Exploring the Mechanisms Underlying Emotional ...

From Sound to Significance: Exploring the Mechanisms Underlying Emotional Reactions to Music PATRIK N. JUSLIN and GONÇALO BARRADAS Uppsala University TUOMAS EEROLA Durham University A common approach to studying emotional reactions to music is to attempt to obtain direct links between musical surface features such as tempo and a listener’s responses. However, such an analysis ultimately fails to explain why emotions are aroused in the listener. In this article we explore an alternative approach, which aims to account for musical emotions in terms of a set of psychological mechanisms that are activated by different types of information in a musical event. This approach was tested in 4 experiments that manipulated 4 mechanisms (brain stem reflex, contagion, episodic memory, musical expectancy) by selecting existing musical pieces that featured information relevant for each mechanism. The excerpts were played to 60 listeners, who were asked to rate their felt emotions on 15 scales. Skin conductance levels and facial expressions were measured, and listeners reported subjective impressions of relevance to specific mechanisms. Results indicated that the target mechanism conditions evoked emotions largely as predicted by a multimechanism framework and that mostly similar effects occurred across the experiments that included different pieces of music. We conclude that a satisfactory account of musical emotions requires consideration of how musical features and responses are mediated by a range of underlying mechanisms.

.FSFTPVOEDBOIBWFQSPGPVOEFďFDUTPOMJTUFOFST It can be argued that sound is more intimate than sight, “more inside our ‘heads’” (Thompson, 2009, Qõöù /PXIFSFJTUIJTJNQBDUNPSFBQQBSFOUUIBOJO UIFDBTFPGNVTJD4UJMM JUJTPGUFOSFHBSEFEBTPOFPG the great mysteries in life that music, which consists only of abstract tone sequences, is able to arouse such TUSPOHFNPUJPOT8IBUJTNPSF MJTUFOFSTEJďFSXJEFMZJOUIFJSFNPUJPOBMSFBDUJPOTUPNVTJD 4MPCPEB õýýú 5IVT (VUIFJM õýùö BTLFEXIFUIFSNVTJDBOE emotion research “can ever reach the goal of science,

namely to discover laws of cause and effect in order to predict the resultsu Qõõ *OUIJTTUVEZ XFUFTUGPVS different causal mechanisms for how music could evoke emotions and show that despite their elusiveness, emotional reactions to music can be predicted UPBHSFBUFSFYUFOUUIBOJTVTVBMMZCFMJFWFE Music and Emotion Research

.VTJDJTVCJRVJUPVTJOUPEBZTTPDJFUZ /PSUI)BSHSFBWFT öôôü *UPGUFOPDDVSTJOBWBSJFUZPGTPDJBM contexts and accompanies people’s activities “from

American Journal of Psychology

'BMMöôõù 7PMõöü /P÷QQöüõm÷ôør¥ƎöôõùCZUIF#PBSEPG5SVTUFFTPGUIF6OJWFSTJUZPG*MMJOPJT

This content downloaded from 130.238.218.35 on Thu, 05 Nov 2015 14:49:29 UTC All use subject to JSTOR Terms and Conditions

UIFDSBEMFUPUIFHSBWFu (SFHPSZ õýýû Qõöø *OB significant proportion of these life episodes (Juslin, -JMKFTUSÕN 7ÅTUGKÅMM #BSSBEBT 4JMWB öôôü

UIF music arouses an emotionJOUIFMJTUFOFS*OEFFE the strong emotions music evokes help to explain why people spend more money on music than on QSFTDSJQUJPOESVHT )VSPO öôôõ Emotions belong to the broader field of affect, which also comprises moods, preferences, and perTPOBMJUZ USBJUT 0BUMFZ ,FMUOFS +FOLJOT öôôú $IBQUFSõ "MUIPVHIUIFSFBSFOVNFSPVTEFêOJUJPOT of emotion FH *[BSE öôõô,MFJOHJOOB,MFJOHJOOB õýüõ

+VTMJO öôõõ PďFSTBXPSLJOHEFêOJUJPOUIBU captures how many contemporary emotion researchFSTDPODFJWFPGUIFQIFOPNFOPO TFFBMTP+VTMJO 4DIFSFS öôôùDG"MUBSSJCB öôõö Emotions are relatively brief, intense, and rapidly changing reactions to potentially important FWFOUT TVCKFDUJWFDIBMMFOHFTPSPQQPSUVOJUJFT in the external or internal environment—often of a social nature—which involve a number of subcomponents (cognitive changes, subjective feelings, expressive behavior and action tendenDJFT UIBUBSFNPSFPSMFTTiTZODISPOJ[FEuEVSJOHBOFNPUJPOBMFQJTPEF Qõõø

In a musical context, emotion is one of several aspects of music experience (which also encompass physical, behavioral, perceptual, cognitive, existenUJBM BOEEFWFMPQNFOUBMBTQFDUTTFF(BCSJFMTTPO öôõõ UIBUNBZPSNBZOPUCFQSFTFOUJOBOZJOEJWJEVBM JOTUBODFPGNVTJDMJTUFOJOH"TGBSBTMBZMJTUFOFSTBSF concerned, however, it may be one of the most important aspects (Juslin, Liljeström, Laukka, Västfjäll, -VOERWJTU öôõõ 4FWFSBMDPNQPOFOUTPGFNPUJPOIBWFCFFOTIPXO to be affected by musical stimuli, including subjective GFFMJOH 1JLF õýûö

FNPUJPOBMFYQSFTTJPO 8JUWMJFU 7SBOB öôôû

QIZTJPMPHZ ,SVNIBOTM õýýû

CSBJO BDUJWBUJPO #SPXO .BSUJOF[ 1BSTPOT öôôø

BOE CFIBWJPSUFOEFODZ 'SJFE#FSLPXJU[ õýûý 5IFre is also evidence of a synchronization between the WBSJPVTDPNQPOFOUT TFF+VTMJO )BSNBU &FSPMB öôõø-VOERWJTU $BSMTTPO )JMNFSTTPO +VTMJO öôôý +VTMJOBOE;FOUOFS öôôö OPUFUIBUiJODMVsion of measures of different sub-components might increase our ability to decide exactly what kind of reBDUJPOIBTPDDVSSFEUPBNVTJDBMFWFOUu Qû 5IJTJT

especially important in experimental studies because emotions are not easy to arouse in artificial laboratory FOWJSPONFOUT 1MVUDIJL õýýø It is arguably less of a problem to obtain cases of “genuine” emotions to music in field studies, which show that musical emotions occur in a wide range PGTFUUJOHTJOFWFSZEBZMJGF4UJMM JUNVTUCFBDLOPXMedged that we often hear music without actually feeling any emotion at all, at least not one aroused by UIFNVTJD"DDPSEJOHUPTPNFFTUJNBUFT UIFNVTJD evokes emotions in only about 55–65% of the episodes, and there seem to be large individual differFODFTJOPWFSBMMQSFWBMFODF +VTMJO-BVLLB öôôø +VTMJOFUBM öôôü .PSFPWFS UIFSFBSFDPOTJEFSBCMF individual differences between listeners regarding the precise emotional state a musical event will arouse (PXFOTNJUI#MPPN õýýû 5IVT BMUIPVHIGFX researchers today would deny “that some music is capable of exciting some emotion in some people TPNFPGUIFUJNFu #BMM öôõô Qöùû

BNPSFEFMJDBUF issue is to explain why the emotion occurred in the êSTUQMBDF*GPOMZTPNFNVTJDBMFWFOUTTVDDFFEJO arousing an emotion, and if different listeners might respond differently to the same piece of music, what are the precise conditions that will cause a specific emotion to occur? Causes of Musical Emotions

.VTJDBMFNPUJPOTBSFPGUFOSFHBSEFEBTQV[[MJOH %PXMJOH)BSXPPE õýüú 1BSUPGUIFQV[[MFJT that the conditions of emotion elicitation in music and outside music appear so different (Krumhansl, õýýû

BUMFBTUPOUIFTVSGBDF*OUIFQBSBEJHNBUJDDBTF an emotion is aroused when an event is appraised as having the capacity to influence the goals or plans of UIFQFSDFJWFS5IFNBJOQSPCMFNJTUIBU XIFOXF listen to music, only rarely does the music per se have JNQMJDBUJPOTGPSPVSHPBMTJOMJGF3FTFBSDIFSTIBWF thus expressed concerns over the limitations of appraisal theory: “Explaining emotional responses to instrumental music is a real problem for appraisal theories and may be a real threat to the generality of appraisals as elicitors of emotion” (Ellsworth, 1994, Qõýù "DIBMMFOHFGPSSFTFBSDIFSTIBTUIVTCFFOUPDPNF VQXJUIBMUFSOBUJWFBDDPVOUT4PNFBVUIPSTIBWFBSgued that musical features affect listeners “directly” 3PCJOTPO öôôù Q÷ýö

CVUUIFVTFPGUIFXPSE

t JUSLIN ET AL .


directly only goes to show that no causal explanation IBTZFUCFFOPďFSFE5IFCFMJFGUIBUPCKFDUTPSFWFOUT “directly” cause our emotions is called the nativist GBMMBDZ 4JMWJB öôõö *UBQQFBSTUPCFQBSUJDVMBSMZ common in music research, maybe because we tend to think of music as abstract sequences of notes, deWPJEPGTFNBOUJDNFBOJOH5IVT NBOZTDIPMBSTIBWF aimed to obtain direct links between surface features PGUIFNVTJDBOEFWPLFEFNPUJPOT#VUTVDIDPSSFMBUJPOTEPOPUDPOTUJUVUFBOFYQMBOBUJPOUIFZTJNQMZ NPWFUIFCVSEFOPGFYQMBOBUJPOGSPNPOFMFWFM i8IZ EPFTUIFTFDPOENPWFNFOUPG#FFUIPWFOTA&SPJDB TZNQIPOZBSPVTFTBEOFTT u UPBOPUIFSMFWFM i8IZ EPFTTMPXUFNQPBSPVTFTBEOFTT u 5IJTBQQSPBDI DPOGVTFT SF EFTDSJQUJPOXJUIFYQMBOBUJPO 8FTVCNJUUIBUSFBMQSPHSFTTJOUIFTUVEZPGNVTJD and emotion can be achieved only if we understand the underlying process that mediates between surGBDFGFBUVSFTBOEBSPVTFEFNPUJPOT"HPPEUIFPSZ of emotion causation should explain both why a givFOFWFOUBSPVTFTBOFNPUJPO FMJDJUBUJPO BOEXIZ the aroused emotion is of a specific kind (differenUJBUJPO 5IFQTZDIPMPHJDBMQSPDFTTUISPVHIXIJDI this is achieved is referred to here as the underlying NFDIBOJTN*UDPNQSJTFTBGVODUJPOBMEFTDSJQUJPOPG XIBUUIFNJOEJTEPJOH JOQSJODJQMF FH iSFUSJFWJOH BNFNPSZu

XIJDITIPVMEOPUCFDPOGVTFEXJUIJUT implementation in the brain, or with the phenomenoMPHJDBMFYQFSJFODFJUTFFLTUPFYQMBJO %FOOFUU õýüû Theories of Underlying Mechanisms

"MUIPVHIGFXTUVEJFTIBWFBUUFNQUFEUPUFTUUIFPries of possible mechanisms underlying emotional SFBDUJPOTUPNVTJD GPSBEJTDVTTJPO TFF+VTMJO 7ÅTUGKÅMM öôôü

TFWFSBMTDIPMBSTIBWFQSPQPTFEQPTsible mechanisms over the years, typically limiting UIFNTFMWFTUPPOFPSBGFX.FZFS õýùú XBTBSHVBCMZ the first modern scholar who understood the important role of psychological theory in unraveling relationships between musical structure and emotional response and offered a useful theory of musical exQFDUBODZ%PXMJOHBOE)BSXPPE õýüú BSHVFEUIBU conditioned responses form part of the composer’s toolkit for evoking emotion, and they provided sevFSBMFYBNQMFTPGUIJTQSJODJQMF#BVNHBSUOFS õýýö proposed that episodic memory plays an important role in emotional responses to music and presented TVSWFZEBUBUPCBDLVQUIFDMBJN8BUFSNBO õýýú

made an early attempt to apply appraisal theory (the WFSTJPOPVUMJOFECZ0SUPOZ $MPSF $PMMJOT õýüü UPNVTJDBMFWFOUT+VTMJO öôôô QSPQPTFEUIBUiFNPtional contagion” via voicelike features of the music BSFSFTQPOTJCMFGPSBTVCTFUPGNVTJDBMFNPUJPOT 4MPCPEBBOE+VTMJO öôôõ BOE4DIFSFSBOE;FOUOFS öôôõ EJTDVTTFETPNFPGUIFTFNFDIBOJTNTCVUEJE not attempt to formulate an integrated framework featuring a set of hypotheses that could guide researchFSTJOUIFêFME5IFNPTUDPNQSFIFOTJWFBUUFNQUUP describe a set of mechanisms was begun in the midöôôôTBOESFTVMUFEJOUIF#3&$7&."GSBNFXPSL +VTMJO öôôù öôõ÷+VTMJO7ÅTUGKÅMM öôôü+VTMJO -JMKFTUSÕN 7ÅTUGKÅMM -VOERWJTU öôõô

OBNFEBGUFS the first letters of the mechanisms included (listed MBUFSJOUIJTBSUJDMF 5IF#3&$7&."GSBNFXPSLUBLFTBTJUTQPJOU PGEFQBSUVSFBOFWPMVUJPOBSZQFSTQFDUJWF"OFWPMVtionary perspective on human perception of sounds suggests that the survival of our ancient ancestors depended on their ability to detect patterns in sounds, derive meaning from them, and adjust their behavJPSBDDPSEJOHMZ1SPDFFEJOHGSPNUIJTBTTVNQUJPO JU is theorized that there are several emotion mechanisms implemented by a number of distinct “brain networks,” which developed gradually and in a particular order during evolution, from simple reflexes to DPNQMFYKVEHNFOUT&BDINFDIBOJTNJTSFTQPOTJWF in its own unique way to specific configurations of information in the music, the listener, and the situation, referred to jointly as “the musical event” (Juslin, öôõ÷ 5IFGSBNFXPSLGFBUVSFTFJHIUNFDIBOJTNT õ #SBJOTUFNSFëFY BIBSEXJSFEBUUFOUJPO response to simple acoustic features such as FYUSFNFPSJODSFBTJOHMPVEOFTTPSTQFFE 4PLPMPW õýú÷

ö 3IZUINJDFOUSBJONFOU BHSBEVBMBEKVTUNFOU PGBOJOUFSOBMCPEZSIZUIN FH IFBSUSBUF UPXBSEBOFYUFSOBMSIZUINJOUIFNVTJD )BSSFS )BSSFS õýûû

÷ &WBMVBUJWFDPOEJUJPOJOH BSFHVMBSQBJSJOHPGB piece of music and other positive or negative stimuli, leading to a conditioned association #MBJS4IJNQ õýýö

ø $POUBHJPO BOJOUFSOBMNJNJDSZPGUIFQFSceived voicelike emotional expression of the NVTJD +VTMJO öôôô

ù 7JTVBMJNBHFSZ JOOFSJNBHFTPGBOFNPUJPOBM character conjured up by the listener through a

EMOTIONAL REACTIONS TO MUSIC t


metaphorical mapping of the musical structure 0TCPSOF õýüô

ú &QJTPEJDNFNPSZ BDPOTDJPVTSFDPMMFDUJPOPGB particular event from the listener’s past that is USJHHFSFECZUIFNVTJD #BVNHBSUOFS õýýö

û .VTJDBMFYQFDUBODZ BSFTQPOTFUPUIFHSBEVBM unfolding of the syntactical structure of the music and its expected or unexpected continuBUJPO .FZFS õýùú

ü "FTUIFUJDKVEHNFOU BTVCKFDUJWFFWBMVBUJPOPG the aesthetic value of the music based on an JOEJWJEVBMTFUPGXFJHIUFEDSJUFSJB +VTMJO öôõ÷

#Z TZOUIFTJ[JOH UIFPSZ BOE EBUB GSPN NBOZ domains mostly outside music, Juslin and Västfjäll öôôü XFSFBCMFUPEFWFMPQUIFêSTUTFUPGIZQPUIFTFT that may help researchers distinguish between the NFDIBOJTNT GPSBOVQEBUF TFF+VTMJO öôõ÷ 0OF implication is that, in order for data to contribute in a cumulative fashion to our knowledge, researchers need to specify as far as possible the underlying NFDIBOJTNJOWPMWFEJOFBDIFNPUJPOFQJTPEF Empirical Approaches to Mechanisms

