"Back in my day...": A Preliminary Study on the Differences in Generational Groups Perception of Musically-evoked Emotion

“BACK IN MY DAY...”: A PRELIMINARY STUDY ON THE DIFFERENCES IN GENERATIONAL GROUPS PERCEPTION OF MUSICALLY-EVOKED EMOTION Ped o L. LOURO([email p o ec ed])(0000-0003-3201-6990)1and Rena o PANDA([email p o ec ed])(0000-0003-2539-5590)1,2 1CISUC, DEI,Uni e si y o Coimb a,Coimb a, LASI, Po ugal 2Ci2 — Sma Ci ies Resea ch Cen e ,Poly echnic Ins i u e o Toma ,Toma , Po ugal ABSTRACT The inc easingly globalized wo ld we li e in oday and he wide a ailabili y o music a ou inge ips ha e led o mo e di e se musical as es wi hin younge gene a ions han in olde gene a ions. Mo eo e , hese dispa i ies a e s ill no well unde s ood, and he ex en o which hey a - ec lis ene s’ p e e ences and pe cep ion o music. Fo- cusing on he la e , his s udy explo es he di e ences in emo ional pe cep ion o music be ween he Millennials and Gen Z gene a ions. In e iews we e conduc ed wi h six pa icipan s equally dis ibu ed be ween bo h gene a ions by eco ding hei lis ening expe ience and emo ion pe - cep ion on wo p e iously compiled se s o songs ep e- sen ing each g oup. Signi ican di e ences be ween gene - a ions and possible con ibu ing ac o s we e ound in he analysis o he conduc ed in e iews. Findings poin o di - e ences in he pe cep ion o ene gy o songs wi h speci ic messages o su e ing o lo e, as well as a endency om he younge g oup o pe cei e a well-de ined emo ion in songs ep esen ing hei gene a ion in con as o neu al esponses om he o he g oup. These indings a e p elim- ina y, and u he s udies a e needed o unde s and hei ex en . Ne e heless, aluable insigh s can be ex ac ed o imp o e music ecommenda ion sys ems. 1. INTRODUCTION In music psychology, many s udies on he ac o s ha con- ibu e o he di e ences when epo ing musically-e oked emo ion show ha hese a e ela ed o backg ound, gen e, pe sonali y, musical educa ion, as e, and epe oi e [1]. These a e no mally i ele an when dealing wi h pe cei ed emo ion, bu e idence shows ha cul u al backg ound may indeed in luence emo ion pe cep ion in music [2]. These s udies a e s ill ela i ely ecen and no mally ocus on he di e ences be ween wo dis inc cul u es. Bu wha i we conside he e olu ion o cul u es? Va ious s ud- ies show ha ce ain gene a ions ha e gene alized music p e e ences, such as Baby Boome s (a ound 1946 h ough 1964), who ha e a Rock p e e ence, o Gene a ion X (a ound 1965 h ough 1980) who p e e G unge and Hip- Copy igh : © 2024. This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License, which pe mi s un e- s ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Hop, which e lec he iews o socie y, wi h o agains i , a each ime pe iod. Despi e his, mainly due o he g owing globaliza ion o cul u e and he widesp ead access o music, p e e ences ha e become di icul o a ibu e o younge gene a ions, such as Millennials and Gene a ion Z (o Zoome s). Al hough o p e ious gene a ions, we can deduce ha music emo ion pe cep ion has some simi- la bias ans e sal o each gene a ion, his is no so clea o he mo e ecen ones [3]. Two esea ch ques ions we e o mula ed in o de o un- de s and po en ial di e ences in pe cep ion ac oss he mo e globalized gene a ions: i) How di e en is he emo- ional pe cep ion be ween Millennials and Zoome s?; ii) Wha ac o s con ibu e o emo ional pe cep ion di e - ences? To answe he i s ques ion, he pe cei ed emo- ion in a se o songs mus be collec ed om a ious mem- be s o bo h gene a ions. This can be done by showing a small exce p o hose songs and explici ly asking o hei emo ional con en , no how i made hem eel. To explain he p e ious answe s and po en ial di e ences, he second ques ion needs o be answe ed, which can be done by ind- ing common hemes be ween he a ious membe s’ eali- ies. To answe hese ques ions, a s udy is conduc ed ollow- ing a phenomenological app oach ha i s con ex ualizes he musical ends ep esen ed by music popula a he end o he gene a ion in o which he subjec s all. The sub- jec s a e asked o ecoun hei expe ience wi h each song and hei o e all expe ience wi h he songs lis ened o. The same is conduc ed o he musical ends o he o he gene - a ion. The e o e, he s udy adop s a social cons uc i ism lens o ake in o accoun he mul iple eali ies desc ibed by he di e en pa icipan s. Con ibu ions om his s udy a e wo old: Fi s , e idence o di e ences in pe cep ion be ween bo h gene a ional g oups is p o ided ac oss he wo gene a ion’s musical ends. Second, possible ac- o s con ibu ing o such di e ences a e desc ibed, such as he di e en pe cep ion o ene gy in songs ha deal wi h he message o su e ing o lo e and he younge gene - a ion’s p eposi ion o pe cei e mo e p onounced emo ions in songs ha con ey low emo ional in ensi y in con as o he olde gene a ion. Figu e 1. Russell’s Ci cumplex Model adap ed om [4]. As seen in he igu e, each quad an ep esen s an emo- ion, bu should hese alues be low enough he emo ion is conside ed neu al. A = A ousal h eshold; V = Valence h eshold. 