesinterestsandabilitiesandthetaskmaybenec-essaryforpassiontodevelop.However,suchamatchwasnotas-sessedinthepresentresearchandshouldbetheobjectoffutureOnthePredictionofHarmoniousandObsessivePassionThepresentresearchalsoshowsthatitispossibletodifferentiatebetweenthetwotypesofpassion.Fourvariablesappeartopredictwhichtypeofpassionwillemerge(i.e.,autonomysupport,prefer-enceforactivityspecialization,parentsvaluationoftheactivity,andidenti“cationwiththeactivity).Autonomysupportconsistentlydis-tinguishedbetweenharmoniouslyandobsessivelypassionatepeople.Peoplewithaharmoniouspassionreportedinteractingwithparentsandothersigni“cantadultswhoweremoreautonomy-supportivethanthosereportedbyobsessivelypassionatepeople.This“ndingwasrobust,asitwasobtainedacrossallthreestudiesusingcorre-lationalandshort-termlongitudinaldesignsandthreedifferentmea-suresofautonomysupport,includingascaleassessedbytheparentsthemselves(Study2).Itwasalsoobservedwithnovice,intermediate,andexpertsamples,suggestingthatautonomysupportmightbeanimportantingredientnotonlyforthedevelopmentbutalsoforthemaintenanceofaharmoniouspassionatallstagesoftalentdevel-opment(Bloom,1985b).InlinewithSelf-DeterminationTheory(Deci&Ryan,1985,2000;Grolnick&Ryan,1987,1989),thesere-sultssupporttheimportanceofanautonomy-supportivestyleinfa-cilitatingoptimalactivityengagement.Futureresearchisneededtobetterunderstandhowcontrollingbehaviorsfromsigni“cantadultsleadtomoreobsessivebehaviorsinchildren.Wearguethatcon-OntheDevelopmentofPassion trollingbehaviorsleadtoobsessivepassionbyteachingchildrenthattoobtainsocialapprovaltheymustengage,orworse,excel,intheiractivity,thuscreatingself-activitycontingencies.Theactivitythenbecomeshighlyimportantbutforself-protectiveanddefensiverea-sonsthatarenotnecessarilycoherentwiththechildstruedesiresandsenseofself.Thisinturnsetsthestageforexperiencinganob-sessivepassion.Similarly,thepresentresultsshowthatparentswhohighlyvaluetheirchildrensactivityseemtofosteramoreobsessivepassion.Althoughparentshelpchildrentofocusontheiractivitybyvaluingtheactivity,itseemsthatitcan,attimes,putadditionalpressureonchildrentoengageintheiractivityandfosterobsessivePreferenceforactivityspecializationalsodistinguishedbetweenthetwotypesofpassion,althoughitsrolechangedacrossBlooms(1985b)phasesofactivityinvolvement.Inthenovicesample(Study3),participantsandparentspreferencesforactivityspecializationpredictedpassioningeneral.However,thesevariablesdistinguishedchildrenwithanobsessivepassionfromthosewithaharmoniouspassionintheintermediatesample(Study2).Itseemsthataschil-drenprogressintheiractivity,strongpreferencesforactivityspe-cialization(theirownortheirparents)maybeexperiencedascoercivebecauseitmightimplythatchildrenmustgiveupotherimportantactivitiesthataredeartothem.Theseresultssuggestthatwhatisappropriatetonurturepassionatagivenstageofdevelop-ment(novice)maynolongerbeappropriateatalaterstage(inter-mediate).FutureresearchisneededtoaddressthisissuemoreSimilarly,theroleofidenti“cationwiththeactivityinthedevel-opmentofpassionalsochangedacrossBlooms(