(JWFO UIF DSVDJBM SPMF PG UIFPSJFT PG VOEFSMZJOH mechanisms in explaining emotions to music, there is an urgent need for empirical studies that attempt UPUFTUTQFDJêDNFDIBOJTNT5PEBUF NPTUEBUBDPNF GSPNêFMETUVEJFTUIBUSFMZPOTFMGSFQPSU#BVNHBSUOFS õýýö BOE+BOBUB 5PNJD BOE3BLPXTLJ öôôû FYQMPSFE FQJTPEJD NFNPSJFT MJOLFE UP NVTJD " broader selection of mechanisms were surveyed by +VTMJOFUBM öôôü BOECZ%JOHMF 4BWJMM 'SBTFS BOE 7JFUI öôõõ "MUIPVHIêFMETUVEJFTDBOPďFSTPNF clues about the occurrence of various mechanisms in real-life contexts, these data should be treated with caution: Field data do not enable researchers to draw strong conclusions with regard to causal relationships because of insufficient experimental DPOUSPM*UIBTCFFOEFNPOTUSBUFEUIBUQFPQMFHFOerally have a poor understanding of the causes of their own behavior and that it is problematic to rely solely on subjective self-reports of causes of emotions, because we could be unaware of the trigger UIBUFMJDJUTUIFFNPUJPO 'PY öôôü 5IJTBQQMJFTUP NVTJDBMFNPUJPOTBMTP*UJTUIFSFGPSFOFDFTTBSZUP conduct experiments in a laboratory setting, where specific mechanisms may be manipulated so as to QSPEVDFJNNFEJBUFFďFDUTPOCFIBWJPSBMNFBTVSFT

This task is challenging however, and requires a set PGTUSBUFHJFT To separate the effects of distinct mechanisms, we need to be able to activate as well as suppress specific mechanisms in each case by manipulating different aspects of the music, the listening situation, and the MJTUFOFS5IJTDPVMECFEPOFJOBUMFBTUUISFFQSJODJQBM ways: First, one might select or manipulate pieces of music in such a manner as to provide or withhold the information required for a certain mechanism to be activated while leaving or removing other information UIFQSJODJQMFPGJOGPSNBUJPOTFMFDUJPO 4FDPOE POF can design the specific test procedure in such a way that it will prevent the type of information processing needed for a mechanism to be activated (the principle PGJOUFSGFSFODF 'PSJOTUBODF POFDBOHJWFMJTUFOFST a task that recruits attentional resources to such an extent that visual imagery, also dependent on these SFTPVSDFT XJMMCFNBEFJNQPTTJCMF5IJSE POFDBO manipulate the listener by creating new memories during the test procedure before presenting the tarHFUTUJNVMVT UIFQSJODJQMFPGQSPDFEVSBMIJTUPSZ

UP enable study of mechanisms such as evaluative conEJUJPOJOH "TTIPVMECFBQQBSFOU FYQMPSJOHNFDIBOJTNT must entail experimental studies, and there are at least two complementary experimental strategies UIBUNJHIUCFVTFE'JSTU POFNBZBUUFNQUUPEJSFDUMZ manipulate musical features or situational circumstances to activate particular mechanisms by means PGIJHIMZDPOUSPMMFETZOUIFTJ[FE PSSFTZOUIFTJ[FE QJFDFT5IJTBQQSPBDIQFSNJUTTUSPOHDPODMVTJPOT about causal relationships, but suffers from low ecoMPHJDBMWBMJEJUZ4QFDJêDBMMZ UIFNVTJDBMTUJNVMJDPVME bear little relation to music as experienced by most MJTUFOFSTJOUIFSFBMXPSME5IVT BTFDPOEBQQSPBDI is to attempt to find existing pieces of music that include musical characteristics relevant to specific NFDIBOJTNT"MUIPVHIUIFJOUFSOBMWBMJEJUZDPVME be reduced, the use of real music makes it easier to arouse intense emotions in listeners, and the ecologiDBMWBMJEJUZJTJODSFBTFE Only the synthesis approach has been used in TUVEJFTTPGBS"QJPOFFSJOHFYQFSJNFOUCZ4UFJOCFJT ,PFMTDI BOE4MPCPEB öôôú VTFETVCKFDUJWF and physiologic measures to capture emotional reacUJPOTUPVOFYQFDUFEIBSNPOJDQSPHSFTTJPOT4UJNVMJ DPOTJTUFE PG UISFF NBUDIFE WFSTJPOT PG TJY #BDI

t JUSLIN ET AL .


MJGF +VTMJOFUBM öôôü öôõõ

XIFSFBTUIFGPVSUI musical expectancy—is often regarded as particularly JNQPSUBOUCZNVTJDSFTFBSDIFST 5IPNQTPO öôôý To activate each of these mechanisms, we selected pieces of music that included extreme acoustic events CSBJOTUFNSFëFY

WPJDFMJLFFNPUJPOBMFYQSFTTJPOT DPOUBHJPO

VOFYQFDUFENVTJDBMTFRVFODFT NVTJDBMFYQFDUBODZ

BOEQJFDFTMJOLFEXJUITJHOJêDBOU MJGFFWFOUTGPSNPTUQFPQMF FQJTPEJDNFNPSZ

SFTQFDUJWFMZ5PBWPJEBOZFďFDUPGBUBSHFUNFDIBOJTN being merely an artifact of a specific piece of music, we included four pieces of music to represent each NFDIBOJTN5IFQJFDFTEJďFSFEJONBOZSFTQFDUT FH JOTUSVNFOUBUJPO QJUDIMFWFM BQBSUGSPNUIF specific types of information that were considered OFDFTTBSZGPSUIFUBSHFUNFDIBOJTN 8FTFMFDUFEQSPNJOFOUQJFDFTPGNVTJDCZFNJnent composers, to be able to produce emotions as JOUFOTFBTQPTTJCMF"UUIFTBNFUJNF XFXBOUFEUP isolate the effects of individual mechanisms as much as possible, so that no mechanisms other than those targeted in a given condition would diffuse the effect PGUIFUBSHFUNFDIBOJTN.PTUQSPCMFNBUJDJOUIJT regard are memories or personal associations with QBSUJDVMBSQJFDFTUIBUIBWFCFFOIFBSEQSFWJPVTMZ To facilitate a selective activation of the nonmemory mechanisms, we featured pieces likely to be unfamilJBSUPUIFMJTUFOFST"TOPUFECZ)BSHSFBWFT õýüú

classical music is a minority interest among music MJTUFOFST FWFOJOUIF8FTUFSOXPSME*O#BVNHBSUOFST õýýö TVSWFZ GFBUVSJOHDPMMFHFTUVEFOUT BMM pieces of music associated with an episodic memory FYDFQUUXPJOWPMWFEQPQSPDLPSGPMLNVTJD5IFSFRationale for the Present Study The aim of this study was to test four of the causal fore, we expected that using classical music would mechanisms hypothesized to underlie emotional re- tend to minimize unwanted memory effects, as verisponses to music using more ecologically valid pieces fied also by having listeners rate their familiarity with PGNVTJD"MUIPVHIXFVTFEFYDFSQUTGSPNUIFFYJTU- each piece and report whether the music evoked any ing repertoire of music, by selecting these carefully NFNPSJFT$POWFSTFMZ UIFNFNPSZDPOEJUJPOXPVME to provide or withhold information theoretically as- feature pieces that were likely to be highly familiar to sumed to be needed for specific mechanisms, we UIFMJTUFOFST aimed to manipulate the mechanisms in a selective 8FVTFEDPOWFSHJOHFWJEFODFGSPNNVMUJQMFNFBmanner, to produce predictable effects on listeners’ sures to draw more valid conclusions about evoked FNPUJPOTUIBOXPVMECFQPTTJCMFGSPNBTJOHMFJOEFY SFTQPOTFT To render different studies comparable, we in- )FODF JOBEEJUJPOUPTFMGSFQPSUTPGFNPUJPOT XF cluded the same four mechanisms as were manipu- obtained post hoc self-reports with regard to mechaMBUFEJO+VTMJOFUBM öôõø 5ISFFPGUIFTFCSBJOTUFN OJTNT .FD4DBMF BOEQTZDIPQIZTJPMPHJDNFBTVSFT reflex, contagion, and episodic memory—appear to TLJODPOEVDUBODFMFWFMBOEGBDJBMFMFDUSPNZPHSBQIZ be among the most commonly occurring in everyday and also used a control condition in the form of a chorales, which differed only in terms of one chord, harmonically either expected, unexpected, or very VOFYQFDUFE3FTVMUTTIPXFEUIBUGFMUUFOTJPO PWFSBMM subjective emotionality, and electrodermal activity all JODSFBTFEXJUIJODSFBTFTJOVOFYQFDUFEOFTT5IFTF results rendered support for the musical expectancy NFDIBOJTNJOUFSNTPGTUZMJTUJDWJPMBUJPOTPG8FTUFSO DMBTTJDBMNVTJD .FZFS õýùú *OBSFDFOUTUVEZ +VTMJOFUBM öôõø

XFBMTPVTFE BTZOUIFTJTBQQSPBDI8FNBOJQVMBUFEBQJFDFPGNVsic to activate four different mechanisms: brain stem reflex, contagion, episodic memory, and musical exQFDUBODZ5IJTXBTEPOFCZNFBOTPGBDBSFGVMFEJUJOH of computerized but natural-sounding performances, inserting or removing the information needed for TQFDJêDNFDIBOJTNT3FTVMUTGSPNBMJTUFOJOHUFTUSFvealed that the target mechanism conditions aroused UIFTQFDJêDFNPUJPOTQSFEJDUFE This was the first experiment to manipulate and contrast different target mechanisms in the inducUJPOPGFNPUJPOTUISPVHINVTJDMJTUFOJOH)PXFWFS UIFTUVEZJODMVEFEPOMZBTJOHMF PSJHJOBM QJFDFPG NVTJD"MUIPVHIJUNJHIUCFBSHVFEUIBUUIFPSJHJOBM piece served mainly as a carrier of different types of information, the results clearly need to be replicated XJUIPUIFSQJFDFT.PSFPWFS UIFTZOUIFTJ[FETUJNVMJ were fairly simple compared with most “real” music JOPSEFSUPBDIJFWFFYQFSJNFOUBMDPOUSPM*UJTUIVT crucial to demonstrate that such effects can be obtained with real pieces of music by real composers, OPUNFSFMZXJUIBSUJêDJBMTUJNVMJ



sumptions that zygomaticus muscle activity in the face might reveal the valence of a response -BOH (SFFOXBME #SBEMFZ )BNN õýý÷ and that the memory condition would arouse IBQQJOFTTBOEOPTUBMHJB JF QPTJUJWFFNPUJPOT

XFQSFEJDUFEUIBUUIFNFNPSZDPOEJUJPO would produce more zygomatic muscle activity than would the contagion condition (sadness JOWPMWFTOFHBUJWFWBMFODF 'JOBMMZ CBTFEPOUIF assumptions that corrugator muscle activity JTSFëFDUJWFPGOFHBUJWFFNPUJPOT -BOHFUBM õýý÷ BOEUIBUUIFDPOUBHJPODPOEJUJPOXPVME arouse sadness, we predicted that the contagion condition would show more corrugator activity UIBOUIFNFNPSZDPOEJUJPO

“neutral” piece of music, to help rule out alternaUJWFFYQMBOBUJPOT5IFGPMMPXJOHQSFEJDUJPOT TJNJMBS BDSPTTUIFFYQFSJNFOUT XFSFUFTUFE HSPVQFEBDDPSEing to type of measure: Emotion ratings: Listeners rated feelings on 15 scales (based on earlier studies of music and FNPUJPO 8FQSFEJDUFEUIBUUIFCSBJOTUFN reflex condition would evoke mainly surprise, the contagion condition would evoke mainly sadness, the expectancy condition would evoke mainly anxiety, and the memory condition would evoke mainly nostalgia and happiness (for a more detailed description, see the Method TFDUJPO .FD4DBMF5IJTTDBMF +VTMJOFUBM öôõø QVSported to capture the mechanisms that had occurred and consisted of eight simple questions, each targeting one of the mechanisms in the #3&$7&.GSBNFXPSL +VTMJOFUBM öôõô

QMVT BQQSBJTBM5IFJEFBXBTUIBUBMUIPVHITPNFPG the mechanisms are implicit in nature, they may co-occur with subjective impressions that can CFSFQPSUFECZMJTUFOFST'PSJOTUBODF BMJTUFOFS influenced by the expectancy mechanism might find the music difficult to predict, whereas a listener who becomes aroused through the episodic memory mechanism might report conTDJPVTSFDPMMFDUJPOTPGUIFQSFWJPVTFWFOU4FMG reports of this type cannot be taken as veridical, CVUUIFZDBODPNQMFNFOUPUIFSJOEJDFT5IF scale was predictive of target mechanism condiUJPOTJOUIFQSFWJPVTTUVEZ +VTMJOFUBM öôõø 5IVT XFFYQFDUFEUIF.FD4DBMFJUFNTUPCF predictive of the target mechanism conditions in UIFQSFTFOUTUVEZBMTP 1TZDIPQIZTJPMPHZ"MUIPVHIQTZDIPQIZTJologic reactions are not related to emotions in a one-to-one fashion (for a review, see Larsen, #FSOUTPO 1PFIMNBOO *UP $BDJPQQP öôôü

it appears feasible to link specific indices to CSPBEEJNFOTJPOTPGBSPVTBMBOEWBMFODF5IVT based on the assumptions that skin conductance level is a reliable measure of autonomic BSPVTBM "OESFBTTJ öôôû BOEUIBUCSBJOTUFN reflexes would arouse surprise (an emotion with BIJHIBSPVTBMMFWFM3VTTFMM õýüô

XFFYQFDUFE that the brain stem reflex condition would produce higher levels of skin conductance than would the contagion condition, which was expected to arouse sadness (an emotion with a MPXFSBSPVTBM 'VSUIFSNPSF CBTFEPOUIFBT-

EXPERIMENTS 1–4 METHOD Participants

4JYUZQBSUJDJQBOUT öýNFOBOE÷õXPNFO BHFõýmùü years, MƎƎöúö SDƎƎûû UPPLQBSUJOUIFTUVEZBTB whole and were either paid or given course credits for UIFJSBOPOZNPVTBOEWPMVOUBSZQBSUJDJQBUJPO.PTU participants were students, who were recruited by NFBOTPGQPTUFSTUISPVHIPVU6QQTBMB6OJWFSTJUZ 4JYUZUISFFQFSDFOUPGUIFQBSUJDJQBOUTQMBZFEBUMFBTU one musical instrument, and 55% had received some NVTJDFEVDBUJPO5IFZXFSFSBOEPNMZEJTUSJCVUFE across the four experiments, with the only provision that there must be an equal number of participants in each experiment: Experiment 1 featured 7 men and 8 women (age 19–44 years, MƎƎöúû SDƎƎû÷

&YQFSJment 2 featured 7 men and 8 women (age 20–36 years, MƎƎöùô SDƎƎøô

&YQFSJNFOU÷GFBUVSFEüNFO and 7 women (age 19–58 years, MƎƎöýý SDƎƎõöõ

and Experiment 4 featured 7 men and 8 women (age 19–27 years, MƎƎööý SDƎƎöö 4UBUJTUJDBMUFTUT BPOFXBZ"/07" CFUXFFOTVCKFDUT GPSUIFBHFWBSJBCMF$PDISBOTOPOQBSBNFUSJDQ test for kƎƎöFYQFSJNFOUBMUSFBUNFOUT CJOBSZDPEFE in the cases of gender, musical instrument, and music FEVDBUJPO SFWFBMFEOPTJHOJêDBOUEJďFSFODFCFUXFFO the four listener samples with regard to gender, age, experience of playing a musical instrument, or music FEVDBUJPO Design

5IFEFTJHOXBTTJNJMBSBDSPTTUIFGPVSFYQFSJNFOUT 8FVTFEBXJUIJOTVCKFDUEFTJHOJODMVEJOHUBSHFU mechanism as independent variable (five levels: brain

t JUSLIN ET AL .


stem reflex, contagion, episodic memory, musical exQFDUBODZ BOEOFVUSBMDPOEJUJPO BOETFMGSFQPSUFE GFFMJOH õùTDBMFT

NFDIBOJTNJNQSFTTJPOT .FD 4DBMF

GBDJBMFYQSFTTJPO [ZHPNBUJDVTBOEDPSSVHBUPS NVTDMFT

BOEBVUPOPNJDBDUJWJUZ TLJODPOEVDUBODF MFWFM BTEFQFOEFOUWBSJBCMFT"OBMZTFTPGGBDJBMBOE autonomic measures also included an additional CBTFMJOFDPOEJUJPO5IFPOMZEJGGFSFODFCFUXFFO the experiments was that different music excerpts were used to represent the four mechanisms in each FYQFSJNFOU8FEFDJEFEUPDPOEVDUGPVSTFQBSBUF experiments as opposed to a single experiment including four blocks because the latter option would require a 3-hr experiment, thereby increasing the risk PGGBUJHVFFďFDUT

attack, frequency spectrum, pitch, and key clarity— BTFTUJNBUFEVTJOHUIF.VTJD*OGPSNBUJPO3FUSJFWBM UPPMCPY -BSUJMMPU 5PJWJBJOFO &FSPMB öôôü *O the following, we list all pieces, grouped according UPUBSHFUNFDIBOJTN BR AIN STEM REFLE X.