2. BACKGROUND 2.1 Emo ion and Music F om he i uals conduc ed by ou a -o ances o s o ad- jus ing ou mood acco ding o he ask a hand in he p esen day, such as in a p o essional con ex , music has always been e y impo an o ou way o li e. Inhe en o hinking o music is he emo ion ha comes a ached o i , ega ded as he main eason why we engage wi h i [5] and which na u e has been a deba e o many cen u ies [6]. Some a gue ha he emo ions e oked by music a e me ely pe cei ed by he lis ene and no ac ually induced wi hin hem; o he s de end he opposi e no ion and he mos ac- cep ed idea ha bo h a e p esen . Howe e , induced emo- ions a e p one o high subjec i i y, which makes i ha d o ind ag eemen when asking subjec s o epo hei expe- iences [7], and his is why pe cei ed emo ion is p e e ed in mos con empo a y quan i a i e s udies as i is conside - ably less subjec i e. Fu he mo e, he e is also a need o ep esen he whole emo ion spec um o collec anno a ions o hese s udies, o which emo ion models exis . These can be ei he ca e- go ical o dimensional. The o me is usually di ided in o a se numbe o clus e s o g oups o ela ed emo ions, he mos commonly used being He ne ’s Adjec i e Ci - cle [8], and is easy o anno a o s o unde s and. Howe e , ca ego ical models do no accu a ely ep esen he emo- ion spec um due o he s ic bounda ies be ween clus e s when some emo ions may lie he e. Dimensional models sol e his p oblem by mapping emo ions in o a con inuous space on se e al planes, such as he Russell’s Ci cumplex Model [9] seen in Figu e 1, which mos commonly is ep- esen ed by wo planes ela ed wi h human biological sys- ems o di e en ia ing emo ions, hese being a ousal, he amoun o ene gy p esen in he s imuli, and alence, how nega i e o posi i e his s imulus is. Mos li e a u e ega ding he ields ha s udy emo ion in music uses he la e model due o i s accu a e spec um ep esen a ion. Howe e , i is known ha anno a o s end o ha e di icul y unde s anding he wo kings o he model and need some p e ious guidance o p o ide accu a e an- no a ions [10]. The wide adop ion o his model can be seen, o example, in de eloping da ase s o Music Emo- ion Recogni ion, such as he DEAM da ase [11]. 2.2 In luences On Emo ion Pe cep ion One o he majo conce ns o music psychology when deal- ing wi h emo ions is he se o ac o s ha in luence sub- jec s’ emo ional esponses. As mos music is no usu- ally con ined o speci ic con ex s, a ious indi idual ac- o s ha e been conside ed, such as age, gende , music edu- ca ion, lis ening habi s, and pe sonali y [1,12,13], o name a ew. Resea che s ha e also ex ensi ely s udied cul u al back- g ound as a c i ical ac o , emphasizing i s impac h ough di e ences in adi ions, socie al no ms, and di e se mu- sic heo ies ac oss he globe. To some ex en , some s udies ha e ound common emo ion pe cep ions be ween Wes e n and indigenous A ican cul u es [14]. These a e, howe e , limi ed o some o he basic emo ions om Ekman’s s ud- ies [15]. Mo e ho ough s udies on Wes e n and Eas e n cul u al di e ences in emo ion pe cep ion showed signi i- can disc epancies [2]. These s udies ocus on di e en cul u es, and o he au- ho s’ knowledge, no s udy ocuses on he di e ences be- ween wo gene a ional g oups wi h simila cul u al back- g ounds. 3. METHOD 3.1 Pa icipan s The pa icipan s o his s udy we e mainly chosen o i in o ha pa icula gene a ion, al hough di e en ages om he al eady in e iewed pa icipan s we e p e e ed la e on. A subjec is conside ed a Millennial o Zoome be ween 43-26 and 25-11 yea s o age, espec i ely. We p oceed o epo he mos ele an in o ma ion om he pa icipan s. The p opo ion o subjec s pe gene a ion o which hey belong was e en, wi h h ee subjec s in each. As o he dis ibu ion o ages wi hin hese gene a ions, Millennials we e mo e encompassing o ha gene a ion han Zoome s, as can be seen by he a e age age being 33, a ound he middle o ha gene a ion. Howe e , o Zoome s, he e is a highe concen a ion a he beginning since he age mean is a ound 23, close o he bounda y p e iously de ined. Al- hough no conside ed ele an o his s udy, he p opo - ion o he subjec ’s gende is hea ily male-skewed, which may lead o some bias on he conclusions aken om he indi idual expe ience. Pa icipan s a e e e ed o as P*, whe e * is a numbe a ibu ed o dec easing age, mean- ing ha P1 e e s o he oldes pa icipan and P6 o he younges . As o music educa ion, hal o he subjec s only ecei ed basic music educa ion as pa o hei school cu iculum (P3, P5, P6), while he o he hal ha e di e en le els o expe ise, om basic music heo y (P2) o