1985b)phasesofactivityinvolvement.WhereastheresultsfromStudy1withexpertperformersrevealedthatidenti“cationwiththeactivityseemscon-ducivetothedevelopmentofanobsessivepassion,thoseofStudy3withnoviceparticipantsfailedtounveilthisrelationship.Identi“-cationwiththeactivitywasinsteadshowntopredictbothtypesofpassion.Overall,theseresultssuggestthatalthoughidenti“cationmayrepresentanin”uentialvariableindeterminingifanactivitywillbecomepassionateornot,itmaybeonlywithtimeandexperiencewiththeactivitythatitin”uencesthedevelopmentofanobsessivepassion.TheseresultsareinlinewithVallerandandhiscolleagues(2003,Study1)“ndings,whichshowedthatforperformerswithMageau,Vallerand,Charest,etal. yearsofexperienceintheactivity,bothtypesofpassionarerelatedtopeoplesidentity,butthecorrelationisstrongerforobsessivepassion.Itwouldthusappearthattodevelopapassionforagivenactivity,oneneedstoseethisactivityasbeingpartofonesidentity.However,ifthisactivitycomestotaketoomuchspaceinonesidentityovertime,anobsessivepassionmaydevelopattheexpenseofharmoniouspassion.Futureresearchisneededinordertoshedlightonthisissue.PassionandOutcomesA“nalimplicationthatcanbederivedfromthepresent“ndingsisthatpassionmatterswithrespecttoatleasttwooutcomes.First,passionateexpertperformerswhowerefoundtobemorepassionatethantheirnovicecounterpartsengagedintheiractivity10timesmorethanthepassionatenovicemusicians.These“ndingsareinlinewithrecentresearch(Vallerandetal.,2007,2008)thatshowsthatbothtypesofpassionrepresentimportantpredictorsofdeliberatepractice(Ericsson&Charness,1994).Asecondoutcomerelatedtopassionisperformance.Resultsfromthethreestudiessuggestthatasonemovesfromthe“rsttothethirdstageofactivityinvolvement,thepercentageofpassionatepeopleincreases(only36%ofthenov-icesamplewaspassionate,whereas100%oftheexpertsamplewas).Furthermore,bothharmoniousandobsessivepassionsincreasewhenwecompareanovicetoanexpertsample(seeTable6).Be-causetheprogressionfromtheinitialtothelaststageentailsamovementtowardexcellenceinperformance,itwouldappearthatpassionmightbeanimportantingredienttoachieveperformance.Thishypothesishasbeensupportedinrecentcorrelationalresearch.Speci“cally,Vallerandandhiscolleagues(2007)foundintwostudiesthatpassionleadstosustainedinvolvementindeliberatepractice(Ericsson&Charness,1994)aimedattaskmastery,whichinturnleadstoobjectivelevelsofperformance.Thecommonbeliefthatobsessive,andnotharmonious,passionisnecessarytoreachexcel-lencehasthusnotreceivedempiricalsupporteitherinthepresentresearchorinpaststudies(Vallerandetal.,2007,2008).Conversely,ourexpert-levelperformerswithanobsessivepassionhadperfor-mancelevelssimilartothosewithaharmoniouspassion.Otherre-searchalsofoundthatbothpassionsubscalesareequallyrelatedtoperformanceindicators(Vallerandetal.,2007,2008).OntheDevelopmentofPassion