The brain stem reflex mechanism is thought to be activated by extreme features such as high sound level, quick attack, and sharp timbre, which occur locally and cannot be predicted from the syntactiDBMTUSVDUVSFPGUIFNVTJD5IJTNFDIBOJTNXBTUIVT targeted by selecting the following pieces of music NFBOMFOHUIøüùT &YQFSJNFOUõ4ZNQIPOZ/PöJO%NBKPS fourth movement (Allegro con spirito

DPNQPTFECZ+PIBOOFT#SBINTJOõüûû QFSGPSNFE CZ#FSMJOFS1IJMIBSNPOJLFS DPOEVDUFECZ )FSCFSUWPO,BSBKBO "GUFSCVTZTPVOEJOHCVU quiet strings, a loud section breaks in suddenly JOCBSö÷XJUIUIFGVMMPSDIFTUSB MFOHUIúúT &YQFSJNFOUö4ZNQIPOZ/Põô 'JSTUNPWFment (Adagio

DPNQPTFECZ(VTUBW.BIMFSJO õýõô QFSGPSNFECZ#PVSOFNPVUI4ZNQIPOZ 0SDIFTUSB DPOEVDUFECZ4JNPO3BUUMF 5IF excerpt features a moment where the restatement of the theme culminates in a shattering EJTTPOBODF BOPSHBOMJLFDIPSE MFOHUIúôT Experiment 3: The Firebird,i*OGFSOBM%BODF PG"MM,BTIDIFJT4VCKFDUT uTFDUJPOPGBCBMMFU and orchestral concert work composed by Igor 4USBWJOTLZJOõýõô QFSGPSNFECZUIF#FSMJO3BEJP4ZNQIPOZ0SDIFTUSB DPOEVDUFECZ-PSJO

Musical Material

4JYUFFOQJFDFTPGNVTJD GPVSJOFBDIFYQFSJNFOU were selected for inclusion in the study because they featured information deemed relevant for the activaUJPOPGFBDIUBSHFUNFDIBOJTN8FVTFEGBJSMZTIPSU musical excerpts (MƎƎûöT GPSUXPSFBTPOT'JSTU XFXBOUFEUIF SFUSPTQFDUJWF TFMGSFQPSUTUPSFëFDU local events in the music, which prevented the use of MPOHFSFYDFSQUT4FDPOE FNPUJPOTDPVMEDIBOHFSBQidly, which means that use of longer excerpts might produce series of emotions, which would overcomQMJDBUFUIFTUBUJTUJDBMBOBMZTFT5IFQJFDFTXFSFFEJUFE with regard to musical form, in order to preserve the JOUFHSJUZPGUIFQJFDFT"TBSFTVMU UIFFYDFSQUTXFSF OPUFYBDUMZJEFOUJDBMJOMFOHUI Table 1 presents overall correlations between the target mechanism conditions and commonly analyzed musical features—tempo, dynamics, tone

TABLE 1. Correlations Between Target Mechanism Conditions and Musical Features for Excerpts, Experiments 1–4 (N = 16) Musical feature Mechanism

Tempo

Dynamics

Attack

Spectrum

Pitch

Key clarity

Brain stem

.09

.68

–.32

.33

.08

.09

Contagion

.11

–.28

.73

–.15

–.11

.05

Memory

–.07

–.15

–.39

.13

.03

.26

Expectancy

–.13

–.25

–.01

–.31

–.01

–.39

Note. Tempo = mean tempo in beats per minute; dynamics = root mean square value of the amplitude; attack = mean attack time in seconds; spectrum = centroid of the frequency spectrum in hertz; pitch = mean pitch in hertz; and key clarity = arbitrary units (higher number denotes higher key clarity). Values show the point-biserial correlations (rpb) between the target mechanism condition (coded dichotomously) and musical feature (coded continuously) in Experiments 1–4. For a more detailed description of how each musical feature is computed, see Lartillot et al. (2008).



.BB[FM 5IFFYDFSQUCFHJOTXJUIBMPVEESVN and brass chord, which is repeated intermitUFOUMZêWFUJNFT MFOHUI÷ôT &YQFSJNFOUø4ZNQIPOZ/PýøJO(NBKPS 4FDPOENPWFNFOU Andante

XSJUUFOCZ'SBO[ +PTFQI)BZEOJOõýûõ QFSGPSNFECZ8JFOFS Philharmonic Orchestra, conducted by LeoOBSE#FSOTUFJO "GPSUFLFUUMFESVNTUSPLFPDDVSTBUUIFFOEPGUIF QJBOJTTJNP SFQFBUPGUIF êSTUTFDUJPO MFOHUI÷üT 4QFDJBMDBSFXBTUBLFOUPDBMJCSBUFUIFTPVOEMFWFM of the target events, but pretesting indicated that the peak sound level did not quite have to reach the levels used in research on the acoustic startle response -FWFOTPO öôôû UPQSPEVDFBSFMJBCMFFďFDUPOUIF MJTUFOFS"QFBLTPVOEMFWFMPGûùE#"XBTTVĐDJFOU For these excerpts, we expected listeners to react primarily with surprise and autonomic arousal to the sudden extreme events, consistent with an early reaction that occurs before any elaborate classification of UIFTPVOEFWFOUIBTUBLFOQMBDF 4JNPOT õýýú 5IF target events are reflected in the correlations involvJOHEZOBNJDTBOETQFDUSVN TFF5BCMFõ CONTAGION.

The contagion mechanism is thought to be activated by a particularly moving emotional expression in the music, and it is assumed that the effect is strengthened by a voicelike lead part, either a real voice or an JOTUSVNFOUSFNJOJTDFOUPGUIFIVNBOWPJDF*UIBT been proposed that the cello and the violin are the closest-sounding instruments to the human voice, in terms of register, tone attack, timbre, and vibrato GPSFNQJSJDBMTVQQPSU TFF.PSFT öôôý

BOEQSFvious results indicate that performances with a sad expression are perceived as particularly expressive +VTMJO õýýû 'JHVSF÷ 8FUBSHFUFEUIJTNFDIBOJTN by selecting the following pieces, which include a TBEFYQSFTTJPO +VTMJO-BVLLB öôô÷ 5BCMFû BOE TPMPWPJDFTQFSGPSNFEPOUIFDFMMP FYDFSQUTõBOE÷ PSUIFWJPMJO FYDFSQUTöBOEø NFBOMFOHUIõôúT Experiment 1: “Prayer,” From Jewish Life/P õ XSJUUFOCZ&SOFTU#MPDIJOõýöø QFSGPSNFE CZ+BZ#BDBM VTJOHUIF7JFOOB4ZNQIPOJD-JCSBSZ "MZSJDBMBOEFYQSFTTJWFQJFDF DPNQPTFE for cello and piano, marked andante moderato BOEFYQSFTTJOHBOJOXBSEGFFMJOHPGTBEOFTT This piece is the same as the contagion version GFBUVSFEJO+VTMJOFUBM öôõø MFOHUIùôT Experiment 2: Concerto for Two Violins in A minor 0Q÷/Pü ** i-BSHIFUUPF4QJSJUPTP u

XSJUUFOCZ"OUPOJP7JWBMEJJOõûõõ QFSGPSNFE CZ"DDBEFNJB;JMJOJBOB GFBUVSJOH'SBOUJØFL 'JHVSB "GUFSBCSJFGJOUSPEVDUJPO BTPMPMJOF played with heavy vibrato on the violin begins and is soon joined by a second violin in a movJOH WPDBMMJLFEVFU MFOHUIõööT Experiment 3: Vocalise,0Q÷ø /Põø XSJUUFO CZ4FSHFJ3BDINBOJOPďJOõýõö QFSGPSNFECZ .JTDIB.BJTLZBOE-JMZ.BJTLZ 0SJHJOBMMZ XSJUUFOGPSWPJDF XJUIPVUMZSJDT XJUIQJBOP accompaniment, this version was arranged for DFMMPBOEQJBOP5IF NPEBM &NJOPSUPOBMJUZ the chromatic motion of the harmony, and the NFMPEZBMMTVHHFTUNFMBODIPMZ MFOHUIõöúT Experiment 4: Heart’s Ease 5ISFFMZSJDT/P õ

XSJUUFOCZ'SBOL#SJEHFJOõýöõ XPSLõúõB QFSGPSNFECZ+BZ#BDBM XJUI7JFOOB4ZNQIPOJD-JCSBSZ "TIPSU TMPX BOEDPOUFNQMBtive piece, written for violin and piano, marked andante tranquillo MFOHUIõöüT For these excerpts, we expected a matching or NJNJDSZSFTQPOTFJOMJTUFOFST*OPUIFSXPSET XF expected that they would show an empathic reaction UPUIFFNPUJPOBMFYQSFTTJPOPGUIFNVTJD#FDBVTF the music featured a sad expression, we expected it UPBSPVTFNBJOMZTBEOFTTJOUIFMJTUFOFST5IFTBE expression of the excerpts is reflected, for instance, in the minor mode, the slow tone attacks, and the TVCUMFEZOBNJDT TFF5BCMFõDG)FWOFS õý÷ù õý÷ú EPISODIC MEMORY.

The episodic memory mechanism is thought to be activated by salient melodic themes, which are associated with emotionally charged events that the MJTUFOFS SFNFNCFST 5P FWPLF NVTJDBTTPDJBUFE episodic memories without having to encode them during this experiment, we selected four pieces likely to be highly familiar to the present listener sample because of their frequent occurrence in social events FH DFSFNPOJFT JO4XFEFO NFBOMFOHUIúôT &YQFSJNFOUõi8FEEJOH.BSDIuJO$NBjor, from Suite of Incidental Music0QúõUP 8JMMJBN4IBLFTQFBSFTQMBZA Midsummer Night’s Dream,XSJUUFOCZ'FMJY.FOEFMTTPIO #BSUIPMEZJOõüøö QFSGPSNFECZ.BSHBSFUB -JOEHSFO 5IJTJTUIFNPTUDPNNPOMZVTFE XFEEJOHNBSDIJO4XFEFO UZQJDBMMZQFSGPSNFE POBDIVSDIQJQFPSHBO MFOHUIùúT Experiment 2: Sommar, Sommar, Sommar, XSJUUFOCZ4UFO$BSMCFSHJOõýùö QFSGPSNFE CZªLF+FMWJOHT0SLFTUFS 'PSNPSFUIBOùô

t JUSLIN ET AL .


years, this piece has been the signature song of a IJHIMZQPQVMBSSBEJPQSPHSBNJO4XFEFO Sommar (Summer

XIJDIJTCSPBEDBTUEBJMZEVSJOH UIFTVNNFS MFOHUI÷ýT Experiment 3: Den Blomstertid Nu Kommer, written by Israel Kolmodin in 1894 (performed CZ"EPMG'SFESJLT#BDILÕS 5IJTJTPOFPGUIF NPTUXFMMLOPXOQTBMNTJO4XFEFO XIJDIJT UZQJDBMMZTVOHEVSJOHHSBEVBUJPOT MFOHUIüõT Experiment 4: Studentsången,XSJUUFOCZ)FSNBO4ÅUIFSCFSH MZSJDT BOE1SJOT(VTUBG NVTJD JOõüùö QFSGPSNFECZ$BQFMMB$BOUJDB This piece, marked “marsch,” is often sung by choirs at joyous events that celebrate graduation JOUIFMBUFTQSJOH MFOHUIúøT The emotions aroused by this mechanism are believed to reflect the emotional tone of the memory FWPLFE5IFFYDFSQUTDIPTFOXFSFUIPVHIUUPCFBTsociated with both nostalgic and happy memories of fun, holidays, relaxation, spring and summer, graduBUJPO BOEGFTUJWJUJFT'PSUIFTFFYDFSQUT XFFYQFDUFE listeners to experience mainly nostalgia and happiness because of the memories evoked by the familiar NVTJD MUSICAL E XPECTANC Y.

The musical expectancy mechanism is believed to be activated by unexpected melodic, harmonic, or SIZUINJDTFRVFODFT )VSPO öôôú.FZFS õýùú Thus, in order to activate this mechanism, and more specifically to confound listeners’ musical expectations, we selected the following pieces of music (mean MFOHUIûùT Experiment 1: The Symphony of Psalms, II, “ExQFDUBOT&YQFDUBWJ%PNJOVN uDPNQPTFECZ *HPS4USBWJOTLZJOõý÷ô QFSGPSNFECZUIF3VTTJBO4UBUF"DBEFNZ0SDIFTUSBBOE$IPJS DPOEVDUFECZ*HPS.BSLFWJUDI 5IJTFYDFSQU GSPN 4USBWJOTLZTOFPDMBTTJDBMQFSJPE DPOTJTUTPGB fugue theme that begins with a four-note cell in UIFPCPFJOUIFêSTUNFBTVSF MFOHUIüøT Experiment 2: Lyric Suite, Three Pieces for String Orchestra,1BSU*** i"EBHJP"QQBTTJPOBUP uXSJUUFOCZ"MCBO#FSHJOõýöú QFSGPSNFE CZ8JFOFS1IJMIBSNPOJLFS DPOEVDUFECZ$MBVEJP"CCBEP 5IFFYDFSQUGPMMPXT CVUEPFTOPU TUSJDUMZBEIFSFUP "SOPME4DIPFOCFSHTõöUPOF practice, which abandons harmonically conDFJWFEUPOBMJUZ MFOHUIûôT Experiment 3: Three Pieces for Orchestra,0Q ú i1SBFMVEJVN uXSJUUFOCZ"MCBO#FSHJOõýõù

QFSGPSNFECZ4UFďFO'BIM XJUI7JFOOB4ZNQIPOJD-JCSBSZ "OJNQSFTTJPOJTUJDQSFMVEF which begins in vagueness with unpitched QFSDVTTJPOTPVOET8IFOUIFLFUUMFESVNTFOUFS in the third measure, indeterminate pitch is reQMBDFECZVODFSUBJOQJUDI MFOHUIúôT Experiment 4: Rite of Spring,1BSUõ i-FT"VHVSFT1SJOUBOJFST uXSJUUFOCZ*HPS4USBWJOTLZ JOõýõ÷ QFSGPSNFECZ#FSMJOFS1IJMIBSNPOJLFS DPOEVDUFECZ)FSCFSUWPO,BSBKBO "OBWBOU garde piece characterized by a repetitive stamping chord in the horns and strings, based on &ëBUTVQFSJNQPTFEPOBOUSJBEPG& (TIBSQ BOE# MFOHUIüúT "OBMZTFTDPOêSNFEUIBUUIFTFFYDFSQUTJOWPMWFEB MPXFSEFHSFFPGLFZDMBSJUZ 5BCMFõ UIBOUIFFYDFSQUT JODMVEFEJOUIFPUIFSDPOEJUJPOT'PSUIFTFFYDFSQUT we expected listeners to respond mainly with anxiety to the unresolved uncertainty created by the syntactic sequences of the musical structure, as theorized by .FZFS õýùú TFFQöûTFFBMTP)VSPO öôôú In addition to these pieces, we selected an unLOPXOQJFDFUIBUTFSWFEBTBOFVUSBMDPOEJUJPO5IF piece, titled “minimalist music,” was composed by the alias Mihangeliago and downloaded from the *OUFSOFU8FDIPTFJUCFDBVTFJUEJEOPUGFBUVSFBOZ type of information deemed necessary to arouse an emotion through one of the mechanisms in the #3&$7&."GSBNFXPSL1JMPUUFTUJOHDPOêSNFE UIBUUIFQJFDFXBTiFNPUJPOBMMZJODPNQFUFOUu*U could be characterized as slow, soft, and monotoOPVT5IFOFVUSBMFYDFSQUXBTUIFTBNFJOBMMGPVS FYQFSJNFOUT MFOHUIùýT Experiential Measures