ins umen - speci ic (P1), o choi -speci ic educa ion (P4). Finally, e- ga ding music as e, he e is a a ie y o p e e ed gen es. When conside ing Millennials, he e is li le o no o e lap Table 1. Pa icipan s’ indi idual in o ma ion. Gene a ion Age Gende P o essional Backg ound P e ious Music Educa ion F equency o Music Lis ening P e e ed Gen es P1 Millennials 42 M S udying Compu e Science; Wo ked on he same ield Sel -lea ned: In e media e (Applied o Gui a ) A leas 3 days pe week Rock, Jazz, Blues, Reggae, Classic P2 31 M S udying Compu e Science; Wo king on A i icial In eligence Sel -lea ned: Basic E e y day Samba, Bossa No a, House P3 27 M S udying Compu e Science; S udied Cinema Basic School Educa ion E e y day Pop, EDM, K-Pop, Punk Rock, T ance, Classic P4 Zoome s 25 M S udied Museology; Wo king in a Scien i ic Cen e Choi -educa ed E e y day Pop, Rock, La ino, KPop P5 24 F S udying Languages and Li e a u e Basic School Educa ion A leas 3 days pe week Pop, Indie Rock, Lo-Fi P6 21 M S udying Ci il Enginee ing Basic School Educa ion Almos e e y day Hea y Me al, Me al, Rock be ween subjec s’ p e e ences, while he e is some o e - lap in Zoome s, speci ically in Rock. This in o ma ion is summa ized in Table 1. 3.2 Ma e ial o Conduc ing he S udy Fo conduc ing he da a collec ion p ocess, a playlis ep- esen ing he o e all music ends o each gene a ion was compiled. P ope ly de ining wha music ends needed o be ep esen ed o each gene a ion would be ano he s udy al oge he , so a mo e s aigh o wa d app oach was ol- lowed. Fi s , a poin in ime ep esen a i e o each g oup had o be de ined. Achie ing his is challenging because he goal is o cap u e mos musical ends wi hin each gene a ion’s de ined 15-yea pe iod. The chosen app oach was o es ab- lish e e ence poin s a he yea s ha signi y he s a o he succeeding gene a ion, i.e., 1997 o Millennials and 2012 o Zoome s. This las one was ex apola ed om p e ious gene a ions’ ends as his poin lacks gene al ag eemen . Ha ing de ined hese poin s, a se o songs om hose yea s needed o be chosen. A his s age, i was decided o only accoun o songs desc ibed as Pop due o hei mains eam appeal. The amoun o songs ha p esen con- lic ing emo ional con en , i.e., ambiguous emo ion, o ha lack emo ion in gene al, is also conside ably educed com- pa ed o o he gen es. Songs we e pulled om Billboa d’s Ho 100 cha om each poin in ime, agg ega ing he op 100 songs played in he USA each week since he 1960s. Billboa d was chosen as he p o ide o his da a as i ep- esen s he mos in luen ial egion o his gen e, he Uni ed S a es o Ame ica, o he ele an ime pe iods. Wi h he songs picked, exce p s ep esen a i e o hose songs’ emo ional con en we e ob ained o conduc he s udy. The 30-second p e iews on Spo i y we e chosen o his pu pose. The choice was e iden since his s eaming pla o m has one o he bigges lib a ies in he ma ke , de- c easing he p obabili y o no ma ch being ound om each cha . I also allows playback o he men ioned p e iews om a web b owse a no cha ge o non-comme cial use. The e is, howe e , he ca ea o he e being no gua an ee ha hese p e iews ep esen he song’s emo ional con en . A e e iewing hem, he au ho s conside hem adequa e o he s udy, bu his poin should be u he add essed in u u e s udies. A sc ip was de eloped o au oma ically c ea e a Spo i y playlis om he p o ided Billboa d’s Ho 100 cha week, whe e all p e iews’ URLs we e ex ac ed and sa ed. The inal playlis s a e shown in Table 2. 3.3 Da a Collec ion All in e iews began by asking o ele an pe sonal in o - ma ion om each pa icipan , including age, gende , place o bi h, and p o essional si ua ion. This in o ma ion is necessa y o place he pa icipan in o one o he s udy’s gene a ions and unde s and possible bias behind hei an- swe s o he ollowing ques ions. In addi ion, we also col- lec ed he le el o musical educa ion, equency o music lis ening, and music p e e ences da a o he same pu pose. The collec ion p ocess con inues wi h he pa icipan lis- ening o he p e iously compiled music exce p s o six songs ep esen a i e o hei gene a ion. A e each exce p , h ee ques ions a e asked: • ‘Do you ha e pas expe iences wi h his song?’, o unde s and i he e is any in luence o hei judgmen on he emo ional pe cep ion o ha exce p de i ed om pas expe iences; • ‘How would you desc ibe you o e all expe ience?’, in an e o o sepa a e he induced om he pe - cei ed emo ion; • ‘Wha is he p edominan emo ion you pe cei e on his exce p ?’. This que y should ou pu he pe cei ed emo ion om he song wi hou in luences om p e ious expe iences and p o ide some insigh in o he ac o s behind i . A he end o he playlis , he pa icipan is ques ioned abou hei o e all expe ience wi h he se o songs. They a e encou aged o desc ibe hei di e ences and simila - i ies, p o iding mo e insigh s in o he emo ion pe cei ed in each song. The p ocess is eplica ed o he playlis o he al e na e gene a ion. To conclude he in e iew, pa ic- ipan s a e asked o e lec on he wo playlis s, mo e speci - ically, iden i y ecu ing hemes and highligh he mos el- e an di e ences. 