ChapterPDF Available

Cognitive Approaches to Analysis of Emotions in Music Listening


Abstract and Figures

In recent years research into music cognition and perception has increasingly gained territory. A fact which is not always realised by music theorists is that, from the perspective of cognitive psychology and empirical methodology, the representatives of the expanding field of cognitive music research frequently address questions and propose theoretical frameworks that ought to have implications for music theory of a more traditional kind. Yet, such cognitive theories and empirical findings have not had radical impact on general analytical practice and teaching of music theory. For theorists interested in musical meaning the emotional impact of music has always been a major concern. In this paper I will explore how multiple cognitive theories and empirical findings can be applied to account for emotional response to three subjectively chosen excerpts of strongly emotion-inducing music: Namely Penderecki’s ‘pain-inducing’ Threnody to the Victims of Hiroshima (1959-61), Wagner’s ‘weepie’ Prelude to Act II from Tristan und Isolde (1859), and the opening bars of Chopin’s ‘shocking’ Scherzo no. 2 (1837). Although ambitious multiple-mechanisms theories have recently been proposed by e.g. Huron (2006) and Juslin & Västfjäll (2008), we still lack a complete and all-embracing theory of musical emotions, and none of the existing ones actually reaches a level of methodological specificity rendering it directly and unambiguously applicable to specific musical scores and recordings. This is an area where music theorists can be instrumental in bridging the gap between cognitive music research and music analysis.
Content may be subject to copyright.
For music theorists with an interest in musical meaning, a natural
way of engaging with the music listening process analytically is to as-
sess the emotional experience resulting from listening to music. This
line of research was already proposed more than 50 years ago by mu-
sicologist Leonard B. Meyer (1918-2007) in his seminal book Emo-
tion and Meaning in Music.1 Amongst other things, Meyer argued that
insights from Gestalt psychology and information theory as well as
the notion of expectations arising from prior stylistic familiarity can
be applied successfully to account for musical narratives. His work
instigated a closer, mutually beneficial relation between music theory
and psychology and has been a source of great inspiration to subse-
quent generations of music scholars2 and music psychologists.3 In the
present climate, research in musical emotions takes place within and
across a multitude of academic fields and disciplines; some examples
are music therapy,4 advertising research,5 film music research,6 music
history,7 aesthetics and philosophical psychology,8 music technology
and engineering,9 and cognitive neuroscience.10
Despite the cross-disciplinary nature inherent to Meyer’s endeav-
ours, the extent to which research findings from specific fields are ad-
equately adopted to inform the work of scholars and scientists working
within other domains may, nevertheless, be challenged. In particular, it
can be argued that research in music cognition has not had noteworthy
influence on the ways in which we teach music theory and analyse mu-
sical pieces. One indication hereof is the fact that empirical research is
hardly ever cited by music theorists and analysts; consequently, such
findings remain largely unknown to the current students and future
scholars of music theory and analysis.
In recent years this topic has been addressed in various ways. For
instance, Hansen discussed the legacy of Lerdahl and Jackendoff’s
Generative Theory of Tonal Music (GTTM) as a manifestation of the
“cognitive revolution” within the field of music theory.11 Narmour, fur-
thermore, speculated that the lack of interaction between music theory
and experimental psychology can be explained by the observation that
the Society for Music Theory—primarily due to the overrepresentation
of composers in this organisation at the time—missed an important mo-
ment in the 1980s by overlooking music psychology and refraining from
encouraging an interest in it.12 This retrospective perspective searching
for historical explanations for the disciplinary divide does, however,
somewhat transcend the scope of the present paper which takes a more
future-oriented direction by pointing out possible ways in which this
divide can be crossed—or possibly even bridged.
More specifically, this paper aims to exemplify how experimental
results can inform music-analytical practice and may eventually in turn
have an impact on the teaching of music theory. For this purpose, three
excerpts of music by Krzysztof Penderecki, Richard Wagner and Fré-
déric Chopin with a strong emotional impact on this author have been
chosen on a purely subjective basis. Empirical findings as well as recent
theories relating to musical emotions—particularly those of Huron13 and
Juslin and Västfjäll14—will be applied to explain emotional experience.
Theoretical frameworks for understanding musical
A few years ago, Patrik N. Juslin and Daniel Västfjäll proposed a
Multiple Mechanisms Theory of musical emotions, which they substan-
tiated with empirical findings and summarised into a number of em-
pirically testable hypotheses.15 According to the authors, the primary
motivation for this theory was that “research on music and emotions
has failed to become cumulative because music researchers have ei-
ther neglected underlying psychological mechanisms or assumed that
musical emotions reflect a cognitive appraisal.”16 Their solution was to
propose six independent mechanisms through which musical emotions
can be induced in the listener. These mechanisms are thought to have
developed through the course of evolution approximately in the order
listed and summarised below:
(1) Brain stem reflexes represent hard-wired brain networks respond-
ing—primarily pre-attentively—to sudden, loud and dissonant
sounds. This may also be described in terms of a situation where
“an emotion is induced by music because one or more fundamen-
tal acoustical characteristics of the music are taken by the brain
stem to signal a potentially important and urgent event.”
(2) Evaluative conditioning characterises situations where “an emo-
tion is induced by a piece of music simply because this stimu-
lus has been paired repeatedly with other positive or negative
(3) Emotional contagion is a cross-modal mechanism resulting in
emotional experience from facial expressions, speech voices
and expressive music alike; in the last case, “because the listener
perceives the emotional expression of the music, and then ‘mim-
ics’ this expression internally.”
(4) Visual imagery represents a mechanism where emotions result
when a listener “conjures up visual images (e.g., of a beautiful
landscape) while listening to the music.”
(5) Episodic memory embodies situations where “an emotion is in-
duced in a listener because the music evokes a memory of a
particular event in the listener’s life.”