8FNFBTVSFEUIFTVCKFDUJWFGFFMJOHDPNQPOFOUPGUIF aroused emotions in listeners by means of 15 adjective TDBMFT XIJDIIBWFCFFOVTFEBU6QQTBMB6OJWFSTJUZ specifically to measure emotions to music (listed in 5BCMFöMBUFSJOUIJTBSUJDMF 5IFTDBMFTSFQSFTFOUB compromise between the response formats currently VTFEJOUIFNVTJDFNPUJPOêFME ;FOUOFS&FSPMB öôõô CFDBVTFUIFTFMFDUFEUFSNTJODMVEFiCBTJDu emotions characteristic of discrete emotion theories *[BSE õýûû

DPWFSBMMGPVSRVBESBOUTPGBDJSDVNQMFY NPEFMJOUFSNTPGWBMFODFBOEBSPVTBM 3VTTFMM õýüô

and feature possibly more music-related terms such as nostalgia, expectancy, and awe +VTMJO-BVLLB öôôø 5IFTFMFDUFEUFSNTSPVHIMZDPWFSUIFOJOF GBDUPSTPG(&.4ý QSPQPTFECZ;FOUOFS (SBOEKFBO BOE4DIFSFS öôôü CVUCFDBVTFUIFSFJTOPWBMJEBUFE



WFSTJPOPG(&.4ýJO4XFEJTI BOEUIFTDBMFMBDLT placed on the palmar surface of the nondominant terms that were needed in this study, such as surhand, at the thenar and the hypothenar eminences prise XFEFDJEFEUPVTFBDVTUPNJ[FETDBMF 5IF 'PXMFTFUBM õýüõ 4LJODPOEVDUBODFXBTSFDPSEFE list features the emotions most commonly reported in in microsiemens (ΩNIP QSFWJPVTTUVEJFT +VTMJO-BVLLB öôôø+VTMJOFUBM #JQPMBSGBDJBMFMFDUSPNZPHSBQIZ &.( SFDPSEöôõõ8FMMT)BLBOFO õýýõ;FOUOFSFUBM öôôü *O ings were made from the left corrugator and zygomataddition to 12 discrete emotions, listeners also rated ic muscle regions in accordance with Fridlund and liking and familiarity for each version, and whether $BDJPQQPT õýüú HVJEFMJOFT#FGPSFBUUBDIJOHUIFø UIFZFYQFSJFODFEBOZiDIJMMTu FH ,POFčOJ 8BOJD NNNJOJBUVSFTVSGBDF"H"H$MFMFDUSPEFT êMMFEXJUI #SPXO öôôûEFêOFEBTQJMPFSFDUJPO PSgåshud &.(HFM (&-õôô #JPQBD4ZTUFNT

XFDMFBOTFE JO4XFEJTIFWFSZEBZUFSNJOPMPHZ "MMSBUJOHTXFSF the participant’s skin to reduce interelectrode impedmade on a unidimensional scale from 0 (not at all BODF"MMJNQFEBODFXBTSFEVDFEUPMFTTUIBOõôLΩ to 4 (a lot

FYDFQUGPSDIJMMT XIJDIXFSFSFQPSUFEJO 'SJEMVOE$BDJPQQP õýüú 5IFFMFDUSPEFTXFSF BEJDIPUPNPVTGBTIJPO DPOOFDUFEUPUIF&.(õôô$BNQMJêFSNPEVMFXJUI In addition to reporting their feelings, particiMPX BOEIJHIQBTTêMUFSTTFUBUùôô)[BOEõô)[ QBOUTêMMFEPVUBTFDPOESFTQPOTFTDBMF .FD4DBMF SFTQFDUJWFMZ BOEOPUDIêMUFSTTFUBUùô)[XFSFVTFE GPSFBDINVTJDBMFYDFSQU TFF"QQFOEJY 5IJTTDBMF UPEJNJOJTIJOUFSGFSFODFXJUIUIFFMFDUSJDNBJOT5IF purported to capture the mechanisms that had ocTBNQMJOHSBUFXBTTFUBUöôôô)['BDJBM&.(XBT curred and consisted of eight simple questions, each measured in microvolts (μ7 BOEBOBMZ[FEVTJOHUIF UBSHFUJOHPOFPGUIFNFDIBOJTNTJOUIF#3&$7&. SPPUNFBOTRVBSF5IFSBX&.(EBUBXFSFêMUFSFE GSBNFXPSL +VTMJOFUBM öôõô QMVTBQQSBJTBMCSBJO using a finite impulse response filter between 28 and stem reflex, rhythmic entrainment, episodic memory, öùô)[ JOPSEFSUPJODSFBTFTJHOBMUPOPJTFSBUJP evaluative conditioning, visual imagery, contagion, .FBOWBMVFTGPSTLJODPOEVDUBODFMFWFMBOE&.( NVTJDBM FYQFDUBODZ BOE DPHOJUJWF BQQSBJTBM "MM [ZHPNBUJDVTBOEDPSSVHBUPSNVTDMFT XFSFDBMDVitems were rated on a scale from 0 (not at all UPø a MBUFEGPSCBTFMJOFBOEFYQFSJNFOUBMDPOEJUJPOT 5IF lot "GPMMPXVQJUFNJO.FD4DBMF OPUTIPXOJOUIF baseline recordings were obtained before the listen"QQFOEJY

XIJDIBQQFBSFEPOUIFDPNQVUFSTDSFFO JOHUFTUEVSJOHSFMBYBUJPOVOEFSTJMFOUDPOEJUJPOT only if a stimulus happened to evoke a memory, asked %VSJOHUIFMJTUFOJOHUFTU UIFSFXBTBCSFBLCFUXFFO whether the memory was mainly positive, mainly musical excerpts to allow levels to return to baseline OFHBUJWF PSBDPNCJOBUJPOPGCPUI5IJTJUFNXBT CFGPSFUIFOFYUTUJNVMVT featured to enable us to check whether the valence of evoked memories was consistent with the valence Procedure PGFNPUJPOTSFQPSUFEJOUIFNFNPSZDPOEJUJPO 8IFOQBSUJDJQBOUTBSSJWFEBUUIFMBCPSBUPSZ UIFZXFSF seated in a comfortable armchair and received the folMPXJOHJOTUSVDUJPOT USBOTMBUFEGSPN4XFEJTI

XIJDI Psychophysiology: Facial Expression were the same in all four experiments: and Autonomic Activity

To increase the validity of the measurement of emoUJPO XFBMTPNFBTVSFEQIZTJPMPHJDJOEJDFT5IFHPBM was to obtain evidence of an emotional response, in order to distinguish felt emotions from mere perceptionPGFNPUJPOT*OUIFGPSNFSDBTF XFXPVMEFYQFDU to discover some changes in physiologic indices (as QBSUPGBOFNPUJPOBMSFBDUJPO

XIFSFBTJOUIFMBUter case there would be no reason to expect such DIBOHFT.PSFPWFS XFBJNFEUPUFTUQSFEJDUJPOT XJUISFHBSEUPTQFDJêDDPOUSBTUTCFUXFFODPOEJUJPOT Psychophysiologic indices were obtained usJOHUIF#*01"$.1õùô4ZTUFN #JPQBD4ZTUFNT 4BOUB#BSCBSB $" BOEUIF"DR,OPXMFEHFWFSTJPO øõTPGUXBSF4LJODPOEVDUBODFMFWFMXBTNFBTVSFE VTJOHUIF(43õôô$&MFDUSPEFSNBM"DUJWJUZ"NQMJfier module and EL507 disposable snap electrodes

8FMDPNFUPUIFNVTJDMBCPSBUPSZ:PVXJMMTPPO MJTUFOUPBTFMFDUJPOPGTIPSUQJFDFTPGNVTJD "GUFSFBDIQJFDFXFXBOUZPVUPEFTDSJCFZPVS FYQFSJFODFPGUIFNVTJD5IJTTIPVMECFEPOF in two ways: first we want you to describe your GFFMJOHTEVSJOHUIFNVTJDPOBDPNQVUFSTDSFFO 5IFTDSFFODPOTJTUTPGUXFMWFFNPUJPOTYour task is to rate how much of each emotion you felt on a scale from 0 (“not at all”) to 4 (“a lot”).:PV also report whether you experienced “chills,” as well as how much you liked the music and how familiarZPVXFSFXJUIJU5IFOXFXBOUZPVUP attend to a second screen, which features eight questions concerning other aspects of your muTJDFYQFSJFODF:PVXJMMBMTPCFêUUFEXJUITPNF electrodes so that we can conduct physiological

t JUSLIN ET AL .


NFBTVSFNFOUT5IFTFFMFDUSPEFTBSFDPNQMFUFMZ harmless and do not emit strong radiation or FMFDUSJDJUZ)PXFWFS JOPSEFSUPPCUBJOBTBDcurate measurements as possible, it is important that you don’t touch any of the electrodes durJOHUIFFYQFSJNFOU8BUDIFTBOESJOHTIBWF to be removed and your cell phone must be TXJUDIFEPď'JSTU ZPVXJMMCFBTLFEUPSFMBYGPS BXIJMFEVSJOHTJMFODF5IFO UIFBDUVBMMJTUFOJOHUFTUCFHJOT8IFOUIFQMBZCBDLPGBQJFDF of music ends, there will be a brief intermission before the next piece begins, to give you time to êMMPVUUIFUXPSFTQPOTFTIFFUT5IFO ZPVXJMM relax again for a while before the next piece beHJOT/PUFUIBUBOZFNPUJPOZPVNBZFYQFSJFODF during listening need not correspond to the muTJDTFNPUJPOBMFYQSFTTJPO5IBUJT ZPVTIPVME rate your own emotions, not what the music exQSFTTFT"GUFSUIFFYQFSJNFOUZPVXJMMCFBTLFE UPSFTQPOEUPTPNFCBDLHSPVOERVFTUJPOT Participants were tested individually in a soundproofed room and listened to the music through a QBJSPGIJHIRVBMJUZMPVETQFBLFST %BMJ*LPOú.,ö %BMJ"4 /SJOHFS %FONBSL 4UJNVMVTBENJOJTtration and data collection was handled using the .FEJB-BCTPGUXBSF5IFTPVOEMFWFMXBTQSFTFUUP a comfortable level, which was held constant across MJTUFOFST4UJNVMVTPSEFSXBTSBOEPNJ[FEGPSFBDI participant, whereas the order of rating scales was LFQUDPOTUBOUBDSPTTQBSUJDJQBOUT"GUFSUIFMJTUFOJOH test, the participants filled out a short questionnaire XJUISFHBSEUPWBSJPVTCBDLHSPVOEWBSJBCMFT FH BHF HFOEFS NVTJDFEVDBUJPO 5IFQBSUJDJQBOUTXFSF not fully debriefed about the purpose of the experiment until all had been tested, to prevent confoundJOHFďFDUT /FBMF-JFCFSU õýüú "OFYQFSJNFOUBM TFTTJPOMBTUFEBCPVUùôNJOVUFT

RESULTS To give a more concise presentation, we report the EBUBGSPNUIFGPVSFYQFSJNFOUTJOBKPJOUTFDUJPO First, we present separate analyses for every measure in each experiment, then we report combined analyses, which capture broader trends across the FYQFSJNFOUT Separate Analyses: Experiments 1–4 EMOTION R ATINGS.

The most important data concern the listeners’ ratJOHTPGGFMUFNPUJPOTPOUIFõùSBUJOHTDBMFT5PFWBMV-

ate the effect of target mechanism on listeners’ selfSFQPSUT XFDPOEVDUFEBO"/07"XJUINFDIBOJTN as within-subject factor (five levels: neutral, brain stem reflex, contagion, episodic memory, musical FYQFDUBODZ POFBDITDBMF8FVTFEBOFYQFSJNFOU XJTF#POGFSSPOJDPSSFDUJPOGPSNVMUJQMFUFTUT nƎƎõù from αƎƎôùUPαƎƎôô÷÷ Table 2 presents a summary of the results for &YQFSJNFOUTõmø"TDBOCFTFFO UIFêWFTDBMFTJOWPMWFEJOPVSQSFEJDUJPOT JF IBQQJOFTTmFMBUJPO sadness– melancholy, surprise– astonishment, OPTUBMHJBmMPOHJOH BOEBOYJFUZmOFSWPVTOFTT TIPXFE significant effects in all instances except one (95%, nƎƎöô BOYJFUZmOFSWPVTOFTTJO&YQFSJNFOUõ5IF rightmost column of Tables 2 presents effect sizes, in terms of η2"TDBOCFTFFO FďFDUTJ[FTGPSUIFQSFdicted emotions ranged from “moderate” (η2ƎƌƎöù to “strong” (η2ƎƌƎúø BDDPSEJOHUP'FSHVTPOT öôôý HVJEFMJOFTGPSJOUFSQSFUBUJPO4JNJMBSMZ UIFOPOEJTcrete scales emotion intensity, liking, and familiarity showed significant effects of mechanism with one exception: intensity fell short of significance in ExQFSJNFOU÷ 5BCMFö "MTPOPUJDFUIFMBSHFFďFDUTPG target mechanism on the familiarity scale (mean η2 BDSPTTFYQFSJNFOUTƎƎüõû Careful inspection of Table 2 suggests that a few additional emotion scales included in the self-report JOTUSVNFOUTIPXFETJHOJêDBOUFďFDUTJOUIF"/07"T /PUFIPXFWFS UIBUUIFFďFDUTPGUIFTFTFWFOTDBMFT were smaller overall (mean η2 across scales and exQFSJNFOUTƎƎöûö UIBOUIPTFGPSUIFêWFTDBMFTPGUIF QSFEJDUFEFNPUJPOT øúô 5IFFďFDUTXFSFBMTPJODPOTJTUFOU5IVT GPSJOTUBODF JOUFSFTUmFYQFDUBODZ and anger–irritation showed no significant effects, and disgust–contempt and admiration–awe showed a significant effect in some experiments, but not in PUIFST5IFPOMZSFDVSSJOHUFOEFODJFTXFSFUIBUDBMNm DPOUFOUNFOUBOEMPWFmUFOEFSOFTTXFSF NPTUMZ TJHOJêDBOUBDSPTTFYQFSJNFOUT3BUJOHTPOUIFGPSNFS scale were inversely correlated with ratings of anxiety– nervousness (rƎƎ¢ùö

XIFSFBTSBUJOHTPOUIFMBUUFS scale were correlated with ratings of nostalgia–longing (rƎƎúô BOEMJLJOH rƎƎùý

SFTQFDUJWFMZ r computed across experiments, NƎƎúô BMMpTƎƎôù #FDBVTFPGUIFHFOFSBMMZTNBMMFSBOENPSFJODPOsistent effects of the additional scales, and in order to give a more concise presentation of the data, we will henceforth focus on the five emotion scales involved JOPVSUIFPSFUJDBMQSFEJDUJPOT3FDBMMUIBUUIFQSFEMOTIONAL REACTIONS TO MUSIC t


TABLE 2. ANOVA for Listeners’ Emotion Ratings, Experiments 1–4 Mean square

F

pa

η2

7.880

9.746

.000005

.410

Sadness–melancholy

18.113

33.425

.000001

.705

Surprise–astonishment

Experiment 1 Happiness–elation

14.167

14.655

.000001

.511

Calm–contentment

3.847

3.068

.023497

.180

Interest–expectancy

2.487

2.467

.055225

.150

Nostalgia–longing

7.513

6.705

.000175

.324

Anxiety–nervousness

4.883

4.160

.005087

.229

Pride–confidence

3.333

3.341

.015964

.193

Anger–irritation

2.113

3.614

.010878

.205

Love–tenderness

9.513

9.455

.000007

.403

Disgust–contempt

0.987

1.791

.143531

.113

Admiration–awe

4.647

5.442

.000897

.280

Emotion intensity

4.413

4.650

.002597

.249

Liking

7.220

7.848

.000043

.359

33.353

71.326

.000001

.836

Happiness–elation

15.420

14.706

.000001

.512


19.913

18.002

.000001

.562


15.447

11.232

.000001

.445

Calm–contentment

11.420

9.821

.000004

.412


1.667

1.064

.383029

.071

Nostalgia–longing

15.113

10.914

.000001

.438


15.147

13.460

.000001

.490

4.920

5.655

.000678

.288

Familiarity Experiment 2

Pride–confidence Anger–irritation

2.533

2.778

.035465

.166

Love–tenderness

17.787

16.368

.000001

.539

Disgust–contempt

1.613

2.630

.043751

.158

Admiration–awe

4.753

3.667

.010097

.208

Emotion intensity

5.680

7.349

.000079

.344

9.153

8.995

.000011

.391

37.720

189.957

.000001

.931

Liking Familiarity

dictions concerned which emotions the four target mechanism conditions would evoke in listeners, for instance that the brain stem reflex condition would FWPLFQSFEPNJOBOUMZTVSQSJTFJOMJTUFOFST5PUFTU this, we conducted planned comparisons (tUFTUT CFUXFFOUIF QSFEJDUFE UBSHFUNFDIBOJTNBOEUIF other four conditions (the three mechanisms and the

OFVUSBMDPOEJUJPO

UPFYQMPSFXIFUIFSUIFQSFEJDUFE NFDIBOJTNSFDFJWFEUIFIJHIFTUNFBOSBUJOH5BCMF÷ TVNNBSJ[FTUIFSFTVMUT$BSFGVMJOTQFDUJPOPG5BCMF÷ TIPXTUIBUJOúûPGUIFüøDPOUSBTUT üôPGUIFDBTFT UIFSFTVMUTXFSFJOMJOFXJUIPVSQSFEJDUJPOTUIBUJT the rating on the scale was significantly higher for the QSFEJDUFENFDIBOJTNUIBOGPSUIFPUIFSDPOEJUJPO

t JUSLIN ET AL .


