an embodied approach to testing musical empathy in
TRANSCRIPT
Anembodiedapproachtotestingmusicalempathyinsubjectswithan
autismspectrumdisorder
LeenDeBruyn1,DirkMoelants1,MarcLeman1
1GhentUniversity,DepartmentofMusicology(IPEM)
Abstract
Wepresentanempiricalandqualitativestudytestingmusicalempathicability insubjectswith
anautismspectrumdisorder(ASD).Fourexperimentsrequiringanincreasinglevelofempathy
withmusic, from synchronization and attuning to emotional empathy,were carried out, using
kinematicdevicesformeasuringembodiedlisteningresponsesandaverbalemotionattribution
task. Results suggest that people with ASD have a corporeal understanding of the affective
featuresofmusic,sincetheyareabletomirrorstructuralandevenaffectivefeaturesofthemusic
into corporeal articulations. However, this corporeal understanding does not give them a
straightforwardaccesstotheemotionalcontentofthemusic.ThesubjectswithASDseemedto
relyondisembodiedcognitiveprocessestoattributeaffectstomusic.
Keywords
autism,music,musicalempathy,embodiment,affectattribution
Oneof themainsymptomsofautismspectrumdisorders isadisturbanceofsocialskills,more
particularlyadisturbanceof theability toshare the feelingsandemotionswithotherpersons.
While this lack of empathy in autism subjects is evident in social contexts, it is less evident
whetheritalsooccursinartisticcontexts.Thispaperaimsattestingtowhatextentsubjectswith
autismcandevelopanempathicrelationshipwithmusicthroughlistening.Thefocusonmusicis
of particular interest because music is known to be very rich in emotional content. Many
(typically developing) people use music for consolation and they often engage in a kind of
emotionalrelationshipwiththemusic(1,2,3).Agoodexampleofthisengagementiswhenpeople
listentosadmusic.Initspropercontext,listeningtosadmusicmayconsoleasubjectbecauseit
allows the subject to find relief in the imitation of themusical emotion, as if themusic is just
another social agent with whom it is possible to share emotions (4). The question is to what
degreesubjectswithautismmayfindasimilarreliefinmusic,orinmoregeneralterms,towhat
degreetheydifferinmusicalempathyfromacontrolgroupoftypicallydevelopingsubjects.
AnumberofstudieshavebeenaddressingtheroleofmusicforpeoplewithASD,therebymainly
focusing on theperception ofmusical structure. Several studies showed thatASD‐subjects are
superior to typically developing subjects when it comes to detailed processing of musical
structures (5). Research of Heaton et al. (6,7,8), for example, showed that children with ASD
outperform their matched controls on pitch discrimination, pitch categorization and pitch
memory tasks.The resultsofBonnel et al. (9) showed that subjectswithASDwere superior to
typicallydevelopingindividualsinthepitchdiscriminationandcategorizationtasks.
Research on ASD and affect perception inmusic aremuchmore rare.Musical affect has been
addressedinastudybyHeatonetal.(10).Inthisstudy,fourteenchildrenwithASDandAsperger
syndromewere tested for theirability to identify theaffectiveconnotationsofmelodies in the
majorandminormusicalmode.Theywererequiredtomatchmusicalexcerptswithschematic
representations of happy and sad faces. The results showed that there were no differences
betweentheclinicalgroupandcontrolsintheirabilitytoascribethemusicalexamplestothetwo
affectivecategories.Theauthorsconcludedthat incontrasttotheirperformanceswithinsocial
andinterpersonalrelations,childrenwithASDshowednodeficitsinprocessingaffectinmusical
stimuli.However, the fact that childrenwithASDcould recognizeemotional expressionat this
simple levelmightnotbesosurprising,given the fact thatemotionrecognition inpeoplewith
ASDisnotalwaysimpairedonalllevelsofcomplexity.Indeed,Cappsetal.(11)andBaron‐Cohen
etal.(12)foundthatchildrenwithASDwereabletounderstandsimpleemotions.Inthestudyof
Capps et al. (11) it was also found that subjects with ASD could understand more complex
emotions,althoughheretheyneededsignificantlymorepromptingandtimethantheirmatched
controls.
In thepresent study, the relationshipbetweenpeoplewithASDandmusic is studied from the
viewpoint of embodied listening and embodied musical empathy. Empathy is traditionally
describedastheabilitytoshareanotherperson’sfeelingsoremotionsasiftheywereyourown.