(6) Musical expectancy leads to emotional experience “because a
specific feature of the music violates, delays, or confirms the
listener’s expectations about the continuation of the music.”17
For each of these mechanisms, Juslin and Västfjäll made an attempt
to formulate testable hypotheses about how they can be distinguished
experimentally from one another. For this purpose, they identified 11
characteristics that are thought to differ between the mechanisms: more
specifically, (1) their evolutionary survival value, (2) the type of infor-
mation processed, (3) their place in human ontogenetic development,
(4) the key brain regions involved, (5) the extent to which they are in-
nate or culturally dependent, (6) what particular affects they induce, (7)
the speed with which induction takes place, (8) the extent of volitional
control, (9) their availability to consciousness, (10) their interdepen-
dence in relation to other psychological processes, and (11) the extent
to which induction varies with musical structure. In addition, Juslin
and Västfjäll tentatively explained the experience of complex, mixed
emotions in terms of co-activation of more than one of their proposed
Reviews of Juslin and Västfjäll’s theory have, amongst other
things, emphasised the primary importance of expectancy in musical
emotions18 and some have furthermore proposed that the mechanisms
be “organized hierarchically with musical anticipation as its guiding
mechanism.”19 Another major point of criticism relates to the authors’
tendency to exaggerate cognitive processes that are not specific to mu-
sic.20 Ultimately, this risks implying that “if these non-musical media-
tors (images, memories, associations) were to be kept constant, there
would be no effect of music on emotion.”21
The criticism regarding lacking reliance on music-specific observa-
tions does by no means apply to another important account of musical
emotions, namely David Huron’s ITPRA Theory.22 Huron takes music
as his very starting point for developing a general psychological theory
offering a more elaborate framework for understanding specifically the
expectancy dimension of musical emotions. Evidently reminiscent of
Meyer’s theories, Huron’s theory is based on the notion that musical
expectations result from internalised probabilistic features in our sen-
sory input. He skilfully applies this principle to provide insights into
important topics within music theory, history and composition relating
to, for instance, meter, syncopation, cadence, tonality and the concepts
of musical “works” and “genres.” The underlying tenet that statistical
learning is significant to musical listening has indeed been demonstrat-
ed experimentally using pitched tone sequences23 as well as more natu-
ralistic musical stimuli.24
According to Huron, five response systems underlie expectation-
elicited emotional responses to music:
(1) Imagination response: Emotions arise from imagining future
outcomes. This system motivates us to take steps to achieve
pleasant situations and avoid unpleasant ones.
(2) Tension response: Emotions result from physiological changes
accompanying heightened arousal (motor preparation) and/or
attention (perceptual preparation) in preparation for a particu-
lar expected event. Such emotions are often negatively valenced
and thus motivate us to avoid wasting energy unnecessarily.
(3) Prediction response: Correct and incorrect predictions produce
positively, respectively negatively, valenced emotions. This sys-
tem motivates us to optimise our internal models of the world
through learning and thus to make optimally accurate predic-
tions about the future.
(4) Reaction response: Outcomes are automatically and pre-atten-
tively assessed by a conservative response system which is ex-
empt from habituation and (un-)learning, always assumes the
worst-case scenario and therefore usually produces negatively
valenced emotions.
(5) Appraisal response: Finally, emotions arise from conscious ret-
rospective evaluation of the implications associated with a given
outcome. This response system takes account of complex social
and situational factors and serves the purpose of reinforcing or
suppressing certain behaviours.
To illuminate the workings of this theory, we could consider a music
lover attending a concert performance of Edvard Grieg’s well-known
Piano Concerto in A-minor, op. 16 (1868). Before going to the concert
hall, this person might experience pleasure from imagining the forth-
coming performance. This positive experience could, for instance, be
based on previous concerts attended and recordings listened to and may
very well extend several days or weeks retrospectively prior to the ac-
tual concert. Perhaps this emotion even motivated our music lover to
purchase her ticket in the first place. Then, when seated in the concert
hall, the silence following the applause after the conductor’s entrance
on stage would most probably evoke a tension response in our music
lover. This effect increases when the initial timpani roll commences, and
the underlying crescendo only intensifies this experience. Being highly
familiar with this piece, our music lover knows very well what will hap-
pen, and this will ensure a positively valenced prediction response when
she hears the pianist playing the first A-minor chord. Nevertheless, she
does not know exactly when this chord will sound. The tension response
thus serves the purpose of preparing her for the unavoidably negative
reaction response that will automatically follow the sforzando chord. On
the other hand, she will subsequently appraise that sudden piano chords
are not harmful at all, and on the longer term she might even look back
at this event as an emotionally loaded memory that could motivate her to
pursue similarly pleasurable experiences later on in her life.
As evident from this example, whereas imagination and tension re-
sponses happen prior to the onset of an event, the prediction, reaction
and appraisal systems represent different types of responses succeeding
the actual outcome. This contributes a sequentially structured, temporal
dimension to the explanation of musical emotions which is largely ab-
sent from Juslin and Västfjäll’s Multiple Mechanisms Theory.
In Huron’s framework, interactions between the five ITPRA mecha-
nisms take place; in fact, the contrastive valence resulting from incon-
gruity between multiple response systems is argued to account for com-
plex emotional phenomena such as “bittersweet” or “mixed” feelings.
As the example above showed, and as the analysis below will confirm,
emotions are sometimes experienced as even stronger when different
mechanisms predict contrastive valence.
Painful Penderecki: Threnody to the Victims of Hiroshima
The first example of emotion-inducing music is Threnody to the
Victims of Hirsohima (1959-61) by Krzysztof Penderecki (b. 1933).
It was composed for 52 string instruments in commemoration of the
atomic bomb over Hiroshima towards the end of World War II. Argu-
ably, this is one of the scariest and most unpleasant pieces of music, and
it often gives the author of this paper a lump in the throat when listening
to it. Having been used as stimuli in psychological research to represent
“threatening music,”25 “unpleasant music,”26 “horrific music,”27, 28 and
“agitation/anger, fear, surprise,”29 the characteristic emotional impact
of this piece appears to be shared by many listeners. The subsequent
musical analysis will throw further light upon which mechanisms un-
derlie this experience.
Obviously, Penderecki’s associative titling capitalises on the fear
of nuclear war which was dominant in the early 1960s; indeed, this
was not long before the Cuban Missile Crisis in 1962. Although most
people were not direct victims of the Hiroshima tragedy, they are very
likely to have strong memories associated with receiving this news or
perhaps learning about it in history classes; and as argued by Juslin
and Västfjäll, episodic memory can indeed be instrumental in emotional
experience to music.
In a study with Western listeners, Balkwill and Thompson found that
string instruments are more successful than flutes in communicating
“anger.”30 Similarly, Behrens and Green reported that the violin is better
than timpani, trumpet and voice in conveying the emotion “scared” in
improvised solo performance.31 The idiomatic playing of vibrato, glis-
sando and eerie, high pitches on string instruments might have had at
least some influence on Penderecki’s choice of instruments.
In fact, the characteristic sounds used throughout Threnody had just
been used one year earlier to evoke horror during the shower stabbing
scene in Bernard Herrmann‘s score for Alfred Hitchcock’s movie Psy-
cho (1960). Thus, a simple conditioned association between the visual
scene and the accompanying music is created.32 Such associations are
likely to extend beyond this particular movie and beyond the horror
genre to other domains including Penderecki’s instrumental music. Jus-
lin and Västfjäll’s evaluative conditioning mechanism offers a suitable
framework for understanding the emotional impact of the instrument-
specific sounds used in this piece.
Furthermore, a number of structural characteristics result in low
predictability and thus in difficulties in forming expectations about how
the music will continue. First of all, fragmented textures (see e.g. pp.