TABLE 2. (continued) Mean square

F

pa

η2

Happiness–elation

5.500

5.347

.001018

.276


5.647

6.452

.000242

.315


17.987

19.058

.000001

.576

Calm–contentment

16.187

15.832

.000001

.530


2.067

1.307

.278505

.085

21.767

21.260

.000001

.603


5.753

6.742

.000168

.325

Pride–confidence

3.980

2.756

.036611

.164

Anger–irritation

0.713

1.450

.229694

.094

Love–tenderness

16.313

15.765

.000001

.530

Disgust–contempt

0.487

1.446

.231135

.094

Admiration–awe

3.767

2.830

.032938

.168

Emotion intensity

3.553

4.164

.005058

.229

Liking

6.447

5.355

.001007

.277

34.047

47.413

.000001

.772

Happiness–elation

8.313

11.314

.000001

.447


5.820

4.557

.002949

.246


19.567

31.011

.000001

.689

Calm–contentment

11.480

11.546

.000001

.452


1.700

2.380

.062469

.145

Nostalgia–longing

20.880

20.703

.000001

.597


11.020

14.180

.000001

.503

6.280

10.963

.000001

.439

Experiment 3

Nostalgia–longing

Familiarity Experiment 4

Pride–confidence Anger–irritation

3.447

4.405

.003629

.239

Love–tenderness

13.833

25.261

.000001

.643

Disgust–contempt

4.420

8.057

.000033

.365

Admiration–awe

4.247

5.798

.000562

.293

Emotion intensity

2.847

6.584

.000205

.320

6.120

7.948

.000038

.362

28.553

37.547

.000001

.728

Liking Familiarity Note. df = mechanism (4), error (56).

Bonferroni-corrected from α = .05 to α = .0033.

a

In the remaining cases, the rating for the predicted mechanism was either still the highest, although not TJHOJêDBOUMZTP õôDBTFT

PSUIFTFDPOEIJHIFTUPGUIF DPOEJUJPOT ÷DBTFT %FTDSJQUJWFTUBUJTUJDTBOENPSF FMBCPSBUJPOBSFQSPWJEFEJOUIFDPNCJOFEBOBMZTJT

The self-report data regarding chills (piloerecUJPO XFSFBOBMZ[FETFQBSBUFMZCFDBVTFPGUIFEJDIPUPNPVTOBUVSFPGUIFEBUB5PFWBMVBUFUIFFďFDUT of target mechanism on the proportion of chills reported, we used Cochran’s Q test, which is a non-



5"#-& Summary of Planned Comparisons Between Predicted Target Mechanism and Remaining Conditions for Emotion Ratings, Experiments 1–4 Scale Happiness–elation



Nostalgia–longing


Contrast

Experiment 1

Experiment 2

Experiment 3

Experiment 4

M vs. B

.486a

.001*

.229

.002*

M vs. C

.006*

.001*

.068

.001*

M vs. E

.002*

.001*

.001*

.001*

M vs. N

.041*

.001*

.001*

.001*

C vs. B

.001*

.188

.004*

.003*

C vs. M

.001*

.001*

.006*

.039*

C vs. E

.003*

.905

.265

.002*

C vs. N

.001*

.001*

.004*

.006*

B vs. C

.001*

.001*

.001*

.001*

B vs. M

.001*

.001*

.001*

.001*

B vs. E

.001*

.001*

.001*

.003*

B vs. N

.001*

.005*

.001*

.001*

M vs. B

.006*

.001*

.001*

.001*

M vs. C

.556

.055

.116

.132

a

M vs. E

.032*

.001*

.001*

.001*

M vs. N

.001*

.001*

.001*

.001*

E vs. B

.103

.499

.849

.001*

E vs. C

.002*

.001*

.010*

.001*

E vs. M

.023*

.001*

.007*

.001*

E vs. N

.018*

.002*

.030*

.003*

a

Note. Data indicate p values. B = brain stem reflex; C = contagion; E = expectancy; M = memory; N = neutral. aContrasts where the predicted target mechanism did not receive the highest mean rating. *p < .05.

parametric test for three or more matched sets of frequencies or proportions where data are dichotoNPVT $POPWFS õýýý "GUFS#POGFSSPOJDPSSFDUJPO (αƎƎôô÷÷

UIFFďFDUPGUBSHFUNFDIBOJTNDPOEJUJPO was not significant in any of the four experiments (QTƎƎ÷ôômõøõ÷ BMMpTƎƎôôû 5IFPWFSBMMUSFOE was that most chills occurred in the brain stem reflex öû DPOEJUJPO GPMMPXFECZUIFDPOUBHJPO öô

NFNPSZ õö

FYQFDUBODZ õô

BOEOFVUSBM ö DPOEJUJPOT)PXFWFS HJWFOUIBUDIJMMTPDDVSSFESBSFMZ overall (MƎƎõø BOEEJEOPUSFMJBCMZEJTDSJNJOBUF between the experimental conditions, these data are OPUEJTDVTTFEGVSUIFS

listeners’ ratings of these items, we conducted one "/07"XJUINFDIBOJTNBTXJUIJOTVCKFDUGBDUPS (five levels: neutral, brain stem reflex, contagion, epiTPEJDNFNPSZ NVTJDBMFYQFDUBODZ GPSFBDIJUFN8F VTFEBOFYQFSJNFOUXJTF#POGFSSPOJDPSSFDUJPOGPS multiple tests (nƎƎü

GSPNαƎƎôùUPαƎƎôôúø5IF results of the analysis indicated significant effects of mechanism for all items in all four experiments (values of F ø Ǝùú ƎƎùõõmùýùõ BMMpTƎôôúø

FYDFQUWJTVBM JNBHFSZ TJHOJêDBOUJO&YQFSJNFOUTömø BOEDPHOJUJWF BQQSBJTBM TJHOJêDBOUPOMZJO&YQFSJNFOUõ 'VSUIFS SFTVMUTDPODFSOJOH.FD4DBMF JODMVEJOHUIFEJSFDUJPOT PGUIFFďFDUT BSFQSPWJEFEJOUIFDPNCJOFEBOBMZTJT

MECSC ALE.

PSYCHOPHYSIOLOGY.

The listeners also responded to eight items, which To evaluate the manipulation of target mechanism on UBSHFUFETQFDJêDNFDIBOJTNT TFFUIF"QQFOEJY 5P QTZDIPQIZTJPMPHZ XFDPOEVDUFEBO"/07"XJUI evaluate the effects of target mechanism condition on mechanism as within-subject factor (six levels: base-

t JUSLIN ET AL .


line, neutral, brain stem reflex, contagion, episodic NFNPSZ NVTJDBMFYQFDUBODZ POFBDIQIZTJPMPHJD NFBTVSF'PSFBDIMJTUFOFS UIFSBXEBUBPGFBDINFBsure were transformed into z residuals before analysis in order to minimize the impact of individual difGFSFODFTJOCBTFMJOFTCFUXFFOUIFMJTUFOFST5IFSFTVMUTGPS&YQFSJNFOUTõmøBSFQSFTFOUFEJO5BCMFø /PUFUIBUNFDIBOJTNZJFMEFETJHOJêDBOUFďFDUTPO all measures in all experiments, except skin conductance level in Experiment 2 and corrugator muscle BDUJWJUZJO&YQFSJNFOU÷*OTQFDUJPOPGUIFSJHIUNPTU column suggests that these effects were “small” to iNPEFSBUFuJOTJ[F 'FSHVTPO öôôý 5PUFTUPVSQSFdictions with respect to specific contrasts between the target mechanism conditions for these measures, we conducted planned comparisons (tUFTUT 5BCMFù

QSFTFOUTUIFSFTVMUT*UDBOCFTFFOUIBUPOMZùüPG the predictions received support, with better results GPS[ZHPNBUJDVT ûù BOEDPSSVHBUPS ûù BDUJWJUZ predictions than for skin conductance predictions öù 0OMZJOUXPPGUIFDPOUSBTUT IPXFWFS XBT the direction of the observed effect contrary to the QSFEJDUJPO%FTDSJQUJWFTUBUJTUJDTBSFQSPWJEFEJOUIF DPNCJOFEBOBMZTJT Combined Analyses, Experiments 1–4

#FDBVTFUIFSFXFSFOPTJHOJêDBOUEJďFSFODFTCFtween the four experimental groups with respect to age, gender, experience of playing an instrument, or music education (see MethodTFDUJPO

BOECFDBVTF there were no significant differences between the groups with regard to how they rated the common

TABLE 4. ANOVA for Psychophysiology, Experiments 1–4 Experiment

Measure

Mean square

F

p

η2

1

Skin conductance level

2.383

2.644

.030128*

.159

1

Zygomatic muscle activity

4.014

5.115

.000469*

.268

1

Corrugator muscle activity

3.539

4.324

.001740*

.236

2


1.823

1.937

.099117

.122

2


2.333

2.579

.033680*

.156

2


4.413

5.836

.000146*

.294

3


6.229

9.944

.000001*

.415

3


5.478

8.055

.000005*

.365

3


0.348

0.333

.891575

.023

4


6.518

10.757

.000001*

.435

4


0.001

2.620

.031404*

.158

4


2.884

3.334

.009294*

.192

Note. df = mechanism (5), error (70). *p < .05.

5"#-& Summary of Planned Comparisons for Psychophysiology, Experiments 1–4 Measure

Contrast

Experiment 1

Experiment 2

Experiment 3

Experiment 4


B vs. C

.225

.648

.001*

.716a

Zygomaticus muscle activity

M vs. C

.007*

.002*

.091

.002*


C vs. M

.001*

.001*

.962a

.038*

Note. Data indicate p values. B = brain stem reflex; C = contagion; M = memory. aContrasts where the predicted target mechanism did not receive the higher mean value. *p < .05.



NVTJDBMTUJNVMVT iOFVUSBMuQJFDF POBOZPGUIFõù SBUJOHTDBMFT BTJOEJDBUFECZPOFXBZ"/07"T CFtween groups, F ÷ Ǝùú ƎƎôôü÷mõýýù pTƎƎõ÷ýû

we thought it was justified to treat the participants as matched subjects and to combine data across exQFSJNFOUTGPSEFTDSJQUJWFBOEFYQMPSBUPSZQVSQPTFT EMOTION R ATINGS.

Table 6 shows correlations between emotion ratings and target mechanism conditions across Experiments õmø5IFTFDPSSFMBUJPOTDPOêSNUIBUUIFSFTVMUTXFSF mostly in line with the predictions but also highlight the problem with regard to a clear separation of emoUJPOT.PTUJNQPSUBOUMZ JUDBOCFTFFOUIBU DPOUSBSZ to our predictions, the contagion mechanism aroused nostalgia–longing, and the expectancy mechanism BSPVTFETBEOFTTmNFMBODIPMZ*OCPUIDBTFT IPXFWFS the correlation for the predicted emotion was significantly larger than the one for the nonpredicted emotion (pƎƎôù *UDBOGVSUIFSCFPCTFSWFEUIBU the “neutral” piece was negatively correlated with all FNPUJPOT1SFTFOUFEJOUIFMPXFSTFDUJPOPG5BCMFú BSFBMTPUIFSFTVMUTGPSJOUFOTJUZ MJLJOH BOEGBNJMJBSJUZ The former confirm that the “neutral” piece yielded a lower emotional intensity than the mechanism conditions, which on average aroused an intense emotional response (MƎƎöúú 5IFCSBJOTUFNSFëFYDPOEJUJPOT QSPEVDFEUIFNPTUJOUFOTFSFBDUJPOT8JUISFHBSEUP liking, it can be seen that the contagion pieces were CFTUMJLFEPWFSBMM JF EFTQJUFUIFGBDUUIFZUFOEFEUP

FWPLFTBEOFTTJOMJTUFOFST

BOEUIFFYQFDUBODZQJFDFT XFSFMFBTUMJLFEPWFSBMM FYDFQUGPSUIFOFVUSBMQJFDF Finally, the correlations for familiarity confirm that only the music in the memory conditions was highly GBNJMJBSUPUIFMJTUFOFST MECSC ALE.

5BCMFûTIPXTDPSSFMBUJPOTCFUXFFO.FD4DBMFJUFNT and target mechanism conditions across Experiments õmø0GQBSUJDVMBSJOUFSFTUBSFUIFDPSSFMBUJPOTUIBU are both statistically significant and positive in direcUJPO5IFSFTVMUTBSFNBJOMZBTDPVMECFFYQFDUFE if the items have predictive value regarding mechanisms: The brain stem reflex condition correlated most strongly with the brain stem item, the contagion condition correlated most strongly with the contagion item, the expectancy condition correlated most strongly with the expectancy item, and the memory condition correlated most strongly with the memory JUFN/PUFUIBUUIFOFVUSBMDPOEJUJPOXBTOFHBUJWFMZ correlated with all items, suggesting that this piece EJEOPUBDUJWBUFBOZPGUIFNFDIBOJTNT )PXFWFS 5BCMFûBMTPTIPXTTPNFDPSSFMBUJPOT JOBEEJUJPOUPUIPTFSFMBUFEUPUIFUBSHFUNFDIBOJTN /PUFJOQBSUJDVMBSUIBUUIFNFNPSZDPOEJUJPOTZJFMEed a larger number of significant correlations than the PUIFSDPOEJUJPOUZQFT5IFNFNPSZDPOEJUJPOTDPSSFlated not only with the episodic memory item but also with the entrainment, conditioning, visual imagery, and appraisal items, although the correlations of these

TABLE 6. Correlations Between Emotion Ratings and Target Mechanism Conditions Across Experiments 1–4 (N = 300) Condition Emotion scale

Neutral

Brain stem

Contagion

Expectancy

Happiness–elation

−.11

.07

−.09

−.31*

.43*


−.27*

.19*

−.23*

.01

−.20*

−.13

.44*

Memory


−.02

.59*

−.39*

Nostalgia–longing

−.30*

−.19*

.28*


−.11

.12

−.17

Intensity

−.41*

.20*

.13

−.04

.11

Liking

−.33*

.02

.32*

−.19*

.18

Familiarity

−.31*

−.16

−.22*

.84

−.15

−.23*

.44*

.42*

−.26*

Note. Values show point-biserial correlations (rpb) between listener’s emotion ratings (coded continuously) and target mechanism conditions (coded dichotomously). Correlations that are both statistically significant and positive in direction are shown in boldface. (Alpha level was Bonferroni corrected from α = .05 to α = .00125.) *p < .00125.

t JUSLIN ET AL .