Despitethecontributionofthepresentresearchtotheunderstandingofthedevelopmentofpassion,itisimportanttounderscoresomelimitations.First,allthreestudiesemployedacorrelationaldesign.Itisthusimpossibletoinfercausalityfromthepresentdata.Second,theparticipantsofthepresentresearchwereallfairlyyoung(typ-icallyinthe10…22yearsoldrange).Althoughthepresentcross-sec-tionalstudiesprovideinformationastothecorrelatesofpassionoveraperiodofabout10yearsofactivityengagement,ourresearchdesignwasnotlongitudinaloverthe10-yearperiod.Futurelongitudinalresearchisneededtoreplicatethepresent“nd-ingsoveranextendedperiodoftime.Inaddition,thepresentsam-plesarenotnecessarilyrepresentativeoftheentirepopulation,andlittledemographicinformationwasavailable.Futureresearchisneededtoinvestigatepotentialassociationswithvariousdemographicvariables(e.g.,socioeconomicstatus).Third,afewmeasuresyieldedreliabilitycoef“cientsinthe.60….70range,whichcouldhavein”uencedthestabilityofour“ndings.Futurere-searchisthusneededtoreplicatesomeofthepresentresults.Fourth,thepresentdataarebasedonself-reports,whichmayin-troduceresponsebiases.Our“ndingswouldbestrengthenediftheywerereplicatedusingbehavioralorobservationaldata.Finally,dis-tinctionsbetweenpeoplewithaharmoniousandanobsessivepas-sionwerenotinvestigatedusinganobjectiveclassi“cationsystem.Peoplewereseparatedingroupsaccordingtotheirscoresonthepassionsubscalesineachsample.Thus,thenumbersofharmo-niouslyandobsessivelypassionatepeoplecannotbecomparedacrossdifferentsamples.Additionalworkisneededtodevelopaclinicallyvalidevaluationandthresholdthatcouldhelpdeterminemoreobjectivelywhohasamoreharmoniouspassionandwhoismoreobsessive.Inconclusion,thepresentpaperunderlinestheimportanceofidenti“cationwiththeactivity,preferenceforactivityspecialization,parentsactivityvaluation,andautonomysupportindeterminingwhowillbecomepassionateornotforagivenactivityaswellasthetypeofpassion(harmoniousvs.obsessive)thatwilldevelop.Furtherlongitudinalresearchisstillneeded,however,inordertobetterun-derstandtheroleofthesevariablesandothersinthedevelopmentofpassionoverthelifespan.Mageau,Vallerand,Charest,etal. REFERENCESAron,A.,Aron,E.N.,&Smollan,D.(1992).Inclusionofotherintheselfscaleandthestructureofinterpersonalcloseness.JournalofPersonalityandSocial,596…612.Black,A.E.,&Deci,E.L.(2000).Theeffectsofinstructorsautonomysupportandstudentsautonomousmotivationonlearningorganicchemistry:Aself-determinationtheoryperspective.ScienceEducation,740…756.Bloom,B.S.(Ed.).(1985a).Developingtalentinyoungpeople.NewYork:Bal-lantineBooks.Bloom,B.S.(1985b).Generalizationsabouttalentdevelopment.InB.S.BloomDevelopingtalentinyoungpeople(pp.507…549).NewYork:BallantineBoggiano,A.,Flink,C.,Shields,A.,Seelbach,A.,&Barrett,M.(1993).Useoftechniquespromotingstudentsself-determination:Effectsonstudentsana-lyticproblem-solvingskills.MotivationandEmotion,319…336.Bonebright,C.A.,Clay,D.L.,&Ankenmann,R.D.(2000).Therelationshipofworkaholismwithwork…lifecon”ict,lifesatisfaction,andpurposeinlife.JournalofCounselingPsychology,469…477.Brown,B.A.,Frankel,B.G.,&Fennell,M.P.(1989).Hugsorshrugs:Parentalandpeerin”uenceoncontinuityofinvolvementinsportbyfemaleadolescents.SexRoles,397…412.Brustad,R.J.(1988).Affectiveoutcomesincompetitiveyouthsport:Thein”u-enceofintrapersonalandsocializationfactors.JournalofSportandExercise,307…321.Cantor,N.,&Mischel,W.