Empathywithmusicwouldmean thathumanshave theability to identify themselveswith the
emotionalexpressivenessembeddedinthemusic.Leman(4)definesseveralstagesinthisprocess
of empathy with music, namely synchronization, corporeal attuning, corporeal imitation and
emotionalempathy.Allthesegradationsarebasedontheideathatthelistenercangetaccessto
the emotional system of themusic through several gradations of corporeal involvement. This
embodiedviewonmusicalempathyisinagreementwithotherfindingsinemotionresearch(13,14,
15) suggesting that themotor systemmay access the emotional system in different degrees of
engagement.Synchronization is oftendisplayed in tapping along thebeatwhile listening (16,17).
Research has shown that subjects, even without being aware, tend to move body parts in
synchrony with patterns in the physical stimulus (18,19,20,21). Synchronizing the body with
structuralfeatures(basicpulse,beat,orrhythm)inmusiccanbeinterpretedasanon‐inhibited
motorre‐synthesisoftheperceivedsonicpatternsinthemusic.Itallowsthelistenertofeeland
corporeally understand a basis structuring element in themusic. In otherwords, it offers the
listener the possibility to ‘get into the music’. Corporeal attuning or harmonizing with music
impliesanactivemotoractivity, involvingcontrol,short‐termpredictionandanengagementin
goal‐directed or intentional processes. In corporeal attuning with music the subject’s body is
alignedwiththemusic’ssonicforms.Thisalignmentoffersanopportunityformirroring,thatis,
for experiencing theworld from the viewpoint of themusic’s sonic forms (melody, harmony).
Corporeal attuning thus provides a perfect context for empathy. Yet it should be noted that
attuning is mainly involving the motor system, and not necessarily the emotional system. In
emotionalempathy,thesubjectentersastageofbeingfullytakenbythemusic,includingboththe
motor and the emotional system. Music listening normally implies a straigthforward process
from synchronization with the music to emotional understanding of the music, without
experiencing a gapbetweenmotor and emotional system.Hence splittingup this process into
discretestepsisartificial.However,inthepresentstudyweaimatinvestigatingwhetherpeople
withASDcansuccesfullyexecutethedifferentstepsofthisprocess,thusisolatingeachstepinthe
empathicprocess,andwhethertheycanestablishadirectlinkbetweenembodiment,cognition
andempathywithmusic.
Intheresearchpresentedhere,therelationshipbetweensubjectswithASDandmusicisstudied
at three levels of empathic involvement, namely, synchronization, corporeal attuning and
emotional empathy. Both synchronization and attuning are studied from the viewpoint of
embodied listening, using kinematic devices for measurement. In contrast, the study on
emotional empathy is a classic study involving verbal reporting. In the first and second
experiment, subjectswereasked to tapanddrawalongwith thebeatof themusic.These two
experimentstestedtheirabilityofsynchronization.Inathirdexperiment,subjectswereaskedto
drawalongwiththemelodiccontourofthemusic,usinganelectronicpenandtabletequipment,
testing their ability of corporeal attuning. In the fourth experiment subjects had to describe
musical excerpts using affect‐related adjectives. This task requires insight in the emotional
content embedded in the music and thus investigates the ability of emotional empathy. Each
subjectinthisstudycompletedthefourexperimentswithinatimespanoftwoweeks.
Experiments
1.Synchronization
Twoexperimentsaimat investigatingtheabilityofpeoplewithASDtosynchronize theirbody
movementstothemusicalbeat.Thefirstexperimentisastudyontappingthebeat,thesecond
on drawing the beat. The studies differ in that tapping ismonitored using a discretemethod
where each tap represents a discretemoment in time. Instead, drawing ismonitored using a
kinematicdevicethatmeasurescontinuousmovements.Thedrawingconditionallowsastudyof
implicitattuningeffects,thatis,ofcertainbehavioraladaptationstothecharacterofthemusic.
1.1Synchronizationbytappingthebeat
Method
ThepresentexperimentisanalmostexactreplicaofanexperimentcarriedoutbyMcKinney&
Moelants(22)withtypicallydevelopingsubjects.Aselectionofmusicalexcerptswasmadeoutof
thelistofexcerptsusedintheexperimentofMcKinney&Moelants(22),selectingthoseexcerpts
with anunambiguousbeatperceptionby themajorityof the typicallydevelopingparticipants.
TheirresultswereusedascontrolgroupfortheresultsoftheclinicalgroupofsubjectswithASD.
Participants Theclinicalgroupwascomposedof11adultparticipants,allpatientsinacentre
specialised in treatment and guidance of people officially diagnosedwithASD. The groupwas
composedof8menand3women(meanage=30)andwithnormaltohighIQ.Onlysubjectswith
adiagnosisofASD,not combinedwithanyotherdisorder (e.g.ADHD, ...)were included in the
experiment.Allsubjectsweretestedindividuallyandintheirfamiliarsurroundings.