13-16 in the score from PWM Edition) cause uncertainty about sound
source localisation. Penderecki also composed an ultimate divisi where
every instrument performs a separate part, unlike the common sym-
phonic texture where the members of each group share parts and, con-
sequently, are fused in the listener’s perception. According to Huron,
successful parsing of the auditory scene is rewarded with an experience
of mild pleasure.33 In that case, it could be argued that inability to do so
may predispose for a negatively valenced emotional response.
In the same passage, bar lines only serve the purpose of facilitating
reading of the score rather than outlining metrical regularities. Various
kinds of tuplets furthermore add to the general low predictability within
the rhythmic dimension.
Research has shown that strong, sudden changes in loudness may
predispose for experience of chills,34, 35 and that noise is particularly
unpleasant when its occurrence is unpredictable or cannot be controlled
by the listener.36 Indeed, the waveform representation in Fig. 1 shows
how dynamic changes appear very suddenly throughout Threnody. This
may cause an emotional response by means of Juslin and Västfjäll’s
brain stem reflex mechanism.
Threnody belongs to a genre for which Polish musicologists have
coined the term “sonorism.”37-39 Whereas most traditional string orches-
tra timbres result from rubbing horse hair (i.e., the bow) against strings
made from steel, the sonorists experimented with various materials
for vibrators (e.g., strings) and inciters (e.g., bow) used for sound pro-
duction.40 In Threnody strings are plucked (pizzicato) and played with
the wooden part of the bow (col legno and legno battuto), the wooden
bridge and tailpiece are bowed, and the sounding board is struck with
fingers or with the nut. Slow glissando effects similarly contribute to a
remarkably unpredictable timbral environment.
Confirmation and violation of listeners’ expectations were impor-
tant to both the theories introduced above. In Huron’s framework, low
predictability with respect to when- and what-related expectations pos-
sibly increases tension response. Since high tension is an unbalanced,
temporary and very stressful state during which vast amounts of energy
Fig. 1. Waveform showing amplitude plotted against time for a
recording of Penderecki’s Threnody to the Victims of Hiroshima
are spent by our organism, it is not surprising that listeners report nega-
tively valenced, fearful and scared emotions when forced to maintain
this state for longer periods like the ten minutes that a performance of
Threnody typically lasts.
Moreover, when listening to Threnody, the listener’s ability to es-
tablish what Huron refers to as “schematic expectations”—based on
knowledge of musical style stored in semantic, long-term memory—
is strongly impeded. Consequently, our brain is not neurochemically
rewarded for making correct predictions about musical structure, and
negatively valenced emotions thus result. This was what Huron referred
to in terms of prediction response.
In addition to emotional experience related to musical expectancy,
Juslin and Västfjäll’s brain stem reflex mechanism also comes into play
owing to different examples of “sensory dissonance.” In the music psy-
chology literature, it is common to distinguish between two separate
causes underlying the subjective experience of dissonance: “musical
dissonance” is determined by acquired cultural norms whereas “sen-
sory dissonance” can be explained psycho-acoustically in the organi-
zation of the peripheral auditory system.41 The latter arises when two
or more complex tones contain many partials with frequencies falling
within a single critical frequency band roughly corresponding to four
semitones. 42 This phenomenon is physiologically reflected in the struc-
ture of the cochlea in the inner ear where each point along the basilar
membrane has a characteristic frequency for which it is most resonant.
If two frequencies are very close to one another, they will be perceived
as having identical pitch. However, if they are further apart—but still
within one critical band, sensory interactions will give rise to the un-
pleasant sensation of sensory dissonance.
In Threnody, frequent alternation between different percussive
sounds (pp. 6-7 in the score) and sul ponticello playing in particular
contribute to sensory dissonance. Unpitched, aperiodic noises or less
clearly pitched notes have indeed been associated with aggression in
animal communication.43 In the case of sul ponticello playing, the string
player takes the bow as near to the bridge as possible. This creates sen-
sory dissonance by amplifying higher harmonics,44 which produces a
thin and glassy sound.
Another prominent feature is Penderecki’s fascination with “white
noise.” This term refers to a signal containing equal power within a
fixed bandwidth. This naturally leads to numerous interactions and thus
to high degrees of sensory dissonance. Compositional approximations
of white noise appear throughout Threnody in terms of quarter note no-
tation, slow vibrato with high frequency difference, ultimate divisi, and
chromatic chords like the final one. This concluding chord, where the
composer assigns individual pitches to all 52 string players comprising
all quarter tones spanning from C3 to C#5, is indeed the densest one in
the whole piece, thus allowing Penderecki’s white noise approximation
to get the very last word.
Empirical findings emphasise the significance of dissonance level
on emotional experience. For instance, using positron emission tomog-
raphy, Blood and colleagues found increased cerebral blood flow in the
right parahippocampal gyrus when listening to dissonant music.45 This
region has previously been associated with unpleasant emotions evoked
by pictures with negative emotional valence.46 Moreover, noise impairs
prefrontal cortical cognitive function in monkeys through a hyper-
dopaminergic mechanism.47 This presumably happens in order to al-
low posterior cortical and sub-cortical structures to regulate behaviour,
probably because the source of auditory sensory input is considered a
potential threat. Presentation of white noise has also been shown to lead
to increases in heart rate48 that differed significantly from those experi-
enced in response to tones.49 Thus, there is reason to believe that sounds
like the ones produced in Penderecki’s Threnody may be directly in-
volved in activation of the sympathetic nervous system which may in
turn result in some of the physiological responses that this author—as
well as other listeners—report when listening to this music.
Weeping with Wagner: Prelude to Act III from
Tristan und Isolde
In the author’s experience, the opening bars from the prelude to
Act III from Richard Wagner’s Tristan und Isolde represents music that
never fails to evoke chills and create a characteristic, sad and solemn
atmosphere (see Fig. 2). This type of emotional reaction is supported by
research suggesting that chills occur more often in connection with sad
than with happy music.50 Physiological changes have similarly been re-
ported in response to sad-sounding music in terms of skin conductance
changes, slower heart rate and increased blood pressure.51 In accounting
for possible underlying mechanisms, contextual as well as structural
factors will be considered.
The dramatic context obviously contributes to musical emotions in
response to opera performances. In this specific case, the two lovers
Tristan and Isolde have just been separated shortly before the overture
to Act III. Tristan has been mortally wounded by his friend Merlot acting
in fidelity to his master, Tristan’s uncle King Marke, who was supposed
to marry Isolde. Subsequently, the famous manifestation of infinite
love, Isoldes Liebestod, follows. In Juslin and Västfjäll’s framework,
emotional contagion is very likely to come into play here; the emotions
expressed by the singers simply affect our own mood state. Addition-
ally, the scenic staging of opera performances automatically creates a
multi-sensory experience. It may be hypothesised that simultaneous
stimulation of visual and auditory senses may additionally stimulate
the visual imagery mechanism proposed by Juslin and Västfjäll.