(1979).Prototypicalityandpersonality:Effectsonfreerecallandpersonalityimpressions.JournalofResearchinPersonality,187…Csikszentmihalyi,M.(1975).Beyondboredomandanxiety.SanFrancisco:Jossey-Csikszentmihalyi,M.(1982).Towardapsychologyofoptimalexperience.InL.Wheeler(Ed.),Reviewofpersonalityandsocialpsychology(Vol.3,pp.13…36).BeverlyHills,CA:Sage.Deci,E.L.,&Ryan,R.M.(1985).Intrinsicmotivationandself-determinationinhumanbehavior.NewYork:Plenum.Deci,E.L.,&Ryan,R.M.(1987).ThesupportofautonomyandthecontrolofJournalofPersonalityandSocialPsychology,1024…1037.Deci,E.L.,&Ryan,R.M.(2000).Thewhatandwhyofgoalpursuits:Humanneedsandtheself-determinationofbehavior.PsychologicalInquiry,227…268.Descartes,R.(1649).Lespassionsdelame.InE.S.Haldane&G.Ross(Eds.,Trans.1972),ThephilosophicalworksofDescartes.Cambridge,MA:Cam-bridgeUniversityPress.Dewey,J.(1913).Interestandeffortineducation.Boston:RiversidePress.,L.,Kairouz,S.,&Jodoin,M.(1997).Lengagement:ungagedebonheur?[Engagement:Apledgeforhappiness?].RevueQuecoisedePsychologieOntheDevelopmentofPassion Eccles,J.S.,&Barber,B.L.(1999).Studentcouncil,volunteering,basketball,ormarchingband:Whatkindofextracurricularinvolvementmatters?JournalofAdolescentResearch,10…43.Eccles,J.S.,&Wig“eld,A.(2002).Motivationalbeliefs,values,andgoals.ReviewofPsychology,109…132.Ericsson,K.A.,&Charness,N.(1994).Expertperformance:ItsstructureandAmericanPsychologist,71…76.Feertchak,H.(1996).LesmotivationsetlesvaleursenpsychosociologiepsychosociologieMotiva-tionsandvaluesinpsychosociology.Paris:ArmandColin.Festinger,L.(1957).Atheoryofcognitivedissonance.Evanston,IL:Row&Pe-Glasser,W.(1976).Positiveaddiction.NewYork:Harper&Row.Gould,D.,Tuffey,S.,Udry,E.,&Loehr,J.(1997).Burnoutincompetitivetennisplayers:II.Aqualitativeanalysis.TheSportPsychologist,341…366.Greenwald,A.G.(1980).Thetotalitarianego:Fabricationandrevisionofper-sonalhistory.AmericanPsychologist,603…618.Grolnick,W.S.,Benjet,C.,Kurowski,C.O.,&Apostoleris,N.H.(1997).Pre-dictorsofparentinvolvementinchildrensschooling.JournalofEducational,538…548.Grolnick,W.S.,&Ryan,R.M.(1987).Autonomyinchildrenslearning:Anexperimentalandindividualdifferenceinvestigation.JournalofPersonalityandSocialPsychology,890…898.Grolnick,W.S.,&Ryan,R.M.(1989).Parentstylesassociatedwithchildrensself-regulationandcompetenceinschool.JournalofEducationalPsychology,143…154.Grolnick,W.S.,Ryan,R.M.,&Deci,E.L.(1991).Theinnerresourcesforschoolachievement:Motivationalmediatorsofchildrensperceptionsoftheirpar-JournalofEducationalPsychology,508…517.Hat“eld,E.,&Walster,G.W.(1978).Anewlookatlove.Reading,MA:Addison-Hodgins,H.S.,&Knee,C.R.(2002).Theintegratingselfandconsciousexpe-rience.InE.L.Deci&R.M.Ryan(Eds.),Handbookofself-determination(pp.87…100).Rochester,NY:TheUniversityofRochesterPress.Houser-Marko,L.,&Sheldon,K.M.(2006).Motivatingbehavioralpersistence:Theself-as-doerconstruct.PersonalityandSocialPsychologyBulletinKasser,T.,&Ryan,R.M.