Materialsandstimuli 20 musical excerpts, with genre ranging from classical to pop, film
music,metal(AppendixA)anddurationof30seconds,wereusedinthisexperiment.Allexcerpts
were stored in .wav‐format and the intensitywas normalized in order to avoid differences in
loudness.
Procedure Participantswereaskedtotapalongwiththebeatofthemusicbypressingthe
spacebaronthekeyboardofthecomputer.Itwasstressedthataregularpulseshouldbetapped,
notamorecomplexrhythmicalpattern.Participantswereaskedtotapsimultaneouslywiththe
music,butwhenneededtheycouldrestarttheexcerptasmanytimesastheybelievednecessary.
Duringtheexperimentparticipantsveryrarelyneededordecidedtotakethissecondchance.
Analysis The tapped beats perminute (BPM)were calculated permusical excerpt and
perparticipant, basedon the calculationof themedianof the tapintervals.These resultswere
then compared to the BPM tapped by the majority of the participants in the experiment of
McKinney&Moelants(22)forthesamemusicalexcerpts.
Results
Results show that a majority (54,5%) of the clinical group succeeded very well in this
synchronizationtask.However,aqualitativeanalysisrevealedthatasmallgroupofsubjectswith
ASD(18,2%)hadgreatdifficultiesinsynchronizingtheirtappingwiththebeatofthemusic.Two
groupscouldbedistinguishedwhohadmoredifficultieswith this synchronization task.A first
group(ASD1,3and11)(27,3%)clearlytriedtosynchronizetheirtappingwiththebeatofthe
excerpts,buttheytappedalittlebittoofastorslow(±20BPM).Thesecondgroup(ASD6and8)
(18,2%)tappedveryirregularly.Apparantly,theywereincapableofsynchronizingtheirtapping
with thebeatof themusic.Duringobservation itwasclear thatASD6wasconstantly thinking,
countingandsearchingforpatternsthatcouldbetapped,suggestingthatsynchronizationwith
the beat is not an intuitive process for this subject. In short, in deviant synchronization two
groupscanbedistinguished,namelyagroupthatclearlytriedtosynchronizewiththeexactbeat
butdidnotsucceedexactlyandagroupthatdidnotsynchronizeatall.Inthelightofthepresent
researchit is interestingtoseethatamajorityof theclinicalgroupiscapableofsynchronizing
withthebeatofthemusicandthusareabletoengagewithmusicatthislevel.
1.2Synchronizationandattuningbydrawing
Method
Participants 38 subjects served as participants in this experiments, 11 subjects diagnosed
with ASD in the clinical group and 27 typically developing subjects in the control group. The
clinicalgroupwascomposedofthesamesubjectsasthepreviousexperiment.Thecontrolgroup
was composed of 16 men and 11 women (mean age = 28,3). All participants were tested
individuallyandintheirfamiliarsurroundings.
Materialsandstimuli Eightmusicalexcerpts,withgenreranging fromclassiscal to jazz,pop
andhouse(AppendixB)anddurationof30to35seconds,servedasstimuliinthisexperiment.
They were selected from a larger set of excerpts based on the annotations of a group of six
musicologists.Thisexpertgroupdeterminedtheperceivedtempoandtheaffectivecharacterof
thestimuli.Theresultingsetconsistedofexcerptsranging fromslowto fast tempi,andwitha
fluent or stuttering character. All excerpts were stored in .wav‐format and the intensity was
normalized.
Procedure Participantswere asked to draw along rhythmicallywith themusic, using an
electronic pen and tablet equipment. Theywere also asked to draw alongwith the flowing or
stutteringcharacterof themusic, forexample, fluidorroundmovements fora flowingrhythm
(legato) and angularmovements for a stuttering rhythm (staccato). The participants heard all
excerptstwotimes.Itwasexpectedthattheparticipantswereabletodrawinanimmediateand
spontaneouswayalongwiththemusic,thusresultingintwodrawingsperexcerpt.
Analysis In a first analysis, based on the actual drawings on the tablet, we studied
whether or not participants were able to attune their drawing to the flowing or stuttering
characterofthemusic.Inasecondanalysis,westudiedthedevelopmentofthedrawingsintime.
In this representation, horizontal and vertical axes were summed up and plotted over time.
Thesedatawereusedforaperiodicityanalysis.TheBPM’sextractedfromthisanalysiswerethen
comparedwithreferenceBPM’sforeachstimulus.