3.4 Da a Analysis In o de o be e analyze he answe s gi en o he pe - cei ed emo ion in each song, each is mapped o Russell’s Ci cumplex model, p e iously p esen ed in Sec ion 2. Fo answe s ha all in one o he quad an s, we conside he o e all emo ion as he p edominan emo ion o ha quad- an . Fo example, should esponses pe cei e sadness and dep ession in one pa icula song, he o e all pe cep ion would be sad since bo h epo ed emo ions all in o Q3, Table 2. The inal playlis s used in he s udy - one o each gene a ion. Millennials Zoome s "Semi-Cha med Li e" by "Thi d Eye Blind" "Wha Makes You Beau i ul" by "One Di ec ion" "Bi ch" by "Me edi h B ooks" "Se Fi e o he Rain" by "Adele" "Qui Playing Games (Wi h My Hea )" by "Backs ee Boys" "Payphone" by "Ma oon 5" "Un-B eak My Hea " by "Toni B ax on" "We Found Lo e" by "Rihanna" "Say Youll Be The e" by "Spice Gi ls" "Call Me Maybe" by "Ca ly Rae Jepsen" "All By Mysel " by "Céline Dion" "Ti anium ( ea . Sia)" by "Da id Gue a" mainly ela ed o sadness. In cases whe e he answe s e- eal unce ain y o con lic s be ween quad an s, he o e all emo ion is ma ked as neu al since i does no i en i ely in o a pa icula quad an . The nex s ep was o ex ac o e all hemes om he in- e iews ha could explain di e ences be ween and wi hin gene a ions. The s eps a e ou lined nex , ollowing he p o- cess desc ibed in C esswell [16]. Each in e iew was ansc ibed om hei espec i e eco dings be o e being b oken up in o smalle meaning ul uni s, i.e., e idences, o u he analysis. The ansc ip s we e ead, and memoing was conduc ed in o de o p o- ide a mo e gene al iew o he da a and iden i y common hemes be ween pa icipan s. Some concep s, i.e., codes, we e iden i ied by c oss- e e encing he ele an e idence in a sp eadshee . The documen o ganiza ion comp ehends he pa icipan , i s gene a ion, and he co esponding playlis ha p oduced he e idence. The eme ging codes om he in e iews ol- lowed hese. When e idence is conside ed ele an , his is ma ked acco dingly. Finally, co ela ions be ween each code we e analyzed, g ouping hese codes in o mo e gen- e al hemes. 4. RESULTS This sec ion p esen s he indings o he p esen s udy. A ho ough desc ip ion o he di e ences in pe cep ion be- ween bo h gene a ions is gi en be o e p esen ing he com- mon hemes ound in he in e iews ha may ac o in o hei answe s. 4.1 O e all Findings Fo assessing he ag eemen be ween gene a ions’ emo- ional pe cep ion o each song, he mode o he answe s was conside ed as he o e all answe o each gene a ion. O he wel e songs used o his s udy, only i e had ull ag eemen . The emo ion accoun s discussed below e e o pe cep ion only i no s a ed o he wise. We conside an ac- coun neu al only i no o he emo ion is pe cei ed by he pa icipan a any poin . Rega ding he Millennials playlis , h ee ou o he six songs do no show ag eemen , namely "Bi ch" by "Me ed- i h B ooks", "Say You’ll Be The e" by "Spice Gi ls", and "Qui Playing Games (Wi h My Hea )" by "Backs ee Boys". Pa icipan s om he Zoome gene a ion pe cei ed all o hese songs as Relaxed, while Millennial pa icipan s did no pe cei e any conc e e emo ion. In o he wo ds, he emo ional con en was pe cei ed as neu al in mos o he songs, excep he las one men ioned abo e, which was pe cei ed as happy. In e es ingly, he e was e en less ag eemen conside ing he Zoome s’ se o songs, om which ou ou o six songs did no ecei e ag eemen be ween he gene a ions, namely "Payphone" by "Ma oon 5", "We Found Lo e" by "Ri- hanna", "Ti anium" by "Da id Gue a ea . Sia", and "Se Fi e o he Rain" by "Adele". S a ing wi h "Payphone", mos o he Zoome gene a ion conside ed his song sad, while Millenians epo ed mo e neu al esponses. The la - e also epo ed neu al esponses o "We Found Lo e" and "Ti anium", while he o me pe cei ed hese songs as happy. As o "Se Fi e o he Rain", Zoome s mos ly pe - cei ed he song as sad, al hough one pa icipan om his g oup pe cei ed i as ang y. Millennials mos ly pe cei ed i as ang y, excep o one pa icipan who pe cei ed a hap- pie emo ion. F om he accoun s ega ding hei own playlis s, i was clea ha he olde gene a ion is mo e a ached o he p o- ided se o songs. When explici ly asked i hey iden i- ied hemsel es wi h hei gene a ion’s playlis , Zoome s ended o discuss how i did no ma ch hei as e. This inding co obo a es he mo e spa se as es in younge gen- e a ions as discussed back in Sec ion 2. 