Finally, in the case of this author, autobiographical episodic memo-
ries from excellent live-performances of this opera most certainly call
Figure 2. The opening bars from the Prelude to Act III from
Wagner’s Tristan und Isolde
to mind previous mood states. Such memories may both influence
imagination responses preceding the music and appraisal responses
following the music; however, the exact nature of this process is likely
to be subject to great individual and contextual variation.
Table 1. Correlations between structural features and sad-sounding
music established by previous research
Table 1 provides an overview of associations, which have been
reported in previous research, between certain structural features and
sad-sounding music.83 The presence of all these features in the Wagner
excerpt is remarkable. More specifically, the violins begin on their low-
est possible note; violas, celli and double basses also approach their
Structural futures Empirical studies
Minor mode Husain, Thompson & Schellenberg (2002, “nega ve
shi in moode“),52 Wedin (1972, “unpleasantness“),53
Hevner (1937)54
Slow tempo Post & Huron (2009),55 Husain, Thompson & Schel-
lenberg (2002, “decreased arousal“)56 Balkwill &
Thomposn (1999),57 Gabrielsson & Juslin (1996),58
Scherer & Oshinsky (1977),59 Rigg (1937, “lamenta-
tion“, “sorrowful longing“),60 Hevner (1937)61
Low pitch Huron (2008),62 Huron, Yim & Chordia (2010),63
Scherer & Oshinsky (1977),64 Wedin (1972,
“unpleasantness“),65 Hevner (1937)66
Narrow pitch range/small
interval size
Balkwill & Thompson (1999),67 Huron (2008)68
Dark od dull mbre Juslin & Laukka (2004),69 Schutz, Huron, Keeton &
Loewer (2008)70
Dissonances Wedin (1972, “unpleasantness“),71 Hevner (1937),72
Rigg (1937)73
Trochaic rhythm Rigg (1937)74
Firm and slow rhythm Wedin (1972, “unpleasantness“)75
Pauses Juslin & Laukka (2004)76
Low or moderatate sound
Gabrielsson & Juslin (1996),77 Turner & Huron
Legato phrasing Gabrielsson & Juslin (1996),79 Rigg (1937)80
Small ar cula on
Juslin & Laukka (2004)81
Slow tone produc on Gabrielsson & Juslin (1996)82
absolute limit; tempo is “mässig langsam;” the key is F-minor; phras-
ing is constantly legato; the melody moves stepwise with a very narrow
ambitus in the first four measures; the first chord has an added sixth; a
double suspension occurs in the violins in bars 2 and 4; a remarkable
pause follows the crescendo in bar 2; and the close spacing in the low
register represented by double stops in the celli create a dull timbre
and results in many adjacent harmonics in the frequency range most
ideal for human perception. Moreover, due to the first and third notes
being much longer than the second and fourth ones as well as the forte-
dynamic on the first note and the slur ending on the fourth note, the
rhythm appears strongly trochaic.
It is, however, essential to note that no simple stimulus-response
pattern can be established between structural features in the music and
a given emotional response. Such associations are by no means causal
or deterministic, but rather correlational or probabilistic.84 Such cultur-
ally acquired associations thus work by means of Juslin and Västfjäll’s
evaluative conditioning mechanism.
As pointed out in a recent review on the pleasures of sad music,
many of the structural features listed in Table 1 are in fact reminiscent
of prosodic cues that have been shown as affective indicators of sadness
in speech.85 More specifically, sad speech tends to be slower,86 quieter,87
and lower in pitch.88 In this perspective, the emotional experience of
sadness caused by listening to this music may in fact result from a kind
of cross-modal mimicking of the sadness expressed by the music itself.
In the context of the Multiple Mechanisms Theory, this process could
be described in terms of emotional contagion.
According to research by Sloboda, melodic appoggiaturas typically
evoke tears.89 The double suspensions in the violins (G and B-flat su-
perimposed on an F-minor chord) in bars 2, 4, 6, 17, 19, and 21 can
be conceived of as incongruities between harmony and melody calling
for a very constrained resolution. In Huron’s framework, high predict-
ability increases the neurochemical reward received for making correct
predictions (i.e. prediction response).
Interestingly, the author’s emotional response tends to be consider-
ably stronger for the second presentation of the main theme (bars 16-
19) in comparison with the first one (bars 1-6). In trying to account
for this, one may search for structural dissimilarities between the two
excerpts. In fact, the score in itself reveals that, whereas crescendo was
quickly followed by decrescendo in bars 1-6, the crescendo continues
throughout the four bars 16-19 spanning all the way from piano to
forte. The waveform representation in Fig. 3 confirms this by reach-
ing higher intensity, entailing a longer crescendo with a much later
peak and accelerating tempo leading to intensification. This effect is
consistent with findings that loudness90 in music correlates positively
with ratings of emotional arousal91, 92 and chill occurrence.93, 94 In an
evolutionary perspective, it may be argued that, in the soundscape of
our ancestors, gradually increasing loudness was associated with an ap-
proaching sound source. This may obviously have entailed a possible
danger, and more attention would thus automatically be assigned to that
particular object.95 In Huron’s terms, an extended crescendo may re-
sult in increased tension response preparing us for facing the challenge.
This phenomenon is also captured by Juslin and Västfjäll’s notion of
a brain stem reflex mechanism responding to “fundamental acoustical
characteristics of the music.”
Fig. 3. Waveforms showing amplitude plotted against time for a
recording of bars 1-6 and 16-21 from Wagner’s Prelude to Act III
from Tristan und Isolde
Summing up, it first of all leaps to the eye that, despite the apparent
activation of many of the same emotional mechanisms as in the first
analytical example, this piece of music by Wagner succeeds in inducing
an emotional experience which is qualitatively considerably different
from that induced by Penderecki’s piece. After the analysis of the last
piece, we will look further into why this might be the case.
Shocking Chopin: Scherzo No. 2
The final example of strongly emotion-inducing music is the open-
ing from Frédéric Chopin’s Scherzo No. 2 in B-flat minor (op. 31)
which typically brings the present author in a mood of anxiety and inse-
curity. This emotional experience is substantially different from the one
evoked by the Wagner excerpt. Further analysis will explore reasons
why this may be the case.
As previously mentioned, changes in loudness predispose for chill
experience.96, 97 More specifically, gradual changes—like the crescendi
in the Wagner excerpt—lead to a response lag of approximately three
seconds whereas responses to sudden changes occur much faster already
less than one second post-stimuli.98 One possible explanation for this
difference would be that the fast reaction represents a “quick and dirty”
response to a potential danger where higher-level, cognitive processing
is circumvented; that is, a reaction response in Huron’s terms and a
brain stem reflex in Juslin and Västfjäll’s terms.99 Distinctively differ-
ent cognitive processes are likely to give rise to qualitatively different
emotions. This may explain why the emotional response tends more
towards anxiety than towards sadness in the case of this Chopin piece.