(2001).Becarefulwhatyouwishfor:Optimalfunc-tioningandtherelativeattainmentofintrinsicandextrinsicgoals.InP.Sch-muck&K.M.Sheldon(Eds.),Lifegoalsandwell-being:Towardsapositivepsychologyofhumanstriving(pp.116…131).Ashland,OH:Hogrefe&Huber.Koestner,R.,Ryan,R.M.,Bernieri,F.,&Holt,K.(1984).Settinglimitsonchildrensbehavior:Thedifferentialeffectsofcontrollingversusinformationalstylesonintrinsicmotivationandcreativity.JournalofPersonality,233…Koestner,R.,&Zuckerman,M.(1994).Causalityorientations,failure,andJournalofPersonality,321…346.Mageau,Vallerand,Charest,etal. Krapp,A.(2002).Structuralanddynamicaspectsofinterestdevelopment:The-oreticalconsiderationsfromontogeneticperspective.LearningandInstruction,383…409.Lagache,D.(1936).Passionsetpsychosespassionnelles[PassionsandpassionatevolutionPsychiatrique,3…29.Mageau,G.A.,&Vallerand,R.J.(2003).Thecoach…athleterelationship:Amotivationalmodel.JournalofSportsSciences,881…954.Mageau,G.A.,&Vallerand,R.J.(2007).Themoderatingeffectofpassionontherelationbetweenactivityengagementandpositiveaffect.MotivationandEmo-,312…321.Mageau,G.A.,Vallerand,R.J.,Rousseau,F.L.,Ratelle,C.F.,&Provencher,P.J.(2005).Passionandgambling:Investigatingthedivergentaffectiveandcognitiveconsequencesofgambling.JournalofAppliedSocialPsychologyMarkus,H.(1977).Self-schemataandprocessinginformationabouttheself.JournalofPersonalityandSocialPsychology,63…78.Nakamura,J.,&Csikszentmihalyi,M.(2003).Theconstructionofmeaningthroughvitalengagement.InC.L.M.Keyes&J.Haidt(Eds.),Positivepsychologyandthelifewell-lived(pp.83…104).Washington,DC:AmericanPsychologicalAssocation.Ommundsen,Y.,&Vaglum,P.(1991).Soccercompetitionanxietyandenjoymentinyoungboyplayers:Thein”uenceofperceivedcompetenceandsigni“cantothersemotionalinvolvement.InternationalJournalofSportPsychologyPaturet,J.-B.(2001).Pouvoirdelapassion[Powerofpassion].InJ.Ašn(Ed.),Passions:Alienation&liberte[Passions:Alienation&freedom](pp.9…22).RamonvilleSaint-Agne,France:EPelletier,L.G.,&Vallerand,R.J.(1996).Supervisorsbeliefsandsubordinatesintrinsicmotivation:Abehavioralcon“rmationanalysis.JournalofPersonal-ityandSocialPsychology,331…340.Ratelle,C.F.,Vallerand,R.J.,Mageau,G.A.,Rousseau,F.L.,&Provencher,P.J.(2004).Whenpassionleadstopathology:Alookatgambling.JournalofGamblingStudies,105…119.Rathunde,K.,&Csikszentmihalyi,M.(1993).Undividedinterestandthegrowthoftalent:Alongitudinalstudyofadolescents.JournalofYouthandAdoles-,1…21.Rip,B.,Fortin,S.,&Vallerand,R.J.(2006).Therelationshipbetweenpassionandinjuryindancestudents.JournalofDanceMedicine&Science.,14…20.Roberts,G.C.(1984).Achievementmotivationinchildrenssport.InJ.NichollsAdvancesinmotivationandachievement(Vol.3,pp.251…281).Green-wich,CT:JAIPress.Rony,J.-A.(1990).Lespassions[ThePassions].Paris:PressesUniversitairesdeRousseau,F.L.,Vallerand,R.J.,Ratelle,C.F.,Mageau,G.A.,&Provencher,P.J.(2002).Passionandgambling:OnthevalidationoftheFrenchversionoftheGamblingPassionScale(GPS).JournalofGamblingStudies,45…66.OntheDevelopmentofPassion Rudy,E.B.,&Estock,P.J.(1990).Runningaddictionanddyadicadjustment.ResearchinNursingandHealth,219…225.Ryan,R.M.