4.2 Fac o s Con ibu ing o Emo ion Pe cep ion F om he da a analysis o he in e iews desc ibed in Sec- ion 3.4, a ious common hemes we e ound ha may no only explain he obse ed di e ences, bu also explain he la ge numbe o neu al esponses men ioned abo e. In addi ion, some hemes also show ha he indi idual ex- pe ience does no in luence he emo ion pe cep ion o he pa icipan s. Desc ip ions o codes and hemes discussed in his sec ion can be ound in Table 4. 4.2.1 Pas Expe iences Do No In luence Judgemen Many accoun s show some in luence o nos algia on he ecoun s o he lis ening expe ience om each pa icipan . Some imes e y explici ly, ‘’A mix o nos algia and hap- piness.’(P5, "Payphone"), ‘’(The song) gi es me nos algia and calm ( eelings)...’(P4," Say You’ll Be The e"), ‘’Nos- algia, simple imes...’(P6, "Semi-Cha med Li e"). Many imes hese expe iences a e bound o some speci ic con ex , such as ime wi h speci ic pe iods o li e, ‘Some hing an angs y eenage would hea .’(P2, "Semi-Cha med Li e"), a social con ex , ‘I associa e ("We Found Lo e") a lo wi h going ou a nigh as a eenage .’(P3), o e en social be- ha io , ’... we wan o dance his wi h ha gi l we like’ Table 3. Emo ions pe cei ed by each subjec o bo h playlis s. Gene a ion Song P1 P2 P3 P4 P5 P6 O e all Millennials O e all Zoome s Millennials “Semi-Cha med Li e” Sad Happy Happy Happy Sad Happy Happy Happy by “Thi d Eye Blind” “Bi ch” Neu al Neu al Neu al Relaxed Happy Relaxed Neu al Relaxed by “Me edi h B ooks” “Qui Playing Games (Wi h My Hea )” Happy Happy Happy Happy Relaxed Relaxed Happy* Relaxed* by “Backs ee Boys” “Un-B eak My Hea ” Sad Sad Sad Sad Sad Sad Sad Sad by “Toni B ax on” “Say You’ll Be The e” Neu al Neu al Happy Happy Relaxed Relaxed Neu al Relaxed by “Spice Gi ls” “All By Mysel ” Sad Ang y Sad Sad Sad Sad Sad Sad by “Céline Dion” Zoome s “Wha Makes You Beau i ul” Happy Happy Happy Happy Happy Happy Happy Happy by “One Di ec ion” “Se Fi e o he Rain” Neu al Ang y Neu al Sad Ang y Sad Ang y Sad by “Adele” “Payphone” Neu al Neu al Neu al Sad Neu al Sad Neu al Sad by “Ma oon 5” “We Found Lo e” Sad Neu al Neu al Happy Neu al Happy Neu al Happy* by “Rihanna” “Call Me Maybe” Neu al Neu al Neu al Neu al Neu al Neu al Neu al Neu al by “Ca ly Rae Jepsen” “Ti anium ( ea . Sia)” Sad Neu al Neu al Neu al Ang y Happy Neu al Happy** by “Da id Gue a” *Emo ions epo ed we e o low in ensi y, bu no neu al. **O e all disag eemen wi hin gene a ion; Ex a commen s om he pa icipan s we e conside ed o map o e all emo ion. (P1, "Un-B eak My Hea "). Ano he common heme, es- pecially in he se o songs o he younge gene a ion, was he e ec o con inuous exposu e o a gi en song ha hea - ily impac ed he lis ening expe ience. The exposu e could ei he lead o a e sion, ‘F us a ion, gi en he imes I hea d his song be o e.’(P2, "Wha Makes You Beau i ul"), o in- di e ence, ‘Plays a lo in he adio... i eels e y neu al o me.’(P1, "Payphone"). E en hough he e a e some biases ega ding he lis ening expe ience, his did no ansla e o he emo ion pe cei ed in he song, as seen in Table 3. This heme does no p o ide a gumen s o he di e ences be- ween gene a ions bu con i ms ha pe cei ed and induced emo ion accoun s we e adequa ely sepa a ed. 4.2.2 Disag eeing On Su e ing Fo Lo e An ex ensi ely s udied opic in music psychology is he pa adox o nega i e emo ions. The pa adox e e s o lis- ene s epo ing pleasu able expe iences om music pieces in which emo ional con en is pe cei ed as nega i e, e.g., ‘Happy o hea his song (desc ibing lis ening expe i- ence)... Dep essi e, o he poin o cu ing my w is s (de- sc ibing pe cei ed emo ion)’(P2, "Un-B eak My Hea "). Such pleasu able expe ience may be explained by o e - whelming ocals, namely "Un-B eak My Hea ", "All By Mysel " and "Se Fi e To The Rain". Howe e , he un- de lying message o su e ing o lo e seems o be he mos impo an ac o in songs pe cei ed di e en ly in he Millennial’s playlis be ween gene a ions. Many accoun s seem o ag ee ha mos o ha gene a ion’s music ocuses on sea ching o and su e ing a e losing ’ he one’. This ansla ed o a happy o neu al pe cep ion o Millenni- als, while o Zoome s, i ansla ed o pe cei ing elaxed emo ions. 4.2.3 Disag eeing on Low-In ensi y Emo ions Mos o he oldes gene a ion ends o espond neu ally o Zoome s’ playlis songs, wi h ou ou o six songs pe - cei ed his way by he majo i y. In con as , only "Call Me Maybe" ecei ed a simila neu al pe cep ion among he younge gene a ion. A mo e conc e e example is he ac- coun s om Millennials abou he song "Payphone". These e eal con usion de i ed om he disconnec be ween he ins umen al and ocal emo ional con en , ‘(Rega ding my lis ening expe ience) I am e y con used... The ins umen- al and ocals a e e y con adic o y’(P2, "Payphone"). The accoun s in Table 3 e lec his in lowe pe cei ed in- ensi y, as all accoun s we e Neu al ac oss he gene a ion. Simila o he p e ious heme, he main ac o is he un- de lying message o sel -empowe men iden i ied by