5"#-& Correlations Between MecScale Items and Target Mechanism Conditions Across Experiments 1–4 (N = 300) Condition Scale item

Neutral

Brain stem

−.32*

Brain stem .75*

Expectancy

Memory

−.20*

−.01

−.23*

Entrainment

−.06

−.16

−.21*

.19*

Memory

−.25*

−.15

−.02

−.11

.54*

Conditioning

−.32*

−.16

−.01

.03

.45*

Visual imagery

−.30*

−.13

.07

−.01

.38*

Contagion

−.48*

.04

.38*

−.12

.18

Expectancy

−.05

Appraisal

−.19*

.23*

Contagion

.28* −.06

−.06

.36*

.05

−.53*

.32

.32*

Note. Values show point-biserial correlations (rpb) between ratings of MecScale items (coded continuously) and target mechanism conditions (coded dichotomously). Correlations that are both statistically significant and positive in direction are shown in boldface. Scale items are shown in the Appendix. (Alpha level was Bonferroni corrected from α = .05 to α = .00125.) *p < .00125.

were significantly smaller than the correlation of the episodic memory item (pƎƎôù

XJUIUIFFYDFQUJPO of the conditioning item (pƎƎôû 5IFGPMMPXVQJUFN to the episodic memory item (see MethodTFDUJPO showed that 76% of the memories were positive in nature, 0% were negative in nature, and 24% were a NJYUVSFPGCPUIQPTJUJWFBOEOFHBUJWF/PUBCMZ UIFTF SFTVMUTBSFDPOTJTUFOUXJUIUIFQPTJUJWF IBQQJOFTT PS iCJUUFSTXFFUu OPTUBMHJB FNPUJPOTSFQPSUFEJOUIF NFNPSZDPOEJUJPOT To test the predictive power of the listeners’ .FD4DBMFSBUJOHT XFDPOEVDUFEBNVMUJQMFEJTDSJNJOBOUBOBMZTJT5IJTBOBMZTJTGPDVTFEPOQSFEJDUJOH the target mechanism condition (four levels: brain TUFNSFëFY DPOUBHJPO NFNPSZ FYQFDUBODZ CBTFE POUIFMJTUFOFSTSBUJOHTPGUIFFJHIU.FD4DBMFJUFNT #FOFGJUJOHGSPNUIFDPNCJOFEBOBMZTJTBDSPTTBMM four experiments featuring 240 cases, we were able to obtain a ratio of 30 observations for each predictor and 60 observations in each category (a ratio of at least 20 observations for each predictor and 20 PCTFSWBUJPOTJOFBDIDBUFHPSZJTSFDPNNFOEFE)BJS "OEFSTFO 5BUIBN #MBDL õýýü 5IFQSFEJDUPST were entered into the analysis using a simultaneous estimation and assuming equal probabilities of occurSFODF öù 8JUIGPVSDBUFHPSJFT XFDPVMEFTUJNBUF three discriminative functions: Function 1, canonical RƎƎü÷ 8JMLTTλõú χ2ƎƎø÷÷ûü pƎƎôôõ'VODtion 2, canonical RƎƎúö 8JMLTTλƎƎùô χ2ƎƎõú÷õö

pƎƎôôõBOE'VODUJPO÷ DBOPOJDBMRƎƎøø 8JMLTT λƎƎüô χ2ƎƎùõöø pƎƎôôõ Table 8 shows the results in a classification maUSJY*UNBZCFTFFOUIBUUIFPWFSBMMIJUSBUJPBDSPTTUIF four emotion categories was 75% correct, which can be compared with the hit ratio of 25% that could be FYQFDUFEGSPNDIBODFBMPOF$MBTTJêDBUJPOBDDVSBDZ ranged from 57% to 94% depending on the target mechanism, with best result for memory and worst GPSFYQFDUBODZ5IFSFBSFOPHFOFSBMHVJEFMJOFTGPS how to interpret the classification accuracy relative

TABLE 8. Classification Matrix for the Multiple Discriminant Analysis: Prediction of Mechanism Condition From MecScale Items (N = 240) Predicted condition Brain stem

Contagion Expectancy Memory

Actual condition Brain stem

88

2

7

3

Contagion

7

62

16

15

Expectancy

22

20

57

1

3

3

0

94

Memory

Note. The percentages (rowwise) of correctly predicted emotions are given on the main diagonal (in bold). The off-diagonal cells show the confusions. Overall accuracy = 75%.



UPDIBODF CVU)BJSFUBM õýýü BSHVFEUIBUUIFBDcuracy should be at least one-fourth greater than UIBUBDIJFWFECZDIBODF ú÷JOUIJTDBTF

BOEUIF currently observed increase in accuracy relative to chance is approximately 12 times greater than this DSJUFSJPO)PXFWFS JUTIPVMECFOPUFEUIBUJOUIFBCsence of a cross-validation procedure due to a small sample, the estimate is likely to be positively biased to TPNFEFHSFF$POTJEFSBUJPOPGUIFJOEJWJEVBMQSFEJDUPST UIF.FD4DBMFJUFNT SFWFBMFEUIBUBMMJUFNTFYcept visual imagery and cognitive appraisal produced TJHOJêDBOUWBMVFTPGQBSUJBM8JMLTTλ, indicating that UIFZNBEFBVOJRVFDPOUSJCVUJPOUPUIFQSFEJDUJPO .PTUWBSJBODFXBTFYQMBJOFECZUIFCSBJOTUFNSFëFY item (FUPSFNPWFƎƎùüýù

GPMMPXFECZUIFNVTJDBM FYQFDUBODZ ÷ôõü

DPOUBHJPO õ÷÷ý

BOEFQJTPEJD NFNPSZ ûùõdfƎƎ÷ ööý BMMpTƎƎôù JUFNT

A

B

PSYCHOPHYSIOLOGY.

Figure 1 presents means and standard errors (zTDPSFT for listeners’ skin conductance level, zygomaticus activity, and corrugator activity as a function of target NFDIBOJTN BDSPTT&YQFSJNFOUTõmø8JUISFHBSEUP skin conductance level, it can be seen that the experimental conditions were clearly separated from baseline, but they were not clearly differentiated from one BOPUIFS)PXFWFS UIFCSBJOTUFNSFëFYBOENFNPSZ conditions tended to show higher levels of skin conductance than did the contagion and expectancy DPOEJUJPOT'BDJBM&.(TVHHFTUFEBDMFBSFSEJďFSFOUJBUJPOCFUXFFOUIFDPOEJUJPOT*OQBSUJDVMBS UIF contagion and expectancy conditions showed lower levels of zygomaticus activity and higher levels of corrugator activity than did the brain stem reflex BOENFNPSZDPOEJUJPOT'VSUIFSNPSF OPUFUIBUUIF contagion conditions showed lower levels of zygomaticus activity than baseline and that the memory conditions showed lower levels of corrugator activity UIBOCBTFMJOF

C

DISCUSSION Summary of Findings

FIGURE 1. Means and standard errors for the listeners’ (a) skin conductance level, (b) zygomaticus muscle activity, and (c) corrugator muscle activity (z scores) as a function of target mechanism condition, Experiments 1–4

In this study, we aimed to selectively manipulate four mechanisms believed to underlie emotional reactions to music, through a careful selection of existing pieces of music, to see whether we would be able to demonstrate predictable effects on listeners’ emotional SFTQPOTFT5IFSFTVMUTGSPNPVSGPVSFYQFSJNFOUT DBOCFTVNNBSJ[FEBTGPMMPXT

t JUSLIN ET AL .


First, we conclude that the target mechanisms aroused emotions in listeners largely in accordance with our theoretical predictions: The listeners’ selfreports revealed that the brain stem reflex conditions aroused the most surprise, the contagion conditions aroused the most sadness, the episodic memory conditions aroused the most nostalgia and happiness, and the musical expectancy conditions aroused the NPTUBOYJFUZ"MUIPVHIUIFFďFDUTWBSJFETMJHIUMZCFtween the experiments, planned comparisons showed that the ratings were in line with our theoretical preEJDUJPOTJOüøPGUIFDPOUSBTUT JF UIFNFBOSBUJOH was significantly higher for the predicted mechanism UIBOGPSUIFPUIFSDPOEJUJPO *OUIFSFNBJOJOHDBTFT the rating for the predicted mechanism was either still UIFIJHIFTU BMUIPVHIOPUTJHOJêDBOUMZTP õö

PSUIF TFDPOEIJHIFTUPGUIFDPOEJUJPOT ø *UTIPVMEBMTP be noted that the effects of the experimental manipuMBUJPOPOUIF QSFEJDUFE FNPUJPOTXFSFiNPEFSBUFu UPiMBSHFu 'FSHVTPO öôôý BOEUIBUSBUFEPWFSBMM intensity was high (MƎƎöúú POBTDBMFGSPNôUPø 4FDPOE UIFTFSFTVMUTXFSFTVQQPSUFECZQTZDIPphysiology in terms of autonomic activity and facial FYQSFTTJPO.FBTVSFTPGTLJODPOEVDUBODFBOE[Zgomaticus and corrugator activity were significantly influenced by target mechanism condition in the NBKPSJUZ ü÷ PGDBTFT5IFTFEBUBBSFJNQPSUBOU for at least two reasons: They validate the conclusion that the listeners actually experienced emotions rather than merely perceiving emotions in the music, and they display patterns consistent with the FNPUJPOSBUJOHT"MUIPVHIUIFTFêOEJOHTXFSFOPU as clear-cut as we had hoped, it could be seen that the conditions that were predicted to evoke negatiWFFNPUJPOT JF DPOUBHJPO→ sadness, expectancy →BOYJFUZ QSPEVDFENPSFDPSSVHBUPSNVTDMFBDUJvity and less zygomaticus muscle activity, than the conditions that were predicted to evoke a neutral CSBJOTUFNSFëFYƎ→Ǝsurprise PSQPTJUJWF NFNPSZ →IBQQJOFTT OPTUBMHJB FNPUJPO BTFYQFDUFEGSPN their relative position in the circumplex model (RusTFMM õýüô 5IFEBUBGPSTLJODPOEVDUBODFEJEOPU distinguish between target mechanism conditions RVJUFBTDMFBSMZ"MMFYQFSJNFOUBMDPOEJUJPOTUFOEFE to produce higher autonomic activity than baseline, but expected contrasts between specific emotions FH TVSQSJTFƎ←Ǝ→ƎTBEOFTT XFSFHFOFSBMMZOPU TJHOJêDBOU/PUF IPXFWFS UIBUXIFOEBUBGSPNUIF FYQFSJNFOUTXFSFDPMMBQTFE 'JHVSFõ

EBUBTVHHFTUFE

that typically high-arousal emotions such as surprise and happiness did produce higher levels of skin conductance than did typically low-arousal emotions TVDIBTTBEOFTT5IFTFêOEJOHTBSFDMFBSMZXPSUIZ of further investigation, ideally using a design with better statistical power, because many of the effects BQQFBSUPCFTVCUMF Third, the results regarding emotion ratings and psychophysiology were extended by the data for .FD4DBMF UIBUJT UIFFJHIUTFMGSFQPSUJUFNTGPDVTJOHPOTVCKFDUJWFJNQSFTTJPOT FH i%JEUIFNVTJD FWPLFBNFNPSZPGBOFWFOUGSPNZPVSMJGF ui8FSF you ‘touched’ by the emotional expression of the muTJD u 3FTVMUTJOEJDBUFEUIBUUIFJUFNTXFSFSFMJBCMZ SFMBUFEUPUIFDPSSFTQPOEJOHUBSHFUNFDIBOJTN" multiple discriminant analysis showed that the items could predict the target mechanism condition with BOPWFSBMMDMBTTJêDBUJPOBDDVSBDZPGûù5IVT FWFO though self-reports regarding causes of emotions cannot be treated as veridical, they are hardly arbitrary FJUIFS*OGBDU JGNVTJDMJTUFOFSTBSFHJWFOIJHIMZUBSgeted questions, their responses provide indices that NBZDPNQMFNFOUPUIFSGPSNTPGFWJEFODF"TDPVME CFFYQFDUFE UIFDMBTTJêDBUJPOCBTFEPO.FD4DBMF worked somewhat better for mechanisms that by defiOJUJPOBSFFYQMJDJUJOOBUVSF FH FQJTPEJDNFNPSJFT than for mechanisms that have been conceptualized BTNPSFJNQMJDJU FH NVTJDBMFYQFDUBODZTFF+VTMJO öôõ÷ 5IJTEJďFSFODFTIPVMECFUBLFOJOUPBDDPVOU when using self-reports to explore underlying mechaOJTNTJOGVUVSFêFMETUVEJFT Problems, Limitations, and Future Directions

This is the first study to contrast mechanisms using only existing pieces of music, and the present results can be compared with those of our previous study, which used computer-manipulated versions of a QJFDF +VTMJOFUBM öôõø 5IJTTUVEZDPSSPCPSBUFTUIF êOEJOHTJOUIBUTUVEZCZTIPXJOHUIBU SFBTPOBCMZ predictable response patterns may be obtained also XJUIFYJTUJOHQJFDFTPGNVTJD5IFJODMVTJPOPGGPVS different pieces to represent each target mechanism, all pieces taken from the actual repertoire of classical NVTJD JODSFBTFTUIFHFOFSBMJ[BCJMJUZPGUIFSFTVMUT 4UJMM JUJTBMTPDMFBSUIBUUIFBSPVTFEFNPUJPOTXFSFOPU BTOFBUMZEJďFSFOUJBUFEIFSFBTJOUIFQSFWJPVTTUVEZ This might be symptomatic of the greater difficulty in clearly separating different mechanisms when using real pieces of music, which typically feature several EMOTIONAL REACTIONS TO MUSIC t


EJďFSFOULJOETPGFNPUJPOBMMZSFMFWBOUJOGPSNBUJPO One promising approach to better separate the effects of distinct mechanisms when using ecologically valid music may be to combine real compositions with reTZOUIFTJT +VTMJO.BEJTPO õýýý

GPSJOTUBODFCZ digitally editing specific features to reduce the effects PGQBSUJDVMBSNFDIBOJTNT In our previous study, which relied on computer NBOJQVMBUJPO +VTMJOFUBM öôõø

UIFNVTJDBMFYQFDtancy version did not arouse anxiety to the extent we had predicted, but in the present investigation, which featured real compositions by highly accomplished composers, the induction of anxiety was far NPSFTVDDFTTGVM&WFOTP NVTJDBMFYQFDUBODZJTPOF of the mechanisms that need closer attention in future XPSL"MUIPVHIJUJTPGUFODMBJNFEUIBUUIJTJTPOF PGUIFNPTUQPXFSGVMNFDIBOJTNT FH 5IPNQTPO öôôý

FďFDUTJOTUVEJFTTPGBSIBWFHFOFSBMMZCFFO NPEFTUDPNQBSFEXJUIUIPTFPGPUIFSNFDIBOJTNT It is possible that the use of an unfamiliar musical genre reduced the amount of stylistic expectancies that listeners could bring to the listening situation, UIFSFCZSFEVDJOHUIFFďFDUPGUIFNBOJQVMBUJPO "MUIPVHIUIFQSFTFOUNBOJQVMBUJPOPGNFDIBOJTNT did arouse the predicted emotions, it is apparent that some other emotions were also aroused to some exUFOU BMCFJUJOXFBLFSGPSN*ONPTUDBTFT UIFTFFďFDUT were consistent with the predicted emotions, for inTUBODFJOUFSNTPGWBMFODF'PSFYBNQMF JUJTOPUTVSQSJsing that the manipulations yielded significant effects on calm–contentment, given that the ratings of this FNPUJPOXFSF OFHBUJWFMZ DPSSFMBUFEXJUIBOYJFUZm OFSWPVTOFTT POFPGPVSQSFEJDUFEFNPUJPOT0UIFS UFOEFODJFT XFSF NPSF JOUSJHVJOH 5IF DPOUBHJPO conditions, in particular, aroused nostalgia–longing, which is more commonly associated with episodic NFNPSJFTMJOLFEUPNVTJD DG+BOBUBFUBM öôôû+VTMJOFUBM öôôü 5IJTêOEJOHBQQFBSTRVJUFQV[[MJOH considering that the music was unfamiliar to the listeners and that they presumably did not have specific NFNPSJFTBTTPDJBUFEXJUIUIFNVTJD0OFQPTTJCMF explanation could be that the nostalgia was a by-effect of the music-evoked sadness rather than an effect of UIFNVTJD*UIBTCFFOSFQPSUFEQSFWJPVTMZUIBUPOF common trigger of nostalgia is negative affects such as TBEOFTT 8JMETDIVU 4FEJLJEFT "SOEU 3PVUMFEHF öôôú