(1995).Psychologicalneedsandthefacilitationofintegrativepro-JournalofPersonality,397…427.Sacks,M.H.,&Sachs,M.L.(1981).Psychologyofrunning.Champaign,IL:HumanKinetics.Scanlan,T.K.,Carpenter,P.J.,Lobel,M.,&Simons,J.P.(1993).Sourcesofenjoymentforyouthsportathletes.PediatricExerciseScienceSchlenker,B.R.(1985).Identityandself-identi“cation.InB.R.Schlenker(Ed.),Theselfandsociallife(pp.65…99).NewYork:McGraw-Hill.vesque,C.,Laliberte,M.-L.N.,Pelletier,L.G.,Blanchard,C.,&Vallerand,R.J.(2003).HarmoniousandobsessivepassionfortheInternet:Theirassociationswiththecouplesrelationship.JournalofAppliedSocial,197…221.Seligman,M.E.P.,&Csikszentmihalyi,M.(2000).Positivepsychology:Anin-AmericanPsychologist,5…14.Sheldon,K.M.,&Elliot,A.J.(1998).Notallpersonalgoalsarepersonal:Com-paringautonomousandcontrolledreasonsforgoalsaspredictorsofeffortandPersonalityandSocialPsychologyBulletin,546…557.Sheldon,K.M.,&Elliot,A.J.(1999).Goalstriving,needsatisfaction,andlon-gitudinalwell-being:Theself-concordancemodel.JournalofPersonalityandSocialPsychology,482…497.Snyder,E.E.,&Spreitzer,E.(1973).Familyin”uencesandinvolvementinsports.ResearchQuarterly,249…255.Spinoza,B.(1953).thique[Ethics].Paris:Flammarion.Spreitzer,E.,&Snyder,E.E.(1976).Socializationintosport:AnexploratorypathResearchQuarterly,238…245.Stein,G.L.,Raedeke,T.D.,&Glenn,S.D.(1999).Childrensperceptionsofparentsportinvolvement:Itsnothowmuch,buttowhatdegreethatsim-JournalofSportBehavior,591…601.Vallerand,R.J.,Blanchard,C.,Mageau,G.A.,Koestner,R.,Ratelle,C.F.,&onard,M.,etal.(2003).Lespassionsdelame:Onobsessiveandharmoniouspassion.JournalofPersonalityandSocialPsychology,756…Vallerand,R.J.,&Houlfort,N.(2003).Passionatwork:Towardanewconcep-tualization.InD.Skarlicki,S.Gilliland,&D.Steiner(Eds.),Socialissuesinmanagement:Vol.3.Emergingperspectivesofvaluesinorganizations(pp.175…204).Greenwich,CT:InformationAgePublishing.Vallerand,R.J.,&Miquelon,P.(2007).Passionforsportinathletes.InD.Lav-e&S.Jowett(Eds.),Socialpsychologyinsport(pp.249…263).Champaign,IL:HumanKinetics.Vallerand,R.J.,Mageau,G.A.,Elliot,A.J.,Dumais,A.,Demers,M.-A.,&Rousseau,F.(2008).Passionandperformanceattainmentinsport.ofSportandExercise,373…392.Mageau,Vallerand,Charest,etal. Vallerand,R.J.,Rousseau,F.L.,Grouzet,F.M.E.,Dumais,A.,Grenier,S.,&Blanchard,C.B.(2006).Passioninsport:AlookatdeterminantsandaffectiveJournalofSport&ExercisePsychology,454…478.Vallerand,R.J.,Salvy,S.J.,Mageau,G.A.,Elliot,A.J.,Denis,P.,&Grouzet,F.M.E.,etal.(2007).Ontheroleofpassioninperformance.JournalofPer-,505…533.Waterman,A.S.(1990).Personalexpressiveness:PhilosophicalandpsychologicalJournalofMindandBehavior,47…74.Waterman,A.S.(1993).Twoconceptionsofhappiness:Contrastsofpersonalexpressiveness(eudaimonia)andhedonicenjoyment.JournalofPersonalityandSocialPsychology,678…691.Waterman,A.S.(2004).Findingsomeonetobe:Studiesontheroleofintrinsicmotivationinidentityformation.Identity:AnInternationalJournalofTheoryandResearch,209…228.Weiss,M.R.,&Hayashi,C.T.(1995).Allinthefamily:Parent-childin”uencesincompetitiveyouthgymnastics.PediatricExerciseSciences,36…48.OntheDevelopmentofPassion 646