pa - icipan s in bo h gene a ions. One o he accoun s nicely sums up his poin : ‘I is always he same hing, nos al- gia, dep ession and he e is no way ou o his ( e e ing o he su e ing o lo e message o he Millennials’ se o songs). Bu in 2012 (Zoome s’ playlis ), i is comple ely di e en because he e is also dep ession, bu hings can u n a ound. The e is mo e han jus sea ching o lo e’ (P5). O e all, Millennials always pe cei e low-in ensi y emo ions as Neu al in he Zoome ’s playlis ega dless o he message. In con as , Zoome s pe cei e happiness de- pending on he unde lying heme. 4.2.4 A Bunch O No hing S ill, on he opic o low-in ensi y emo ions, his seems no o be comple ely cons ained o he Zoome ’s playlis . Some o he e idence poin s o a lack o emo ional engage- men wi h Millennials’ songs, e.g., ‘’Danceable... Has a lo less ene gy han he o he s ( e e ing o he o he songs on he Millennials playlis )’(P2, "Qui Playing Games"), akin o some accoun s ega ding he Zoome ’s playlis , e.g., ‘I s danceable. Wi h his ype o song ( e e ing o We Found Lo e), i is always he same hing’(P4, "Ti anium"). In andem wi h low in ensi y, many accoun s seem o poin o an emo ional dissocia ion wi h ce ain songs. This ap- pea s o lead o neu al esponses in he lis ening accoun s and he pe cei ed emo ion ones, wi h no clea dis inc ion be ween gene a ional g oups. Table 4. Resul s o coding p ocess. The desc ip ions o each code a e based on pa icipan obse a ions, wi h some codes being di ec ly men ioned by one o mo e. Themes we e ound by agg ega ing codes wi h simila concep s. Theme Code Desc ip ion Mi o o Socie y Mi o o Socie y An o e a ching opic ound is he impac o he cu en s a e o socie y on music ends in bo h playlis s. The Millennial’s playlis ocuses on he pu sui o omance and he su e ing a e a b eakup, while in he Zoome ’s playlis he message is abou o e coming ha dships and emb acing you insecu i ies. Disag eeing on Lo e Su e In he Name o Lo e The unde lying message o songs whe e his code was ound is abou he end o oman ic ela ionships, and, mo e o en han no , how ha d i is o he a is o deal wi h i . High-Impac Vocals Mos o he songs ha induced an emo ion on he pa icipan s had some ype o powe ul ocals, many ypes accompanied by messages o sadness, pain o e ol . Some hing ha has been ex ensi ely s udied in music psychology is how songs ha Pa adox o con ey powe ul nega i e emo ions e oke pleasu able eelings on lis ene s. Nega i e Emo ions Due o his pa adox, mos s udies conce ning musically-e oked emo ions a e cons ained o pe cei ed emo ions. Sel -empowe emen The unde lying message o hese songs e okes eelings o sel -wo h in lis ene s and in i es hem o celeb a e hei insecu i ies. Digi al Lo e In mo e ecen songs, ex ensi e use o digi al enhancemen has become he Disag eeing on Low no m in music p oduc ion. Many accoun s om bo h g oups exp ess he In ensi y Emo ions loss o essence de i ed om such hea y use, ega ding he olde songs as mo e soul ul and mo e emo ionally engaging. Ins umen al- ocal Some songs om he Zoome ’s playlis p oduced accoun s o con usion Disco dance de i ed om he emo ional disco dance be ween he ins umen al and ocal pe o mance o ce ain songs. A Bunch o No hing Emo ional Disconnec Many songs om bo h playlis s ailed o elici any emo ion when pa icipan s ecoun ed hei lis ening expe iences, many a ibu ing his o he o mulaic cons uc ion and epea ed ideas in hem. Lack o Emo ional In ensi y When desc ibing hei lis ening expe ience, pa icipan s om bo h gene a ions men ion he lack o ac ual emo ion in he con en o he song, which made hem unsu e o he emo ion hey pe cei ed in he song. Nos algia Googles This code encompasses all accoun s o p e ious expe iences ha may in luence he pe cep ion o emo ion in a gi en song. Many accoun s, especially on music om he Zoome ’s playlis , e eal a Pas Expe iences Replaybili y-based o ced epea ed lis ening o ce ain songs, which is e lec ed Do No In luence A e sion/Indi e ence in he expe ience ecoun s as o e ly nega i e o neu al, Judgemen dis ega ding he ac ual con en o he song. Con ex -bound Expe ience In some cases, he expe ience ecoun o a song was bound o some speci ic con ex , such as pa y music o music o dealing wi h an hea b eak, which may in luence he pe cep ion o emo ion. 4.2.5 Mi o O Socie y Taking in o accoun he ocus o his s udy, we expec o ind some di e ences be ween he gene a ions when ask- ing o he pe cei ed emo ion o many songs, bu he e o- lu ion o he Pop gen e i sel con ibu es e en mo e o his dispa i y. We al eady discussed he di e en messages p esen in bo h se s o songs, and his heme poin s o he cu en s a e o socie y being e lec ed in hese mes- sages a hose poin s in ime, which we can see in s a e- men s such as ‘In oday’s socie y, people a e mo e indi- idualis ic... An awa eness o ou sel es ( e lec ed in he Zoome s’ playlis )’(P3) and ‘’The di e ences ha e much o do wi h social ne wo ks because you ex e nalize you emo ions mo e (in 2012 and beyond). This was mo e he ole o music be o e ( e e ing o he Milennials gene a- ion and be o e)’(P5). A pa icula s a emen gi es an in- e es ing ision o he neu al esponses om Millennials o mos songs on he younge gene a ion playlis , ‘Songs o his e a ( e e ing o Zoome s) ha e mo e di e si y... The e is no pa e n as dema ca ed as in he 90s... (The e is) mo e di usion and accessibili y o music, which is good and no good, since mo e quan i y does no mean be e quali y.’