BOEUIFTFFNPUJPOTIBWFCFFOIBSEUPEJTFOtangle in previous studies (Vuoskoski, Thompson, .D*MXBJO &FSPMB öôõö 5IVT +VTMJOFUBM öôõø

speculated that once the listener becomes sad through contagion, this emotion may evoke nostalgia–longing BMTP5IJTOPUJPODPVMEQFSIBQTCFUFTUFECZVTJOH even shorter musical excerpts or by adopting a contiOVPVTSFTQPOTFUFDIOJRVF 4DIVCFSU öôõô It could seem odd that the memory conditions XIJDIFWPLFENBJOMZQPTJUJWFFNPUJPOT SFDFJWFE lower ratings of liking than the contagion conditions XIJDIFWPLFENBJOMZOFHBUJWFFNPUJPOT )PXFWFS this finding may arguably reflect that the two processes—preference and emotion—are partly indepenEFOU5IVT UIFMJTUFOFSTFYQFSJFODFEQPTJUJWFFNPtions as a result of the memories evoked, even though UIFZEJEOPUQBSUJDVMBSMZMJLFUIFNVTJD$POWFSTFMZ the pieces featured in the contagion conditions were liked to a greater extent, despite the fact that they BSPVTFEBHSFBUEFBMPGTBEOFTT5IJTDBOCFSFMBUFEUP previous findings that some people actively seek out NVTJDUIBUNBLFTUIFNFYQFSJFODFTBEOFTT (BSSJEP 4DIVCFSU öôõõ The most uniform results across the present experiments occurred for the brain stem reflex mechaOJTN5IJTJTXIBUXFXPVMEFYQFDU CBTFEPOUIF #3&$7&."GSBNFXPSL +VTMJO öôõ÷

XIJDIIPMET that this mechanism is mainly hard-wired and subject UPMJUUMFFďFDUPGJOEJWJEVBMFYQFSJFODF$POWFSTFMZ XF would expect larger variability and effects of personal experience with regard to the memory condition, XIJDIJTBMTPXIBUXFGPVOE.FD4DBMFEBUBJOEJDBUFE that the memory condition “scattered” its effects to a larger extent than the other three conditions, evoking not only episodic memories but also more general BTTPDJBUJPOTBOEJNBHFT*UJTOPUTVSQSJTJOHUIBUUIF memory condition correlated with the visual imagery JUFNCFDBVTFFQJTPEJDNFNPSJFTPGUFOJOWPMWFJNBHFSZ )BEUIFSFQPSUFEJNBHFSZOPUSFTVMUFEGSPNFQJTPEJD memories, we would have expected to find much greater incidence of imagery in the other mechanism conditions that did notFWPLFNFNPSJFT5IFDPSSFMBUJPOXJUI cognitive appraisal can seem surprising, but it makes sense because memories often involve memories of the BQQSBJTBMUPUIFPSJHJOBMFWFOU &MMTXPSUI õýýø "HBJO had the reported appraisals not been tied to episodic memories, we would have expected to find a greater JODJEFODFPGBQQSBJTBMTJOUIFPUIFSDPOEJUJPOTBMTP Of the episodic memories, 25% were described as both QPTJUJWFBOEOFHBUJWFJOWBMFODF5IJTNBZCFSFMBUFE to the fact that nostalgia is commonly regarded as a iCJUUFSTXFFU uNJYFEFNPUJPO 8JMETDIVUFUBM öôôú

t JUSLIN ET AL .


One limitation of this study is that the listener sample was small and featured listeners from only BTJOHMF8FTUFSODVMUVSF5IFSFJTBOVSHFOUOFFE for cross-cultural studies adopting a psychological perspective to explore the role of various mechanisms JOEJWFSTFDVMUVSFT)PXNBZQTZDIPMPHJDBMNFDIBnisms manifest themselves in diverse cultures? JusMJO öôõö BSHVFEUIBUBOBDDPVOUPGUIFJOEVDUJPOPG emotions can be cross-culturally valid at the level of mechanisms, despite cross-cultural diversity in musiDBMTVSGBDFGFBUVSFTBOEFWPLFEFNPUJPOT "MUIPVHI music that arouses nostalgia in listeners in one culture may sound distinct from music that arouses nostalgia in listeners in another culture, this does not rule out that the emotion was aroused for the same reasons in CPUIDBTFT )PXFWFS JUTFFNTQMBVTJCMFUIBUEJďFSFOU mechanisms are important in different cultures depending on both the music and the functions of the NVTJD 4BBSJLBMMJP öôõö 5IFSFGPSF UIFQSFWBMFODF of different mechanisms and its relationship to the functions of the music in various cultures warrant GVSUIFSJOWFTUJHBUJPO Primary Role of Mechanisms

"OJNQPSUBOURVFTUJPOJTXIBUEFUFSNJOFTXIJDI NFDIBOJTN JGBOZ JTBDUJWBUFECZBQBSUJDVMBSNVTJDBM FWFOU5IFBOTXFSEFQFOETPOTFWFSBMGBDUPST4PNF JOGPSNBUJPODPVMECFQSPWJEFECZUIFNVTJD FH FYUSFNFTPVOEFWFOUT FNPUJPOBMFYQSFTTJPOT

PUIFS JOGPSNBUJPONJHIUEFSJWFGSPNUIFDPOUFYU FH BO BFTUIFUJDGSBNJOH PSUIFMJTUFOFS FH UIBUBQJFDFIBT PGUFOPDDVSSFEJOUIFQFSTPOTMJGF 5IJTTUVEZJTMJNited because it does not take individual or contextual WBSJBCMFTJOUPBDDPVOU"NVTJDBMFWFOUNBZiBďPSEu (JCTPO õýûý BQBSUJDVMBSFNPUJPOBMSFTQPOTFCZ featuring information relevant to a particular mechanism, but whether this information will activate the mechanism depends on the listener’s attention, which in turn may depend on the context (focused listenJOHPSCBDLHSPVOENVTJD

BOEPOXIBUPUIFSUZQFT PG QPUFOUJBMMZDPNQFUJOH JOGPSNBUJPOPDDVSTBUUIF TBNFUJNF*USFNBJOTUPCFFYQMPSFEXIBUUZQFTPG JOGPSNBUJPOIBWFQSJPSJUZBOEXIZ'PSJOTUBODF XJMMB melodic theme associated with a certain emotion for a listener be overridden by a perceptually more salient drum stroke that happens to occur at the same time? 8FNBZBTTVNFUIBUJGBHJWFONVTJDBMFWFOUGBJMT to include information relevant for the activation of any mechanism, then consequently no emotion will

be aroused, which does seem to be common (see +VTMJOFUBM öôôü *OPUIFSXPSET JGUIFNVTJDEPFT notJODMVEFFYUSFNFTPVOEFWFOUT CSBJOTUFNSFëFY

BQSPOPVODFEBOEDBUDIZSIZUIN FOUSBJONFOU

B QBTTJPOBUFBOEWPJDFMJLFFYQSFTTJPO DPOUBHJPO

B structural feature that invites metaphorical analogies UPFYUFSOBMFWFOUT WJTVBMJNBHFSZ

BOVOFYQFDUFE tonal, harmonic, or rhythmic sequence (musical exQFDUBODZ

PSBOBFTUIFUJDRVBMJUZTVDIBTWBTUCFBVUZ BFTUIFUJDKVEHNFOU IBTOPUCFFOMJOLFEXJUIFNPtionally laden life events (evaluative conditioning, FQJTPEJDNFNPSZ PS MFTTQMBVTJCMZ EPFTOPUIBWF DSVDJBMJNQMJDBUJPOTGPSPOFTHPBMTJOMJGF BQQSBJTBM

then chances are slim that the music will evoke an FNPUJPO1SFDJTFMZUIFTFDJSDVNTUBODFTBQQMJFEUPUIF neutral piece used in this study, which apparently did OPUBDUJWBUFBOZPGUIFNFDIBOJTNT5PBWFSZNPEFTU extent, at least, we were able to switch on mechanisms BUXJMMJOPSEFSUPBSPVTFTQFDJêDFNPUJPOT"TLFQUJD might argue that, because the excerpts were selected to tap into specific mechanisms, it is trivial that the FYDFSQUTQSPEVDFEEJďFSFOUSFTQPOTFT)PXFWFS JUJT crucial to note that the manipulation did not simply QSPEVDFEJďFSFOUFNPUJPOTXFDPVMEBMTPMBSHFMZpredictUIFQSFDJTFFNPUJPOT/PUPOMZDPVMEXFQSFEJDU the emotions, we could also predict target mechanism conditions from self-reported listener impressions .FD4DBMF EJSFDUMZSFMFWBOUUPFBDINFDIBOJTN FH that the listener experienced episodic memories or was touched by the emotional expression of the muTJD 5BLFOUPHFUIFS UIFTFêOEJOHTTUSPOHMZJOEJDBUF that music may evoke emotions in listeners through each of the four causal processes outlined at the outTFUPGUIFTUVEZ#FJOHBCMFUPQSFEJDUBOEDPOUSPM aroused emotions in terms of specific mechanisms is the ultimate evidence of a valid process descripUJPO5PEFTDSJCFUIFGFBUVSFTPGUIFNVTJDJTPOMZB êSTUTUFQUPXBSEBQTZDIPMPHJDBMFYQMBOBUJPO8IBU matters is not the musical features as such but what meaning they are given by our mechanisms—the disUJODUJPOCFUXFFOTPVOEBOETJHOJêDBODF Concluding Remarks

(VUIFJM õýùö OPUFEUIBUiFNPUJPOBMSFBDUJPOTXIJDI music may arouse are as numerous as the individuals reacting” and that “the subjectivity of emotional exQFSJFODFTJTUIFDPSFPGPVSQSPCMFNu Qõù &Npirical studies of specific mechanisms are still in their JOGBODZ)PXFWFS FWFOUIPVHIFNPUJPOBMSFBDUJPOT EMOTIONAL REACTIONS TO MUSIC t


to music continue to defy simple conclusions, the results of the present study are sufficiently encouraging to suggest that the multiple-mechanism approach is a promising avenue toward understanding the mystery PGFNPUJPOBMSFBDUJPOTUPNVTJD APPENDIX: RESPONSE SHEET FOR MECHANISM INDICES (MECSCALE)

#SBJO4UFN3FëFY

õ %JEUIFNVTJDGFBUVSFBOFWFOUUIBUTUBSUMFEZPV 3IZUINJD&OUSBJONFOU

ö %JEUIFNVTJDIBWFBTUSPOHBOEDBQUJWBUJOHQVMTF rhythm? &QJTPEJD.FNPSZ

÷ %JEUIFNVTJDFWPLFBNFNPSZPGBOFWFOUGSPNZPVS life? &WBMVBUJWF$POEJUJPOJOH

ø %JEUIFNVTJDFWPLFNPSFHFOFSBMBTTPDJBUJPOT 7JTVBM*NBHFSZ

ù %JEUIFNVTJDFWPLFJNBHFTXIJMFZPVXFSFMJTUFOJOH $POUBHJPO

ú8FSFZPVUPVDIFECZUIFFNPUJPOBMFYQSFTTJPOPGUIF music? .VTJDBM&YQFDUBODZ

û 8BTJUEJĐDVMUUPHVFTTIPXUIFNVTJD FH NFMPEZ would continue over time? $PHOJUJWF"QQSBJTBM

ü%JEUIFNVTJDIBWFBOZQSBDUJDBMDPOTFRVFODFTGPS your goals or plans in life? NOTES

5IFQSFTFOUTUVEZXBTTVQQPSUFECZUIF4XFEJTI3FTFBSDI $PVODJMUISPVHIBHSBOUUP1BUSJL/+VTMJO "EESFTTDPSSFTQPOEFODFBCPVUUIJTBSUJDMFUP1BUSJL/ +VTMJO %FQBSUNFOUPG1TZDIPMPHZ 6QQTBMB6OJWFSTJUZ #PY õööù 4&ûùõøö6QQTBMB 4XFEFO FNBJMQBUSJLKVTMJO! QTZLVVTF REFERENCES

"MUBSSJCB + öôõö &NPUJPOBOENPPE0WFSõöôZFBSTPG contemplation and exploration in the American Journal of PsychologyAmerican Journal of Psychology, 125, 409– øöö "OESFBTTJ +- öôôû Psychophysiology: Human behavior & physiological response ùUIFE )JMMTEBMF /+&SMCBVN #BMM 1, öôõô The music instinct: How music works and why we can’t do without it. -POEPO &OHMBOE#PEMFZ )FBE #BVNHBSUOFS ) õýýö 3FNFNCSBODFPGUIJOHTQBTU.VTJD BVUPCJPHSBQIJDBMNFNPSZ BOEFNPUJPOAdvances in Consumer Research, 19,úõ÷múöô

#MBJS .& 4IJNQ 5" õýýö $POTFRVFODFTPGBO VOQMFBTBOUFYQFSJFODFXJUINVTJD"TFDPOEPSEFSOFHBUJWFDPOEJUJPOJOHQFSTQFDUJWFJournal of Advertising, 21, ÷ùmø÷ #SPXO 4 .BSUJOF[ .+ 1BSTPOT -. öôôø 1BTsive music listening spontaneously engages limbic and QBSBMJNCJDTZTUFNTNeuroreport, 15,öô÷÷möô÷û $POPWFS 8+ õýýý Practical non-parametric statistics ÷SEFE /FX:PSL /:8JMFZ %FOOFUU %$ õýüû The intentional stance. Cambridge, .".*51SFTT %JOHMF (" 4BWJMM 4 'SBTFS $ 7JFUI ) öôõõ Music that moves us: In search of the mechanisms linking music listening with emotional response. Poster presented at UIFNFFUJOHPGUIF*OUFSOBUJPOBM4PDJFUZGPS3FTFBSDIPO &NPUJPOT ,ZPUP +VMZöôõõ %PXMJOH 8+ )BSXPPE %- õýüú Music cognition. /FX:PSL /:"DBEFNJD1SFTT &MMTXPSUI 1$ õýýø -FWFMTPGUIPVHIUBOEMFWFMTPGFNPUJPO*O1&LNBO3+%BWJETPO &ET

The nature of emotion: Fundamental questions QQõýömõýú 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT 'FSHVTPO $+ öôôý "OFďFDUTJ[FQSJNFS"HVJEFGPSDMJOJDJBOTBOESFTFBSDIFSTProfessional Psychology: Research and Practice, 40,ù÷ömù÷ü 'PXMFT %$ $ISJTUJF .+ &EFMCFSH 3 (SJOHT 88 -ZLLFO %5 7FOBCMFT 1) õýüõ 1VCMJDBUJPOSFDPNNFOEBUJPOTGPSFMFDUSPEFSNBMNFBTVSFNFOUTPsychophysiology, 18,ö÷ömö÷ý 'PY & öôôü Emotion science.#BTJOHTUPLF 6,1BMHSBWF .BDNJMMBO 'SJEMVOE "+ $BDJPQQP +5 õýüú (VJEFMJOFTGPSIVNBOFMFDUSPNZPHSBQIJDSFTFBSDIPsychophysiology, 25, ùúûmùüý 'SJFE 3 #FSLPXJU[ - õýûý .VTJDUIBUDIBSNTBOE DBOJOëVFODFIFMQGVMOFTTJournal of Applied Social Psychology, 9,õýýmöôü (BCSJFMTTPO " öôõõ Strong experiences with music: Music is much more than just music. Oxford, England: Oxford 6OJWFSTJUZ1SFTT (BSSJEP 4 4DIVCFSU & öôõõ *OEJWJEVBMEJďFSFODFTJO UIFFOKPZNFOUPGOFHBUJWFFNPUJPOJONVTJD"MJUFSBUVSF SFWJFXBOEFYQFSJNFOUMusic Perception, 28,öûýmöýù (JCTPO ++ õýûý The ecological approach to visual perception#PTUPO .")PVHIUPO.JđJO (PXFOTNJUI 8/ #MPPN -+ õýýû 5IFFďFDUTPG IFBWZNFUBMNVTJDPOBSPVTBMBOEBOHFSJournal of Music Therapy, 34,÷÷møù (SFHPSZ ") õýýû 5IFSPMFTPGNVTJDJOTPDJFUZUIFFUIOPNVTJDPMPHJDBMQFSTQFDUJWF*O%+)BSHSFBWFT"$ /PSUI &ET