Moreover, the musical texture of the scherzo is governed by many paus-
es followed by very loud dynamics. Such features are likely to increase
tension response and similarly affect brain stem reflex mechanisms.
In bar 22, a syncopated base note appears on G-flat emphasised with
sfzorzando and very short articulation (see Fig. 4). Using a method of
annotation applied by Lerdahl and Jackendoff where the number of dots
represents the degree of expectedness of a given rhythmical event in
the hypermetrical hierarchy,100 it can be ascertained that the onset of an
event at the beginning of bar 21 is highly expected. Nevertheless, this
event is postponed, not only one bar, but one bar and a beat. This makes
the onset of G-flat extraordinarily unexpected.
Furthermore, the dominant harmony (F-major) does not resolve
to the tonic (B-flat minor), but rather deceptively to the sixth degree.
Huron provides tables of first-order transitional probabilities for scale
degrees as well as harmonies.101 Based on the presumption that sty-
Fig. 4. The opening from Chopin’s Scherzo No. 2 with Lerdahl and
Jackendoff’s (1983) annotation showing metrical structure
listic knowledge is acquired through statistical learning, Huron argues
that these probabilities correspond more or less to the expectations of
typical listeners. Importantly, the statistics referred to by Huron were
computed from samples of Baroque music in the major mode and may
thus not generalise to music by Chopin composed in the minor mode.
However, one may reliably assume that deceptive cadences are indeed
less likely than authentic ones. Thus, the great unpredictability and sud-
den dynamic change in bar 22 might predispose for a negatively va-
lenced emotional experience by means of Huron’s prediction response.
In support hereof, Sloboda found that a “new or unprepared harmony”
typically evokes shivers.102
In bar 46 a seemingly similar passage occurs (see Fig. 5). Interest-
ingly, however, the present author never experiences chills when listen-
ing to this part of the scherzo. Since the metrical displacement, dynam-
ics, and articulation are completely identical, it may be hypothesised
that the difference could be explained in terms of higher degrees of
melodic/harmonic predictability; indeed, the preceding octave E1/E2 is
an appoggiatura incongruent with the underlying harmony (F-minor).
Fig. 5. Bars 37-48 from Chopin’s Scherzo no. 2 with Lerdahl and
Jackendoff’s (1983) annotation showing metrical structure
Therefore, contrary to the deceptive cadence in bar 22, F1/F2 in bar
46 does not represent a chord change and is indeed highly expected.
Confirmation of expectancy is thus rewarded with a high prediction re-
sponse, and this may predispose for a positively rather than negatively
valenced emotional experience.
Post-emotional aftermath
The preceding analyses of excerpts from Penderecki, Wagner and
Chopin have demonstrated how two empirically based theories of mu-
sical emotions—namely the ITPRA Theory and the Multiple Mecha-
nisms Theory—can be applied to account for strongly emotion-induc-
ing listening experiences. Table 2 sums up how each of the underlying
mechanisms proposed by Huron and Juslin and Västfjäll were argued
to contribute to the emotions experienced by this author when listening
to these particular pieces. This concluding section of the paper evalu-
ates the two theories in the objective light that arises when the pain has
resolved, the tears have dried out, and the shock has worn off.
As evident from the table, the six mechanisms of the Multiple Mech-
anisms Theory all appeared to play some role. Nevertheless, not all of
them were of equal importance; in particular, the two mechanisms most
strongly related to musical structure—brain stem reflexes and musical
expectancy—were clearly also most extensively involved. This may
be indicative of a greater importance of structural features—at least
for listeners with relatively high levels of musical sophistication—than
originally envisioned by Juslin and Västfjäll. Such a possible underly-
ing hierarchical organisation of the induction mechanisms was already
anticipated in the previously mentioned review by Vuust and Frith.103
Huron’s approach to illuminating the musical expectancy mechanism
also seemed of high relevance to the present purpose. Especially the ten-
sion, prediction and reaction responses proved outmost illustrative in ac-
counting for musical narrative as communicated by common phenomena
such as surprise, appoggiaturas, pauses and deceptive cadences.
Table 2. Overview of the relevance of the different theoretical
subcomponents to the three musical pieces analysed here
Juslin and
Väs jäll
(2006) Penderecki Wagner Chopin
Brain stem
re ex
Beyond the
scope of
the theory
Sudeden dynamics
Sensory dissonance
Alterna ve tech-
“White noise“
Crescendi Sfz e ects
Evalua ve
condi oning
String instruments
associated with
Structural fea-
tures associated
with sad music
Emo onal
Opera c perfor-
Prosodic cues
Mul sensory
scenic staging
Title Autobiography
Tension Low predictability
(stretched out)
ability (in-
Predic on Surprise (nega ve) Appoggiaturas
(posi ve)
resolu ons
(nega ve)
resolu ons
(posi ve)
Reac on Sudden dynamics Sfz e ects
Appraisal (Autobiography)
A clear overlap between Juslin and Västfjäll’s brain stem reflex
mechanism and Huron’s tension and reaction responses leaped to the
eye. This raises an important question for future research to address,
namely to what extent the mechanisms of the two theories could in fact
be teased apart or if they should rather be characterised as more or less
identical. In the latter case, there seems to be reasons for preferring Hu-
ron’s theory because it gives a more detailed and explanatory account
that also draws the influence of sequential temporality into question,
which is evidently essential in music understood as a temporal art form.
For instance, low predictability (associated with tension response) and
sudden sounds (associated with reaction response) would both fall un-
der the same umbrella term (brain stem reflexes) in Juslin and Väst-
fjäll’s framework even though they differ both in terms of subjective
experience (uncertainty vs. surprise) and temporally sequential relation
to musical events (pre-outcome vs. post-outcome).
Moreover, phenomena like sudden dynamics and sensory disso-
nance can easily be regarded in a predictive framework as instances of
schematically unexpected events resulting in prediction error. Placing
these under brain stem reflexes, which is distinct from musical expec-
tancy, draws a somewhat problematic distinction between musical and
non-musical expectancy violations, which seems difficult to uphold.
What one listener experiences as noise, may indeed be perceived as a
highly meaningful musical event by another listener. This is an example
of something that was also pointed out by the analytical results as a
whole: namely that emotional experience relies heavily on schematic
expectations. Stylistic familiarity thus seems crucial to the listener’s
ability to decipher the “emotional code” of music and may indeed ex-
plain possible differences between the emotional experiences of listen-
ers with different levels of expertise.