Results
Resultsshowthatbothclinicalgroup(76%)andcontrolgroup(83%)wereabletosynchronize
theirdrawingswith the flowingor stuttering characterof themusic.Aminorityof the clinical
grouphadproblemssynchronizing theirdrawings to the characterof themusic (ASD3,6and
11).ASD6andASD11didnotadapttheirdrawingsatalltothechangingcharacterofthemusic
(Figure1a).ASD3didchangethedrawingmovementsoverdifferentstimuli,buttheywerenotin
accordance with the flowing/stuttering character of the music (Figure 1b). The qualitative
analysisof theactualdrawingsalsorevealed thatapartof theclinicalgroup(27%)andof the
controlgroup(37%)instinctivelyadaptedcertainfeaturesoftheirdrawingstothebold/tender
characterofthemusic,althoughtheexperimenttaskdidnotpromptthemtodothis.
Results also show that amajority of the clinical group (64%) and of the control group (85%)
wereabletosynchronizetheirdrawingswiththebeatofthemusicalexcerpts.Theseresultsare
in complete accordance with the results of the previous tapping experiment. The same
participantsoftheclinicalgrouphavedifficultieswiththesynchronizationtask(ASD1,3,8and
11). ASD6 also had great difficulties with the first rhythmical tapping experiment, but he did
succeedinsynchronizinghisdrawingswiththebeat.Hisdrawingshoweverdorevealhissearch
forpatternsinthemusicandaperseveranceforrepetitivemovement(Figure2).
Discussion
Overall,insynchronizeddrawingtomusicpeoplewithASDdisplaylittleornodifferencewitha
control group of typically developing subjects. In both groups, the majority of the subjects
succeedinsynchronizingwiththebeatofthemusic.Moreover,thestudyrevealsthatpeoplewith
ASD are able to corporeally attune to the flowing/stuttering and bold/tender character of the
music.WecanthusconcludethatpeoplewithASDareabletoengagewithmusicatthislevelof
musicalempathy.
2.Graphicalattuningtomelody
Method
Participants Both clinical group and control group are the same as in the previous
experiment.Allparticipantsweretestedindividuallyandintheirfamiliarsurroundings.
Materialsandstimuli 11monophonicmusicalexcerpts,withgenreranging fromclassical to
jazz, pop and filmmusic (AppendixC) andwith a duration of 30 to 35 seconds,wereused as
stimuli in this experiment.Themonophonicmelodies range fromslowandmoreor less static
melodicevolution tomorecomplexmelodicevolution.Allexcerptswerestored in .wav‐format
andtheintensitywasnormalized.
Procedure Participants were asked to draw along with the melodic line of the unison
musicalexcerpts,usinganelectronicpenandtabletequipment.Allparticipantswereinstructed
tomove the pen alongwith the horizontal axis of the tablet.Moving the pen upwards is then
interpretedasanascendingpitchandmovingthependownwardsasadescendingpitch.Given
thatthevastmajorityoftheparticipantsinbothclinicalandcontrolgroupneverhadanymusical
education, all participantswere given an introduction by the researcher and time to practice.
Duringtheexperiment,participantsheardallexcerptstwotimesandwereaskedtodrawalong
duringeachpresentationoftheexcerpt,resultingintwodrawingsperexcerpt.
Analysis Aseriesofplotsaimedatvisualizingthedevelopmentofthedrawingovertime.
Thesedrawingswerethencomparedwiththemelodiccontouroftheexcerpts.Thiscontourwas
extractedfromtheaudiosignal,usingtheMAMIMelodyTranscriber(23).Thesubject’sdrawings
were then compared to the extracted melodic contour. This was done manually and on a
qualitativeratherthanquantitativebasis.
Results
Analysisshowsthatbothclinicalgroupandcontrolgroupcanbedividedintothreesubgroups,
namelyagroupthatattunedtothemelodyinaverypreciseway,asecondgroupthatattunednot
verypreciselyandathirdgroupthatattunedinaverygeneralway(Figure3).
Acomparisonbetweentheclinicalgroupandthecontrolgrouprevealsthatahigherpercentage
ofpeoplewithASDattunesinaveryprecisewaytothemelodiclineofthemusicalexcerpts.In
thecontrolgroup,mostofthesubjectsattuneinaverygeneralwaytothemelodicline(Table1).
Discussion
ThiscorporealattuningexperimentshowsthatthemajorityoftheASD‐subjectsareverycapable
toattunetheirmovementswiththemelodiccontourofthemusicalexcerpts.Comparisonwitha
control group even shows that the subjects with ASD outperform the subjects in the control
group.Theseresultsareinaccordancewithresultsofpreviousstudies(6,9)andcanbeexplained
by theweakurge for central coherence that is characteristic forpeoplewithASD. In theweak
central coherence theory (24), it is assumed that peoplewithASDhave a cognitive stylewhich
biasesprocessinginfavouroflocalfeaturesandpredictsenhancedperformanceontaskswhere
good featural processing conveys an advantage, such as in the present attuning task. This
experimentshowsthatpeoplewithASDtendtodisplaymoreattentiontothedifferentnotesof
themelody,ratherthantothemelodiccontourwhenlisteningtomusic.