The social psychology of music QQõö÷mõøô 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT (VUIFJM &" õýùö *OUSPEVDUJPO*O"$BSQVSTP 7Ǝ3Ǝ'JTJDIFMMJ -(JMNBO &"(VUIFJM +58SJHIU 'Ǝ1BQFSUF &ET

Music and your emotions: A practical

t JUSLIN ET AL .


guide to music selections associated with desired emotional responses QQýmõ÷ /FX:PSL /:-JWFSJHIU )BJS +' "OEFSTPO 3& 5BUIBN 3- #MBDL 8Ǝ$ õýýü Multivariate data analysis ùUIFE 6QQFS4BEEMF3JWFS /+1SFOUJDF)BMM )BSHSFBWFT %+ õýüú The developmental psychology of music.$BNCSJEHF &OHMBOE$BNCSJEHF6OJWFSTJUZ 1SFTT )BSSFS ( )BSSFS ) õýûû .VTJD FNPUJPO BOEBVUPOPNJDGVODUJPO*O.$SJUDIMFZ3")FOTPO &ET

Music and the brain: Studies in the neurology of music QQ öôömöõú -POEPO &OHMBOE8JMMJBN)FJOFNBOO )FWOFS , õý÷ù 5IFBďFDUJWFDIBSBDUFSPGUIFNBKPSBOE NJOPSNPEFTJONVTJDAmerican Journal of Psychology, 47,õô÷mõõü )FWOFS , õý÷ú &YQFSJNFOUBMTUVEJFTPGUIFFMFNFOUTPG FYQSFTTJPOJONVTJDAmerican Journal of Psychology, 48, öøúmöúü )VSPO % öôôõ *TNVTJDBOFWPMVUJPOBSZBEBQUBUJPO Annals of the New York Academy of Sciences, 930,ø÷múõ )VSPO % öôôú Sweet anticipation: Music and the psychology of expectation.$BNCSJEHF .".*51SFTT *[BSE $Ǝ& õýûû Human emotions/FX:PSL /:1MFOVN *[BSE $& öôõô 5IFNBOZNFBOJOHTBTQFDUTPGFNPUJPOT %FêOJUJPOT GVODUJPOT BDUJWBUJPO BOESFHVMBUJPOEmotion Review, 2,÷úõm÷ûô +BOBUB 1 5PNJD 45 3BLPXTLJ 4, öôôû $IBSBDUFSJ[BUJPOPGNVTJDFWPLFEBVUPCJPHSBQIJDBMNFNPSJFT Memory, 15,üøùmüúô +VTMJO 1/ õýýû 1FSDFJWFEFNPUJPOBMFYQSFTTJPOJOTZOthesized performances of a short melody: Capturing the MJTUFOFSTKVEHNFOUQPMJDZMusicae Scientiae, 1,ööùmöùú +VTMJO 1/ öôôô 7PDBMFYQSFTTJPOBOENVTJDBMFYQSFTTJPO 1BSBMMFMTBOEDPOUSBTUT*O",BQQBT &E

Proceedings of the Sixteenth Conference of the International Society for Research on Emotions QQöüõmöüø 2VFCFD$JUZ 2$ *43&1VCMJDBUJPOT +VTMJO 1/ öôôù How does music arouse emotions? Paper presented at the Thirteenth Conference of the InternaUJPOBM4PDJFUZGPS3FTFBSDIPO&NPUJPOT #BSJ *UBMZ +VMZ õõmõù öôôù +VTMJO 1/ öôõõ .VTJDBOEFNPUJPO4FWFORVFTUJPOT TFWFOBOTXFST*O*%FMJÍHF+%BWJETPO &ET

Music and the mind: Essays in honour of John Sloboda QQõõ÷mõ÷ù 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT +VTMJO 1/ öôõö "SFNVTJDBMFNPUJPOTJOWBSJBOUBDSPTT cultures? Emotion Review, 4,öü÷möüø +VTMJO 1/ öôõ÷ 'SPNFWFSZEBZFNPUJPOTUPBFTUIFUJD FNPUJPOT5PXBSEBVOJêFEUIFPSZPGNVTJDBMFNPUJPOT Physics of Life Reviews, 10,ö÷ùmöúú +VTMJO 1/ )BSNBU - &FSPMB 5 öôõø 8IBUNBLFT music emotionally significant? Exploring the underlying NFDIBOJTNTPsychology of Music, 42,ùýýmúö÷ +VTMJO 1/ -BVLLB 1 öôô÷ $PNNVOJDBUJPOPGFNPUJPOTJOWPDBMFYQSFTTJPOBOENVTJDQFSGPSNBODF%JG-

ferent channels, same code? Psychological Bulletin, 129, ûûômüõø +VTMJO 1/ -BVLLB 1 öôôø &YQSFTTJPO QFSDFQUJPO BOEJOEVDUJPOPGNVTJDBMFNPUJPOT"SFWJFXBOEBRVFTUJPOOBJSFTUVEZPGFWFSZEBZMJTUFOJOHJournal of New Music Research, 33,öõûmö÷ü +VTMJO 1/ -JMKFTUSÕN 4 -BVLLB 1 7ÅTUGKÅMM % -VOERWJTU -0 öôõõ &NPUJPOBMSFBDUJPOTUPNVTJD JOBOBUJPOBMMZSFQSFTFOUBUJWFTBNQMFPG4XFEJTIBEVMUT 1SFWBMFODFBOEDBVTBMJOëVFODFTMusicae Scientiae, 15, õûømöôû 4QFDJBM*TTVFPO.VTJDBOE&NPUJPO

+VTMJO 1/ -JMKFTUSÕN 4 7ÅTUGKÅMM % #BSSBEBT ( 4JMWB " öôôü "OFYQFSJFODFTBNQMJOHTUVEZPGFNPUJPOBMSFBDUJPOTUPNVTJD-JTUFOFS NVTJD BOETJUVBUJPO Emotion, 8,úúümúü÷ +VTMJO 1/ -JMKFTUSÕN 4 7ÅTUGKÅMM % -VOERWJTU -0 öôõô )PXEPFTNVTJDFWPLFFNPUJPOT &YQMPSJOHUIF VOEFSMZJOHNFDIBOJTNT*O1/+VTMJO+"4MPCPEB &ET

Handbook of music and emotion: Theory, research, applications QQúôùmúøö 0YGPSE &OHMBOE0YGPSE 6OJWFSTJUZ1SFTT +VTMJO 1/ .BEJTPO ( õýýý 5IFSPMFPGUJNJOHQBUterns in recognition of emotional expression from musical QFSGPSNBODFMusic Perception, 17,õýûmööõ +VTMJO 1/ 4DIFSFS ,3 öôôù 7PDBMFYQSFTTJPOPG BďFDU*O+")BSSJHBO 33PTFOUIBM ,34DIFSFS &ET

The new handbook of methods in nonverbal behavior research QQúùmõ÷ù 0YGPSE &OHMBOE0YGPSE 6OJWFSTJUZ1SFTT +VTMJO 1/ 7ÅTUGKÅMM % öôôü &NPUJPOBMSFTQPOTFTUP NVTJD5IFOFFEUPDPOTJEFSVOEFSMZJOHNFDIBOJTNT Behavioral and Brain Sciences, 31,ùùýmùûù +VTMJO 1/ ;FOUOFS .3 öôôö $VSSFOUUSFOETJOUIF TUVEZPGNVTJDBOEFNPUJPO0WFSUVSFMusicae Scientiae, Special Issue 2001–2002,÷möõ ,MFJOHJOOB 13 ,MFJOHJOOB ". õýüõ "DBUFHPSJ[FE list of emotion definitions, with a suggestion for a consenTVBMEFêOJUJPOMotivation and Emotion, 5,÷øùm÷ûõ KonečOJ 7+ 8BOJD 3" #SPXO " öôôû &NPUJPOBM and aesthetic antecedents and consequences of musicJOEVDFEUISJMMTAmerican Journal of Psychology, 120, úõýmúø÷ ,SVNIBOTM $- õýýû "OFYQMPSBUPSZTUVEZPGNVTJDBM FNPUJPOTBOEQTZDIPQIZTJPMPHZCanadian Journal of Experimental Psychology, 51,÷÷úm÷ùö -BOH 1+ (SFFOXBME ., #SBEMFZ .. )BNN "Ǝ0 õýý÷ -PPLJOHBUQJDUVSFT"ďFDUJWF GBDJBM WJTDFSBM BOECFIBWJPSSFBDUJPOTPsychophysiology, 30,öúõmöû÷ -BSTFO +5 #FSOUTPO (( 1PFIMNBOO ,. *UP 5" $BDJPQQP +5 öôôü 5IFQTZDIPQIZTJPMPHZPGFNPUJPO*O.-FXJT +.)BWJMBOE+POFT -'#BSSFUU &ET

The handbook of emotions ÷SEFE QQõüômõýù /FX:PSL /:(VJMGPSE -BSUJMMPU 0 5PJWJBJOFO 1 &FSPMB 5 öôôü ".BUMBC UPPMCPYGPSNVTJDJOGPSNBUJPOSFUSJFWBM*O$1SFJTBDI



)#VSLIBSEU -4DINJEU5IJFNF 3%FDLFS &ET

Data analysis, machine learning, and applications: Studies in classification, data analysis, and knowledge organization QQöúõmöúü #FSMJO (FSNBOZ4QSJOHFS -FWFOTPO 38 öôôû &NPUJPOFMJDJUBUJPOXJUIOFVSPMPHJDBMQBUJFOUT*O+"$PBO++#"MMFO &ET

Handbook of emotion elicitation and assessment QQõùümõúü 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT -VOERWJTU -0 $BSMTTPO ' )JMNFSTTPO 1 +VTMJO 1 / öôôý &NPUJPOBMSFTQPOTFTUPNVTJD&YQFSJFODF FYQSFTTJPO BOEQIZTJPMPHZPsychology of Music, 37,úõmýô .FZFS -# õýùú Emotion and meaning in music. Chicago, *-$IJDBHP6OJWFSTJUZ1SFTT .PSFT 3 öôôý )VNBOWPJDF"TQBSTF NFBOJOHGVMBOE DBQBCMFSFQSFTFOUBUJPOPGTPVOET*O..#PPOF &E

Proceedings of the NAG/DAGA International Conference on Acoustics, Rotterdam, March 2009 QQüûùmüûü #FSMJO (FSNBOZ(FSNBO"DPVTUJDBM4PDJFUZ /FBMF +. -JFCFSU 3. õýüú Science and behavior ÷SEFE &OHMFXPPE$MJďT /+1SFOUJDF)BMM /PSUI "$ )BSHSFBWFT %+ öôôü The social and applied psychology of music. Oxford, England: Oxford 6OJWFSTJUZ1SFTT 0BUMFZ , ,FMUOFS % +FOLJOT + öôôú Understanding emotions öOEFE 0YGPSE &OHMBOE#MBDLXFMM 0SUPOZ " $MPSF ( $PMMJOT " õýüü The cognitive structure of emotions. Cambridge, England: Cambridge 6OJWFSTJUZ1SFTT 0TCPSOF +8 õýüô 5IFNBQQJOHPGUIPVHIUT FNPUJPOT TFOTBUJPOT BOEJNBHFTBTSFTQPOTFTUPNVTJDJournal of Mental Imagery, 5,õ÷÷mõ÷ú 1JLF " õýûö "QIFOPNFOPMPHJDBMBOBMZTJTPGFNPUJPOBM FYQFSJFODFJONVTJDJournal of Research in Music Education, 20,öúömöúû 1MVUDIJL 3 õýýø The psychology and biology of emotion. /FX:PSL /:)BSQFS$PMMJOT 3PCJOTPO + öôôù Deeper than reason: Emotion and its role in literature, music, and art0YGPSE &OHMBOE$MBSFOEPO 3VTTFMM +" õýüô "DJSDVNQMFYNPEFMPGBďFDUJournal of Personality and Social Psychology, 39,õõúõmõõûü 4BBSJLBMMJP 4 öôõö $SPTTDVMUVSBMBQQSPBDIFTUPNVTJDBOE IFBMUI*O3.BD%POBME (,SFVU[ -.JUDIFMM &ET

Music, health, and well-being QQøûûmøýô 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT 4DIFSFS ,3 ;FOUOFS .3 öôôõ &NPUJPOBMFďFDUTPG NVTJD1SPEVDUJPOSVMFT*O1/+VTMJO+"4MPCPEB &ET

Music and emotion: Theory and research QQ÷úõm ÷ýö 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT 4DIVCFSU & öôõô $POUJOVPVTTFMGSFQPSUNFUIPET*O1Ǝ/ +VTMJO+"4MPCPEB &ET

Handbook of music and

emotion: Theory, research, applications QQöö÷möù÷ 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT 4JMWJB 1+ öôõö )VNBOFNPUJPOTBOEBFTUIFUJDFYQFSJFODF "OPWFSWJFXPGFNQJSJDBMBFTUIFUJDT*O"14IJNBNVSB 4&1BMNFS &ET

Aesthetic science: Connecting minds, brains, and experience QQöùômöûù 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT 4JNPOT 3$ õýýú Boo! Culture, experience, and the startle reflex.0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT 4MPCPEB +" õýýú &NPUJPOBMSFTQPOTFTUPNVTJD"SFWJFX*O,3JFEFSFS5-BIUJ &ET

Proceedings of the Nordic Acoustical Meeting QQ÷üùm÷ýö )FMTJOLJ5IF "DPVTUJDBM4PDJFUZPG'JOMBOE 4MPCPEB +" +VTMJO 1/ öôôõ 1TZDIPMPHJDBMQFSTQFDUJWFTPONVTJDBOEFNPUJPO*O1/+VTMJO+"4MPCPEB &ET

Music and emotion: Theory and research QQ ûõmõôø 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT 4PLPMPW &/ õýú÷ )JHIFSOFSWPVTGVODUJPOTUIFPSJFOUJOHSFëFYAnnual Review of Physiology, 25,ùøùmùüô 4UFJOCFJT / ,PFMTDI 4 4MPCPEB +" öôôú 5IFSPMF of harmonic expectancy violations in musical emotions: Evidence from subjective, physiological, and neural reTQPOTFTJournal of Cognitive Neuroscience, 18,õ÷üômõ÷ý÷ 5IPNQTPO 8' öôôý Music, thought, and feeling. Understanding the psychology of music. Oxford, England: 0YGPSE6OJWFSTJUZ1SFTT 7VPTLPTLJ +, 5IPNQTPO # .D*MXBJO % &FSPMB 5 öôõö 8IPFOKPZTMJTUFOJOHUPTBENVTJDBOEXIZ Music Perception, 29,÷õõm÷õû 8BUFSNBO . õýýú &NPUJPOBMSFTQPOTFTUPNVTJD*NQMJDJU BOEFYQMJDJUFďFDUTJOMJTUFOFSTBOEQFSGPSNFSTPsychology of Music, 24 ù÷múû 8FMMT " )BLBOFO &" õýýõ 5IFFNPUJPOBMVTFTPG QPQVMBSNVTJDCZBEPMFTDFOUTJournalism Quarterly, 68, øøùmøùø 8JMETDIVU 5 4FEJLJEFT $ "SOEU + 3PVUMFEHF $ öôôú /PTUBMHJB$POUFOU USJHHFST GVODUJPOTJournal of Personality and Social Psychology, 91,ýûùmýý÷ 8JUWMJFU $7 7SBOB 43 öôôû 1MBZJUBHBJO4BN Repeated exposure to emotionally evocative music polarises liking and smiling responses, and influences other BďFDUJWFSFQPSUT GBDJBM&.( BOEIFBSUSBUFCognition & Emotion, 21,÷möù ;FOUOFS .3 &FSPMB 5 öôõô 4FMGSFQPSUNFBTVSFTBOE NPEFMT*O1/+VTMJO+"4MPCPEB &ET

Handbook of music and emotion: Theory, research, applications QQ õüûmööõ 0YGPSE &OHMBOE0YGPSE6OJWFSTJUZ1SFTT ;FOUOFS .3 (SBOEKFBO % 4DIFSFS ,3 öôôü Emotions evoked by the sound of music: CharacterizaUJPO DMBTTJêDBUJPO BOENFBTVSFNFOUEmotion, 8, 494– ùöõ

t JUSLIN ET AL .