Despite the somewhat superior explanatory power of the ITPRA
Theory, this theory may, nevertheless, be criticised on grounds of its
apparent tendency towards self-contradiction and lack of falsifiability
because it sometimes hypothesises strong emotional responses to ex-
pected as well as to unexpected events. For instance, it accounted for
emotional responses both to highly predictable resolutions of melodic
appoggiaturas in the Wagner analysis and to metrically and harmoni-
cally unexpected events in the Chopin analysis. In defence of Huron’s
theory, this is, however, one of the very important characteristics of
the prediction response that distinguishes it from both the tension and
reaction responses. More specifically, whereas reaction and tension re-
sponses almost per definition entail negatively valenced emotions, the
prediction response may result in more complex emotions, arguably
spanning the two extremes of the valence spectrum. The relationship
of expectancy confirmation and violation to emotional experience cer-
tainly constitutes a highly relevant topic for future research. Interesting-
ly, the tension response also seemed to embrace qualitatively different
emotions depending on whether tension was stretched out in time as in
the extended crescendi in the Wagner excerpt or resulted from more in-
stantaneous unpredictability as in the Penderecki and Chopin excerpts.
The analysis furthermore established that expressive performance
cues undoubtedly affect the workings of the musical expectancy mecha-
nism. Although this analysis should not be misinterpreted as supportive
of performance prescriptions, a basic understanding of the principles of
music perception and cognition may indeed inform artistic interpreters
and expand their expressive potential considerably.
The imagination and appraisal subcomponents of the ITPRA Theo-
ry were not explored in the same depth as the tension, prediction and re-
action responses in the current analysis. Indeed, given the fact that their
influence is theoretically unbounded backwards (imagination) and for-
wards (appraisal) in time, these mechanisms necessitate a longitudinal
approach which was not taken here and which may after all be difficult
to implement in music analysis. Also, along with evaluative condition-
ing, emotional contagion, visual imagery, and episodic memory from
the Multiple Mechanisms Theory, imagination and appraisal responses
are likely to be subject to considerable individual and contextual differ-
ences. Many of such differences are related to extra-musical factors that
are quite challenging to address in further detail for the music analyst
without assistance from an experimental psychologist who conducts
behavioural experiments.
Extra-musical factors also expose one of the other ambiguities—or
possibly shortcomings—of Juslin and Västfjäll’s theory. Specifically,
the temporal underpinnings of episodic memory seem fairly vague. In
fact, episodic memories can just as easily be involved in imagination
responses prior to the music as in appraisal responses long after the
actual music. Ignoring the time aspect of this mechanism altogether
seems inappropriate.
Finally, the methodology demonstrated in the analysis above took
the author’s subjective emotional experience as a starting point and
then searched for possible explanations. Not surprisingly, this is prob-
lematic since no criteria were provided for deciding when, for instance,
a sudden change in dynamics was forceful enough to evoke a given
emotional response. A more valid approach for future follow-up studies
would therefore be to turn the tables and predict emotional responses
directly from music theory and empirical research in music cognition
and then test whether those predictions hold true or not.
Hence, although some useful attempts have indeed been made, no
one has yet managed to establish a complete theory explaining how mu-
sic elicits emotions in listeners. Huron’s ITPRA Theory and Juslin and
Västfjäll’s Multiple Mechanisms Theory are both important contribu-
tions in this respect, but subsequent reviews as well as the analysis here
still point out necessary further developments. As he openly admits,
Huron only deals with one of the underlying mechanisms; Juslin and
Västfjäll, on the other hand, do not sufficiently account for the inter-
actions between their mechanisms and relate them clearly to the tem-
poral aspects of music experience. Importantly, neither of the theories
achieves a level of methodological specificity where the theory lends
itself directly and unambiguously to analysis of specific musical scores
and recordings.104 Therefore, when accounting for emotional experi-
ence of music we are still left to base our analysis on a wide number
of sources of which some are mutually exclusive, and all are somehow
incomplete in describing this phenomenon in its entirety. Perhaps the
development of a complete theory of musical emotions with applicabil-
ity to musical analysis represents a true challenge where music theorists
could be instrumental in bridging the gap between cognitive music re-
search and music analysis.
1. Leonard B. Meyer, Emotion and Meaning in Music (Chicago: Chicago Uni-
versity Press, 1956).
2. Leonard B. Meyer, Eugene Narmour, and Ruth A. Solie, Explorations in
Music, the Arts, and Ideas: Essays in Honor of Leonard B. Meyer (Hillsdale:
Pendragon Press, 1988).
3. Lola L. Cuddy, “Tributes to Leonard B. Meyer: Editor’s Introduction,” Music
Perception 25/5 (2008): 477.
4. Leslie Bunt and Mercedes Pavlicevic, “Music and Emotion: Perspectives from
Music Therapy,” in Music and Emotion: Theory and Research, eds. Patrik N.
Juslin & John A. Sloboda, 181-201 (New York: Oxford University Press, 2001).
5. Mark F. Zander, “Musical In uences in Advertising: How Music Modi es
First Impressions of Product Endorsers and Brands,” Psychology of Music 34/4
(2006): 465-80.
6. Annabel J. Cohen, “Music as a Source of Emotion in Film,” in Handbook of
Music and Emotion: Theory, Research, Applications (Series in Affective Sci-
ence), eds. Patrik N. Juslin and John A. Sloboda (New York: Oxford University
Press, 2010).
7. Jessica C. E. Gienow-Hecht, Sound Diplomacy: Music and Emotions in Trans-
atlantic Relations, 1850-1920 (Chicago: University of Chicago Press, 2009).
8. Jenefer Robinson, Deeper Than Reason: Emotion and Its Role in Literature,
Music, and Art, (Oxford: Clarendon Press, 2005).
9. Eds. Mathieu Barthet and Simon Dixon, Music and Emotions: Proceedings
of the 9th International Symposium on Computer Music Modelling and Re-
trieval (London: Queen Mary University of London, 2012), http://cmmr2012. les/pdf/CMMR2012Proc-
ceedingsFinal.pdf (accessed April 17, 2013).
10. Stefan Koelsch, “Towards a Neural Basis of Music-Evoked Emotions,”
Trends in Cognitive Sciences 14/3 (2010): 131-7.
11. Niels Chr. Hansen, “The Legacy of Lerdahl and Jackendoff’s a Generative
Theory of Tonal Music: Bridging a Signi cant Event in the History of Music
Theory and Recent Developments in Cognitive Music Research,” Danish Year-
book of Musicology 38 (2011): 33-55.
12. Eugene Narmour, “Our Varying Histories and Future Potential: Models and
Maps in Science, the Humanities, and in Music Theory,” Music Perception 29/1
(2011): 1-21.
13. David Huron, Sweet Anticipation: Music and the Psychology of Expectation
(Cambridge, MA: MIT Press, 2006).
14. Patrik N. Juslin and Daniel Västfjäll, “Emotional Responses to Music: The
Need to Consider Underlying Mechanisms,” Behavioral and Brain Sciences 31
(2008): 559-75.