3.Attributionofaffectivequalities
The findings of the previous experiment raise the question whether a particular focus on
structural features (rhythm, pitches, a.o.), as seems to be typical for people with ASD, would
implya restriction toa low‐level comprehensionof themusicalexpressionor themusical rich
emotionalcontent.
Method
Participants This experiment is an almost exact replica of an experiment carried out by
Lesaffre(25)withtypicallydevelopingparticipants.Resultsfromthelatterexperimentareusedas
controlfortheresultsfromtheclinicalgrouptestedinthisexperiment.
The clinical group is identical to the clinical group in the previous experiments. The subjects
wereagaintestedindividuallyandintheirfamiliarsurroundings.
Materialsandstimuli 20musicalexcerptswereselectedfromthe160musicalexcerptsused
in the experiment of Lesaffre (25) (Appendix D). Preference was given to those excerpts with
unanimityofaffectiveappreciationamongthetypicallydevelopingparticipants.Allexcerptshad
adurationof30seconds,wereconvertedto.wav‐formatandtheintensitywasnormalized.
Procedure Usingpaperquestionnaires,participantswereaskedtodescribethe20musical
excerpts using affect‐related adjectives. Per excerpt seven unipolar affect‐related adjectives
(anxious, passionate, sad, aggressive, cheerful, careless, tender) had to be evaluated on a five‐
pointscale,andfourbipolaradjectives(hard‐soft,slow‐fast,dynamic‐static, flowing‐stuttering)
hadtobeevaluatedonascalefrom‐4to4.Tolimittheimpactofabilityoflanguageprocessing,
the affect‐related adjectives were visualized using photographs of facial expressions (26). The
recognitionoftheaffectivefacialexpressionsbytheparticipantswastestedbeforeexecutionof
thisexperiment.Theexperimenthadnotimelimitations.
Results
Resultsshowthat theclinicalgroup inthisexperimentappreciatedthemusicalexcerpts inthe
samewayasthetypicallydevelopingsubjects intheexperimentofLesaffre (25).Analysisofthe
calculated correlations for the data set of each participant also show that there is great
consistencyintheanswersoftheparticipants.WecanthusconcludethatpeoplewithASDhave
theabilitytoperceiveaffectinmusicandcanengageinanempathicrelationshipwithmusic.
However,sincetheexperimentdidnothaveanytimelimitations,thedurationoftheexperiment
isalsoanimportantfactortotakeintoaccount.IntheexperimentofLesaffre(25)participantshad
to appreciate160musical excerptsduring four sessionswith40musical excerptsper session.
Dependingontheamountoftimestheparticipantslistenedtoeachmusicalexcerpt,theduration
oftheexperimentforonesessionof40stimulivariedbetween60and120minutes.Themean
time taken to evaluate each excerpt is about two minutes. In the present experiment with a
clinicalgroupofpeoplewithASD,only20musicalexcerptswereoffered.Fortheclinicalgroup,
thedurationof theexperimentvariedbetween140and280minutes (executed in two to four
differentsessions).Inthemean,eachexcerpttookabout10,5minutestoevaluate.
Discussion
ThisexperimentshowsthatpeoplewithASDareabletoassignaffect‐relatedadjectivestomusic,
buttheyneedalotmoretimeandpromptingtocompletethistaskincomparisonwithtypically
developingsubjects.ThepresentfindingsareinaccordancewiththeresultsofHeatonetal.(6)in
which itwas shown that childrenwithASD showednodeficits in processing affect inmusical
stimuli.However,whenwetake intoaccountthedurationof theexperiment,averysignificant
differenceappears.Typicallydevelopingsubjectsareable toevaluate theaffectivecharacterof
musicveryfast,whereassubjectswithASDseemtoreason,evaluateandcompare.Itisafifthey
rely less on corporeal imitation of emotions but more on observation and associated
disembodiedprocessing.Intypicallydevelopingindividuals,themotoractivity,synchronization
and imitation,offersastraigthforwardandspontaneousaccesstoget intothemusic. Inpeople
with ASD, it seems as if this link between corporeal articulation and empathy is less clearly
established. InsteadtheyrelyuponcognitivereasoningandthisexplainswhypeoplewithASD
needalotmoretimeforthistask.