(P2). Ul ima ely, all accoun s ela ed o his heme emphasize he di e ences be ween gene a ions. In addi- ion, he same accoun s s ongly sugges ha changes in socie y, which ickle down o mains eam music, also ha e a subs an ial impac . 5. DISCUSSION Be o e conduc ing his s udy, we expec ed o ind di e - ences in how di e en gene a ions pe cei e emo ions in music. These di e ences a e likely due o each g oup’s unique expe iences du ing c ucial momen s in hei li es and he p e ailing ends and socie al ideas a hose imes. Howe e , we did no make assump ions abou he speci ic ac o s con ibu ing o hese di e ences. The unexpec edly low ag eemen a e may s em om he i s au ho ’s pe spec i e, being pa o he younges gen- e a ion and aligning mo e closely wi h he p edic ions o hose pa icipan s. Addi ionally, he au ho ’s expe ise in Music Emo ion Recogni ion migh ha e led o an o e sigh o he po en ial o many neu al esponses when epo ing he song’s pe cei ed emo ion. This o e sigh is likely be- cause esea ch in his a ea o en excludes songs wi hou a clea dominan emo ion. The indings ela ed o he ac o s con ibu ing o such di e ences indica e ha he gene a ional g oup has some conside able impac . Conside ing Russell’s model, we can no e signi ican di e ences in how hese g oups pe cei e a ousal when lis ening o songs ha con ey he heme o lo e- ela ed su e ing. Millennials end o expe ience highe a ousal, placing hem in he quad an associa ed wi h p edominan ly happie emo ions. On he o he hand, Zoome s end o expe ience lowe a ousal, placing hem in he quad an associa ed wi h p edominan ly elaxed emo- ions. Bo h g oups gene ally ag ee on he alence o songs. This is also e iden in "Se Fi e To The Rain," a song om he Zoome ’s playlis ha con eys a simila message. As be- o e, bo h g oups ag eed on he nega i e conno a ion bu disag eed on he le el o exci emen . Millennials pe cei ed highe exci emen , p ima ily eeling ang y emo ions, while Zoome s el lowe exci emen , p ima ily expe iencing sad emo ions. The emaining songs om he Zoome s’ playlis s e- cei ed mos ly neu al esponses om Millennials due o a pe cei ed lack o in ensi y, as p e iously discussed. In con as , Zoome s seemed o pe cei e a mo e p onounced emo ion. Since his g oup has g own wi h songs p esen ing hese ai s and a ended con ex s ha e oke he pe cei ed emo ions, like he pa y s aples "We Found Lo e" and "Ti- anium", he accoun s a e mo e conc e e, e en when e- po ing ha hey a e de oid o subs ance, akin o Millen- nials. Howe e , hese conclusions a e no inal due o he p elimina y na u e o he p esen s udy. Fu he in es iga- ion needs o be conduc ed wi h a la ge sample size. 6. LIMITATIONS AND FUTURE WORK The esul s we discussed a e p omising bu ha e some lim- i a ions, especially conce ning he numbe o pa icipan s. Bo h gene a ional g oups p o ided a g ea deal o e idence o he hemes ound, bu some o he co ela ions we e weak. Rega ding he pa icipan s, he e is a lack o di e - si y in his s udy sample. Only one pa icipan iden i ied as a woman in he Zoome s g oup, while all pa icipan s in he Millennial g oup iden i ied as men. Despi e his, he la e has a be e age dis ibu ion han he o me , concen a ing on he gene a ion’s beginning. O he indi idual cha ac e is ics seem o be well ep e- sen ed om he au ho ’s pe spec i e. These issues should be add essed o mo e con idence in esul s om simila s udies. Rega ding da a collec ion, he in e iews could ha e p o ided mo e in o ma ion had hey been aped. Since he s udy ollows a phenomenological app oach, physical eac ions o he songs could ha e also p o ided mo e e idence o he hemes ound. The mos gla ing limi a ion is he se o playlis s used o conduc ing his s udy. Since hey a e based on one pa icula momen in ime, hey ha dly ep esen he en- i e y o wo gene a ions. Add essing his poin would be ano he s udy al oge he , as al eady men ioned in Sec ion 3.2. S ill, in u u e wo k, i would be in e es ing o ex end his s udy o o he gene a ions and unde s and i simila hemes eme ge since some o he ound hemes seem o be independen o gene a ion, namely he ones discussed in Sec ions 4.2.1, 4.2.4 and 4.2.5. Ano he in e es ing app oach would be o conduc his s udy