15. Ibid.
16. Ibid., 561.
17. Ibid., 564-8.
18. Carol L. Krumhansl and Kat R. Agres, “Musical Expectancy: The In uence
of Musical Structure on Emotional Response,” Behavioral and Brain Sciences
31/5 (2008).
19. Peter Vuust and Chris D. Frith, “Anticipation Is the Key to Understanding
Music and the Effects of Music on Emotion,” Behavioral and Brain Sciences
31/05 (2008): 599-600.
20. Guy Madison, “What About the Music? Music-Speci c Functions Must
Be Considered in Order to Explain Reactions to Music,” Behavioral and Brain
Sciences 31/5 (2008): 587.
21. Vladimir J. Konečni, “A Skeptical Position on Musical Emotions and an
Alternative Proposal,” Behavioral and Brain Sciences 31/5 (2008): 582-4.
22. Huron, Sweet Anticipation.
23. Jenny R. Saffran et al., “Statistical Learning of Tone Sequences by Human
Infants and Adults,” Cognition 70, no. 1 (1999): 27-52, doi: 10.1016/S0010-
24. Niels Chr. Hansen and Marcus T. Pearce, “Predictive uncertainty re ects
statistical learning of sensory input,” in review.
25. Jan W. Van Strien and Claudia A. Boon, “Altered Visual Field Asymmetry
for Lexical Decision as a Result of Concurrent Presentation of Music Frag-
ments of Different Emotional Valences,” Journal of the International Neu-
ropsychological Society 3/5 (1997): 473-9.
26. Stefan Evers and Birgit Suhr, “Changes of the Neurotransmitter Serot-
onin but Not of Hormones During Short Time Music Perception,” European
Archives of Psychiatry and Clinical Neuroscience 250/3 (2000): 144-7, doi:
27. Sherman D. VanderArk and Daniel Ely, “Biochemical and Galvanic Skin
Responses to Music Stimuli by College Students in Biology and Music,” Per-
ceptual and Motor Skills 74/3 (1992): 1079-90.
28. Sherman D. VanderArk and Daniel Ely, “Cortisol, Biochemical, and Gal-
vanic Skin Responses to Music Stimuli of Different Preference Values by Col-
lege Students in Biology and Music,” Perceptual and Motor Skills 77/1 (1993):
29. Chad Louis Stephens, “Autonomic Differentiation of Emotions: A Cluster
Analysis Approach” (MSc dissertation, Virginia Polytechnic Institute and State
University, 2007).
30. Laura-Lee Balkwill and William Forde Thompson, “A Cross-Cultural In-
vestigation of the Perception of Emotion in Music: Psychophysical and Cultural
Cues,” Music Perception 17/1 (1999): 43-64.
31. Gene Ann Behrens and Samuel B. Green, “The Ability to Identify Emotion-
al Content of Solo Improvisations Performed Vocally and on Three Different
Instruments,” Psychology of Music 21/1 (1993): 20-33.
32. Stuart Fischoff, “The Evolution of Music in Film and Its Psychological
Impact on Audiences,” lmmusic.
pdf (accessed May 17, 2013).
33. David Huron, “Tone and Voice: A Derivation of the Rules of Voice-Leading
from Perceptual Principles,” Music Perception 19/1 (2001): 1--64.
34. Oliver Grewe, Frederik Nagel, Reinhard Kopiez, and Eckhardt Altenmüller,
“How Does Music Arouse ‘Chills’?” Annals of the New York Academy of Sci-
ences 1060/1 (2005): 446-9, doi: 10.1196/annals.1360.041.
35. Frederik Nagel, Reinhard Kopiez, Oliver Grewe, and Eckhardt Altenmüller,
“Psychoacoustical Correlates of Musically Induced Chills,” Musicae Scientiae
12/1 (2008): 101-13.
36. David C. Glass and Jerome E. Singer, Urban Stress: Experiments on Noise
and Social Stressors (New York: Academic Press, 1972).
37. Thomas Adrian, “Penderecki, Krzysztof,” in The Oxford Companion to Mu-
sic: Oxford Music Online, ed. Alison Latham (Oxford: Grove Music Online,
38. Danuta Mirka, “Texture in Penderecki’s Sonoristic Style,” Music Theory
Online 6/1 (2000).
39. Danuta Mirka, “To Cut the Gordian Knot: The Timbre System of Krzysztof
Penderecki,” Journal of Music Theory 45/2 (2001): 435-56.
40. Mirka, “To Cut the Gordian Knot.”
41. William Forde Thompson, Music, Thought, and Feeling: Understanding
the Psychology of Music (New York: Oxford University Press, 2009), 48-50.
42. In fact, though stable within the pitch range of most music, the critical
band width measured in semitones is not completely constant across the whole
pitch range of human hearing; see David Huron and Peter Sellmer, “Critical
bands and the spelling of vertical sonorities,” Music Perception 10/2 (1992),
43. Eugene S. Morton, “Sound Symbolism and Its Role in Non-Human Ver-
tebrate Communication,” in Sounds Symbolism, eds. Leanne Hinton, Johanna
Nichols, and John J. Ohala (Cambridge: Cambridge University Press, 1994),
44. Anonymous, “Sul ponticello,” in The New Grove Dictionary of Music and
Musicians, 2nd ed., eds. Stanley Sadie & John Tyrrell (Oxford: Oxford Music
Online, 2001).
45. Anne J. Blood, Robert Zatorre, Patrick Bermudez, and Alan C. Evans,
“Emotional Responses to Pleasant and Unpleasant Music Correlate with Activ-
ity in Paralimbic Brain Regions,” Nature Neuroscience 2/4 (1999): 382-7, doi:
46. Richard D. Lane, Eric M. Reiman, Margaret M. Bradley, Peter J.
Lang,Geoffrey L. Ahern, Richard J. Davidson, and Gary E. Schwartz, “Neuro-
anatomical Correlates of Pleasant and Unpleasant Emotion,” Neuropsychologia
35/11 (1997).
47. Amy F. T. Arnsten and Patricia S. Goldman-Rakic, “Noise Stress Impairs
Prefrontal Cortical Cognitive Function in Monkeys: Evidence for a Hyper-
dopaminergic Mechanism,” Archives of General Psychiatry 55/4 (1998):
48. Marianne Frankenhaeuser and Ulf Lundberg, “The In uence of Cognitive
Set on Performance and Arousal under Different Noise Loads,” Motivation and
Emotion 1/2 (1977): 139-49, doi: 10.1007/BF00998516.
49. Frances K. Graham and Diana Arezzo Slaby, “Differential Heart Rate
Changes to Equally Intense White Noise and Tone,” Psychophysiology 10/4
(1973): 347-62, doi: 10.1111/j.1469-8986.1973.tb00792.x.