When specifically asked, peoplewithASDare capable of recognizing and attributing affects to
music, they are able to theorize about affective features in themusic, but they clearly rely on
disembodied,cognitiveprocessestocompletethistask.SinceparticipantswithASDarecapable
ofmirroringmusical affects into corporeal articulations but lack a straightforward process to
social cognition, a distinction should be made between a corporeal/embodied engagement –
kinesictheoryofmind–withthemusicontheonehandandemotionalunderstanding–semantic
theoryofmind–ofthemusicontheotherhand(4).Thefindingsofthepresentstudiesshowthat
peoplewithASDarecapableofkinesic theoryofmind, since theyareable to synchronizeand
attunetheirbodywithstructuralandaffectivefeaturesofthemusic,butthisdoesnotenablea
directlinkwithemotionalunderstandingoftheseaffectivefeaturesinthemusic.
Conclusion
Fourexperiments,aimedatinvestigatingthedegreeinwhichpeoplewithASDareabletoengage
withmusic,werecarriedout.Basedontheideathatcoherentperceptionofmusicandemotional
involvementwithmusicispossiblethroughcorporealmediation,theconceptofmusicalempathy
was divided into several gradations, ranging from synchronization and attuning to emotional
involvement.
Theresultsshowthatinsynchronizedtappinganddrawingtomusic,subjectswithASDdisplay
little or no difference with a control group of typically developing subjects. The study also
revealedthatpeoplewithASDareabletocorporeallyattunetheirdrawing,bothinacontrolled
andspontaneousway,toaffectivefeaturesinthemusic.ThemajorityofpeoplewithASDattune
theirmovementswith themelodic contour of themusical excerpts and they outperform their
non‐autistic controlsbymoving inamorepreciseway inaccordwith themelody.Finally, ina
study on verbal description of musical affects we found that people with ASD are capable of
attributing affective qualities tomusic, but they needed significantlymore time than typically
developingcontrols.
The present research shows that people with ASD have a corporeal understanding of the
affectivefeaturesofmusic,sincetheyareabletomirrorstructuralandevenaffectivefeaturesof
the music into corporeal articulations. However, when musical empathy involves emotional
processing,wefoundcleardifferenceswithtypicallydevelopingcontrols.PeoplewithASDseem
to rely on disembodied cognitive processes to attribute affects to music, indicating that their
corporealunderstandingdoesnotgivethemastraightforwardaccesstotheemotionalcontentof
themusic.Furtherstudieswithlargersamplesizesshouldbeconductedtoelaboratefurtheron
thisconcept.
Thepresentstudyshowsthatmusiccanbeusedasatooltodiagnosecertainaspectsofempathic
involvement. Research on musical empathy in people with ASD offers the outline of a new,
comprehensive approach to a better understanding, and possibly diagnosis and treatment of
people with this pervasive social disorder and can also offer a more profound insight in the
processes that involve processing of affective musical features and hence contribute to the
domain of music therapy. This embodied approach to musical empathy could lead to a new
methodologyinmusictherapy.Bytrainingtheembodiedunderstandingofaffectivefeaturesin
music, and clarifying and training the link between this embodiment and the semantic
understanding of the emotions in the music, music therapy may establish and enhance the
automatedprocessofsynchronizationtoemotionalempathywithmusicinsubjectswithASD.
Acknowledgements
Theauthorswish to thanktheFondsvoorWetenschappelijkOnderzoekVlaanderen(FWO) for
fundingthisresearch.Wewould liketothankFrankDesmet,MichelineLesaffre,DirkMoelants
andFrederikStynsfortheirhelpinthesetupandanalysisoftheexperiments.Wealsowantto
thankChristelEmbrechtsofthemusictherapydepartmentofvzwCaritasandthepsychologists
andparticipantsofvzwHetAnker.
References
1 DeNoraT. (2001). Aesthetic agency andmusical practice:Newdirections in the sociology of
musicandemotion.InP.N.Juslin&J.A.Sloboda(Eds.)MusicandEmotion:TheoryandResearch
(pp.161‐180).Oxford:OxfordUniversityPress.
2ThompsonW.F.(2009).Music,thought,andfeeling:Understandingthepsychologyofmusic(pp.
141‐144,253).Oxford:OxfordUniversityPress.
3 JuslinP.N. (2009).Emotionalresponses tomusic. InS.Hallam, I.Cross&M.Thaut(Eds.)The
Oxfordhandbookofmusicpsychology(pp.131‐140).Oxford:OxfordUniversityPress.
4LemanM.(2007).Embodiedmusiccognitionandmediationtechnology.Cambridge,MA:TheMIT
Press.