wi h o he gen es. Howe e , his may be challeng- ing since some gen es a e no ypically appealing o he masses o a e less emo ional han Pop gene ally is. F om he ou pu o his wo k, he e is some in e es in unde - s anding he ac o s behind he mo e conc e e accoun s o pe cei ed emo ion om Zoome s ega ding hei gene a- ional playlis compa ed wi h Millennials and o wha ex- en he di e ences in a ousal pe cep ion can be obse ed. 7. CONCLUSIONS Music psychology has long s udied di e ences in he pe - cep ion o emo ion in music. Howe e , he ocus on di - e ences be ween gene a ional g oups and co esponding music ends is limi ed. Six pa icipan s om wo gene a ional g oups, i.e., Mil- lennials and Zoome s, lis ened o wo playlis s ep esen - ing hei gene a ion’s music ends o explo e he di e - ences and con ibu ing ac o s. A e ecoun ing hei lis- ening expe ience, pa icipan s we e asked o iden i y he p edominan emo ion hey pe cei ed in each song. The e- po ed emo ional pe cep ions we e mos ly uni o m wi hin each gene a ion bu mo e a ied be ween g oups han ex- pec ed ini ially. Va ious common hemes eme ged a e conduc ing an open coding app oach o he ansc ibed in e iews. One no able inding was he disc epancy in he ene gy con- eyed by songs wi h a message abou su e ing o lo e. Zoome s ended o pe cei e mo e p onounced emo ions in songs om hei playlis , while some epo ed a lack o emo ion du ing he lis ening expe ience ac oss bo h gene - a ions. These indings poin o e ol ing pe cep ions o emo ion in music in successi e gene a ions and conside able di - e ences in music ends. The e a e di e ences no only ega ding he pa icipan gene a ion bu also in he songs hemsel es. Fu u e wo k should expand he sample size o each g oup o consolida e hese indings. Acknowledgmen s This wo k is unded by FCT - Founda ion o Science and Technology, I.P., wi hin he scope o he p ojec s: MERGE - PTDC/CCI-COM/3171/2021 inanced wi h na- ional unds (PIDDAC) ia he Po uguese S a e Budge ; and p ojec CISUC - UID/CEC/00326/2020 wi h unds om he Eu opean Social Fund, h ough he Regional Op- e a ional P og am Cen o 2020. Rena o Panda was sup- po ed by Ci2 - FCT UIDP/05567/2020. 8. REFERENCES [1] Y.-H. Yang, Y.-F. Su, Y.-C. Lin, and H. H. Chen, “Mu- sic emo ion ecogni ion: The ole o indi iduali y,” in P oceedings o he In e na ional Wo kshop on Human- Cen e ed Mul imedia, 2007, pp. 13–22. [2] X. Hu and J. H. Lee, “A c oss-cul u al s udy o music mood pe cep ion be ween ame ican and chinese lis en- e s,” in P oceedings o he 13 h In e na ional Socie y o Music In o ma ion Re ie al Con e ence, 2012, pp. 535–540. [3] T. G een, “muzic app ecia ion: B idging he gap be ween gene a ion z & classical music,” 2018. [Online]. A ailable: h ps://digi alcommons.libe y. edu/mas e s/490/ [4] K. Py o olakis, P. Tzou eli, and G. S amou, “Mul i- modal song mood de ec ion wi h deep lea ning,” Sen- so s, ol. 22, pp. 387–392, 2009. [5] P. N. Juslin and P. Laukka, “Exp ession, pe cep ion, and induc ion o musical emo ions: A e iew and a ques ionnai e s udy o e e yday lis ening,” Jou nal o New Music Resea ch, ol. 33, no. 3, pp. 217–238, 2004. [6] A. Gab ielsson, “Emo ion pe cei ed and emo ion el : Same o di e en ?” Musicae Scien iae, ol. 5, no. 1, pp. 123–147, 2001. [7] A. Pannese, M.-A. Rappaz, and D. G andjean, “Me apho and music emo ion: Ancien iews and u- u e di ec ions,” Consciousness and Cogni ion, ol. 44, pp. 61–71, 2016. [8] K. He ne , “Expe imen al s udies o he elemen s o exp ession in music,” The Ame ican Jou nal o Psy- chology, ol. 48, pp. 246–268, 1936. [9] J. Posne , J. A. Russell, and B. S. Pe e son, “The ci cumplex model o a ec : An in eg a i e app oach o a ec i e neu oscience, cogni i e de elopmen , and psychopa hology,” De elopmen and Psychopa hol- ogy, ol. 17, pp. 715–734, 2005. [10] T. Ee ola and J. K. Vuoskoski, “A compa ison o he disc e e and dimensional models o emo ion in music,” Psychology o Music, ol. 39, no. 1, pp. 18–49, 2011. [11] A. Aljanaki, Y.-H. Yang, and M. Soleymani, “De el- oping a benchma k o emo ional analysis o music,” PLOS ONE, ol. 12, no. 3, pp. 1–22, 2017. [12] C. P. Heinlein, “The a ec i e cha ac e s o he majo and mino modes in music.” Jou nal o Compa a i e Psychology, ol. 8, no. 2, p. 101, 1928. [13] K. He ne , “The a ec i e cha ac e o he majo and mino modes in music,” ol. 47, no. 1, pp. 103–118, 1935. [14] T. F i z, S. Jen schke, N. Gosselin, D. Sammle , I. Pe e z, R. Tu ne , A. D. F iede ici, and S. Koelsch, “Uni e sal ecogni ion o h ee basic emo ions in mu- sic,” Cu en Biology, ol. 19, no. 7, pp. 573–576, 2009. [15] P. Ekman, “An a gumen o basic emo ions,” Cogni- ion and Emo ion, ol. 6, no. 3, pp. 169–200, 1992. [16] J. W. C eswell and C. N. Po h, Quali a i e inqui y and esea ch design: Choosing among i e app oaches. Sage Publica ions, 2016.