5ApplebaumE.,EgelA.,KoegelR.,andImhoffB. (1979).Measuringmusicalabilitiesofautistic
children.JAutismDevDisord,9(4),279‐285.
6HeatonP.,HermelinB.,andPringL.(1998).Autismandpitchprocessing:Aprecursorforsavant
musicalability?MusicPercept,15(3),291‐305.
7 Heaton P., Pring L., andHermelin B. (2001).Musical processing in high functioning children
withautism.BiolfoundMusic,930,443‐444.
8HeatonP. (2005). Intervalandcontourprocessing inautism. JAutismDevDisord,35(6),787‐
793.
9BonnelA.,MottronL.,PeretzI.,TrudelM.,GallunE.,andBonnelA.‐M.(2003).Enhancedpitch
sensitivity in individualswithautism:A signaldetectionanalysis. J CognNeurosci, 15(2),226‐
235.
10 Heaton P., Hermelin B., and Pring L. (1999). Can childrenwith autistic spectrum disorders
perceiveaffectinmusic?Anexperimentalinvestigation.PsycholMed,29(6),1405‐1410.
11CappsL.,YirimiyaN.,andSigmanM.(1992).Understandingofsimpleandcomplexemotionsin
non‐retardedchildrenwithautism.JChildPsycholPsychiatry,33,1169‐1182.
12 Baron‐Cohen S., Tager‐Flusberg H., and Cohen D. (1993). Understanding others’ minds:
Perspectivesfromautism.Oxford:OxfordUniversityPress.
13 Niedenthal P.M., Augustinova M., and Rychlowska M. (2010). Body and Mind: Zajonc’s
(re)introductionofthemotorsystemtoemotionandcognition.EmotionRev,2(4),340‐346.
14 Carr L., Iacoboni M., Mazziotta J., and Lenzi G. (2003). Neural mechanisms of empathy in
humans:A relay fromneural systems for imitation to limbicareas.Proceedingsof theNational
AcademyofSciencesoftheUnitedStatesofAmerica,100(9),5497‐5502.
15LeslieK.,Johnson‐FreyS.,andGraftonS.(2004).Functionalimagingoffaceandhandimitation:
towardsamotortheoryofempathy.Neurol.,21(2),601‐607.
16DesainP.,andWindsorW.(2000).Rhythmperceptionandproduction.Lisse:Swets&Zeitlinger.
17ThautM.(2005).Rhythm,musicandthebrain:Scientific foundationsandclinicalapplications.
NewYork:Routledge.
18FraisseP.(1963).Thepsychologyoftime.Westport,CT:GreenwoodPress.
19MichonJ.(1967).Timingintemporaltracking.Assen,TheNetherlands:VanGorcum.
20 Schachner A., Brady T.F., Pepperberg I.M., Hauser M.D. (2009). Spontaneous motor
entrainmenttomusicinmultiplevocalmimickingspecies.CurrBiol,19,831‐836.
21LargeE.W.(2000).Onsynchronizingmovementstomusic.HumMovementSci,19,527‐566.
22McKinneyF.,andMoelantsD.(2006).Ambiguityintempoperception:Whatdrawslistenersto
differentmetricallevels?MusicPercept,24(2),155‐166.
23 De Mulder T., Martens J., Lesaffre M., Leman M., De Baets B., and De Meyer H. (2003). An
auditory model based transcriber of vocal queries. In Proceedings of the fourth international
conferenceonmusicinformationretrieval(ISMIR2003),245‐246.
24FrithU.(1989).Autism:Explainingtheenigma.Blackwell.
25 Lesaffre M. (2005).Music information retrieval: Conceptual framework, annotation and user
behaviour.Unpublisheddoctoraldissertation,GhentUniversity.
26EkmanP.(1992).Anargumentforbasicemotions.CognitionEmotion,6,169‐200.
Correspondingaddress
LeenDeBruyn,GhentUniversity,DepartmentofMusicology(IPEM),Blandijnberg2,9000Gent,
Belgium,Email:[email protected]
Tables
Subgroups ASD CG
general 18% 44%
precise 27% 19%
veryprecise 55% 37%
Table1: PercentagesofASD(clinicalgroup)andCG(controlgroup)thatattune
tomelodyinageneral,preciseandveryprecisewaytothemelodicline.
Figures
Figure1: (a)Rhythmical drawings ofASD6 for the bold and stuttering excerpt 3 (Nitro
doll – Rocket City Riot) and the tender and flowing excerpt 4 (Pavan ‘Lord
Salisbury’–OrlandoGibbons).(b)RhythmicaldrawingsofASD3forexcerpts3
and4.
Figure2: Rhythmical drawing of ASD6 for excerpt 2, showing a perseverance for a
repetitivepatterninthemusic.
Figure3: Visual representations of the graphical attuning to the melodic contour for
excerpt1.Inthecolumnsyouseeacomparisonofsubjectsintheclinicalgroup
(ASD)andthecontrolgroup(CG).Intherowsyouseeacomparisonofsubjects
thatattunedtothemelodyinaveryexact,exactandgeneralway.
Appendix
ExcerptNumber Title Performer/Composer BPM
Excerpt1 Lavalsed’Amélie YannTiersen 49
Excerpt2 AoiHitomi MaayaSakamoto 52
Excerpt3 Zij MarcoBorsato 56
Excerpt4 Artichoke CiboMatto 60
Excerpt5 Devilgotmywoman SkipJames 80
Excerpt6 Crockett’sTheme JanHammer 96
Excerpt7 Inch’Allah McSolaar 103
Excerpt8 Beencaughtstealing Jane’saddiction 104
Excerpt9 Respect OtisRedding 114
Excerpt10 Highwaytohell AC/DC 115
Excerpt11 Azzuro AdrianoCelentano 118
Excerpt12 Comedi PaoloConte 130
Excerpt13 Threeo’three Publicenergy 140
Excerpt14 TheMessage Zolex 140
Excerpt15 Cometakemyhand 2brothersonthe4thfloor 168
Excerpt16 I’lltakeyoualltofuckinghell Scarface 170
Excerpt17 Torremolinos Sttellla 176
Excerpt18 Sabeldans AramKhachaturian 178
Excerpt19 SilverRocket SonicYouth 185
Excerpt20 Elgatolopez Ska‐P 220
AppendixA: Listofstimuliusedinthesynchronizationbytappingexperiment.
Excerpt
Number
Title Performer/Composer Character BPM
Excerpt1 Triosonateop.2,no.
2ing‐minor
(allegro)
GeorgFriedrichHändel Stuttering 94
Excerpt2 Disconnected Emma’sMini Flowing 88
Excerpt3 Nitrodoll RocketCityRiot Stuttering 193
Excerpt4 Pavan“Lord
Salisbury”
OrlandoGibbons Flowing 77
Excerpt5 TranceOdyssey DJMarkitos Stuttering 150
Excerpt6 Heynowbabus TomPaul Flowing 83
Excerpt7 MaloSelo Kitka Flowing 54
Excerpt8 Einfach,
Davidsbündlertänze
op.6,no.11
RobertSchumann Flowing 91
AppendixB: Listofstimuliusedinthesynchronizationbydrawingexperiment.
ExcerptNumber Title Performer/Composer
Excerpt1 Oquammirabilisest HildegardvonBingen
Excerpt2 SequenzaIforflutesolo LucianoBerio
Excerpt3 Alleluia Chantdel’égliseMilanese
Excerpt4 SoloLuc OllaVogala
Excerpt5 LaDispute YannTiersen
Excerpt6 Saxophonesolo /
Excerpt7 Flutesolo /
Excerpt8 Wampumprayer ToriAmos
Excerpt9 Violinandrecorder /
Excerpt10 Double‐basssolo /
Excerpt11 LeMoulin YannTiersen
AppendixC: Listofstimuliusedinthemelodicattuningexperiment.
ExcerptNumber Title Performer/Composer
Excerpt1 Ah,vitabella LucillaGeleazzi
Excerpt2 LeSacreduPrintemps IgorStravinski
Excerpt3 Concertoforviolinandorchestra,Canzonetta PjotrTchaikovski
Excerpt4 St.Anger Metallica
Excerpt5 Therobots Kraftwerk
Excerpt6 Manymeetings HowardShore
Excerpt7 PianoconcertoNr.1inc,op.18,3rdmovement SergeiRachmaninov
Excerpt8 Achterbank DeNieuweSnaar
Excerpt9 TearsinHeaven EricClapton
Excerpt10 Tame ThePixies
Excerpt11 Traffic DJTiesto
Excerpt12 OneStep TamEchoTam
Excerpt13 Atisketatasket EllaFitzgerald
Excerpt14 MessadeRequiem,DiesIrae GuiseppeVerdi
Excerpt15 DieWalküre,Vorspiel RichardWagner
Excerpt16 GymnopédieNo.1 EricSatie
Excerpt17 MyWorld Metallica
Excerpt18 Fromthecradletoenslave CradleofFilth
Excerpt19 Libertango AstorPiazolla
Excerpt20 TicTicTac Carrapicho
AppendixD: Listofstimuliusedintheaffectattributionexperiment.