50. Jaak Panksepp, “The Emotional Sources Of ‘Chills’ Induced by Music,”
Music Perception 13/2 (1995): 171-207.
51. Carol L. Krumhansl, “An Exploratory Study of Musical Emotions and Psy-
chophysiology,” Canadian Journal of Experimental Psychology 51/4 (1997):
52. Gabriela Husain, William Forde Thompson, and Glenn Schellenberg, “Ef-
fects of Musical Tempo and Mode on Arousal, Mood, and Spatial Abilities,”
Music Perception 20/2 (2002): 151-71.
53. Lage Wedin, “A Multidimensional Study of Perceptual-Emotional Quali-
ties in Music,” Scandinavian Journal of Psychology 13/1 (1972): 241-57, doi:
54. Kate Hevner, “The Affective Value of Pitch and Tempo in Music,” The
American Journal of Psychology 49 (1937): 621-30.
55. Olaf Post and David Huron, “Western Classical Music in the Minor Mode
Is Slower (except in the Romantic Period),” Empirical Musicology Review 4/1
56. Husain, Thompson, and Schellenberg, “Effects of Musical Tempo.”
57. Balkwill and Thompson, “A Cross-Cultural Investigation.”
58. Alf Gabrielsson and Patrik N. Juslin, “Emotional Expression in Music Per-
formance: Between the Performer’s Intention and the Listener’s Experience,”
Psychology of Music 24/1 (1996): 68-91.
59. Klaus R. Scherer and James S. Oshinsky, “Cue Utilization in Emotion At-
tribution from Auditory Stimuli,” Motivation and Emotion 1/4 (1977): 331-46.
60. Melvin G. Rigg, “Musical Expression: An Investigation of the Theories of
Erich Sorantin,” Journal of Experimental Psychology 21/4 (1937): 442-55.
61. Hevner, “The Affective Value.”
62. David Huron, “A Comparison of Average Pitch Height and Interval Size in
Major- and Minor-Key Themes: Evidence Consistent with Affect-Related Pitch
Prosody,” Empirical Musicology Review 3/2 (2008): 59.
63. David Huron, Gary Yim, and Parag Chordia, “The Effect of Pitch Expo-
sure on Sadness Judgments: An Association between Sadness and Lower Than
Normal Pitch” (paper presented at The 11th International Conference on Music
Perception and Cognition, Seattle, USA, 2010).
64. Scherer and Oshinsky, “Cue Utilization.”
65. Wedin, “A Multidimensional Study.”
66. Hevner, “The Affecive Value.”
67. Balkwill and Thompson, “A Cross-Cultural Investigation.”
68. Huron, “A Comparison.”
69. Patrik N. Juslin and Petri Laukka, “Expression, Perception, and Induc-
tion of Musical Emotions: A Review and a Questionnaire Study of Every-
day Listening,” Journal of New Music Research 33/3 (2004): 217-38, doi:
70. Michael Schutz, David Huron, Kristopher Keaton, and Greg Loewer, “The
Happy Xylophone: Acoustics Affordances Restrict an Emotional Palate,” Em-
pirical Musicology Review 3/3 (2008): 126-35.
71. Wedin, “A Multidimensional Study.”
72. Hevner, “The Affective Value.”
73. Rigg, “Musical Expression.”
74. Ibid.
75. Wedin, “A Multidimensional Study.”
76. Juslin and Laukka, “Expression, Perception.”
77. Gabrielsson and Juslin, “Emotional Expression.”
78. Ben Turner and David Huron, “A Comparison of Dynamics in Major-and
Minor-Key Works,” Empirical Musicology Review 3 (2008): 64-8.
79.Gabrielsson and Juslin, “Emotional Expression.”
80. Rigg, “Musical Expression.”
81. Juslin and Laukka, “Expression, Perception.”
82. Gabrielsson and Juslin, “Emotional Expression.”
83. For a review of early research in the connection between musical fea-
tures and emotions, see Rigg, “Musical Expression,” whereas recent results
are summed up by Patrik N. Juslin, “From Mimesis to Catharsis: Expression,
Perception, and Induction of Emotion in Music,” in Musical Communication,
eds. Dorothy Miell, Raymond A. R. MacDonald, and David John Hargreaves,
85-115 (Oxford: Oxford University Press, 2005).
84. Juslin, “From Mimesis.”
85. David Huron, “Why Is Sad Music Pleasurable? A Possible Role for Prolac-
tin,” Musicae Scientiae 15/2 (2011): 146-58.
86. Caterina Breitenstein, Diana Van Lancker, and Irene Daum, “The Contribu-
tion of Speech Rate and Pitch Variation to the Perception of Vocal Emotions in a
German and an American Sample,” Cognition & Emotion 15/1 (2001): 57-79.
87. Aron W. Siegman and Stephen Boyle, “Voices of Fear and Anxiety and
Sadness and Depression: The Effects of Speech Rate and Loudness on Fear and
Anxiety and Sadness and Depression,” Journal of Abnormal Psychology 102/3
(1993): 430.
88. Grant Fairbanks and Wilbert Pronovost, “An Experimental Study of the
Pitch Characteristics of the Voice During the Expression of Emotion,” Com-
munications Monographs 6/1 (1939): 87-104.
89. John A. Sloboda, “Music Structure and Emotional Response: Some Empiri-
cal Findings,” Psychology of Music 19/2 (1991): 110.
90. Loudness measured in terms of A-weighted dB.
91. Alf Gabrielsson and Erik Lindström, “The In uence of Musical Structure
on Emotional Expression,” in Music and Emotion: Theory and Research, eds.
Patrik N. Juslin & John A. Sloboda, 223-248 (New York: Oxford University
Press, 2001).
92. Emery Schubert, “Modeling Perceived Emotion with Continuous Musical
Features,” Music Perception 21/4 (2004): 561-85.
93. Grewe et al., “How Does Music.”
94. Nagel et al., “Psychoacoustical Correlates.”
95. Ibid.
96. Grewe et al., “How Does Music.”
97. Nagel et al., “Psychoacoustical Correlates.”
98. Schubert, “Modeling Perceived Emotion.”
99. This would similarly be the case for the shock effects applied extensively
in Penderecki’s Threnody.
100. Fred Lerdahl and Ray Jackendoff, A Generative Theory of Tonal Music
(Cambridge: MIT Press, 1983).
101. Huron, Sweet Anticipation, 158-59, 251.
102. Sloboda, “Music Structure,” 114.
103. Vuust and Frith, “Anticipation is the Key.”
104. Similar criticism has previously been expressed in reviews of Huron’s
Sweet Anticiopation; see William Forde Thompson, “Review: David Hu-
ron, Sweet Anticipation: Music and the Psychology of Expectation,” Empirical
Musicology Review 2/2 (2007), 67-70; Adam Ockelford, “Review: D. Huron,
Sweet Anticipation: Music and the Psychology of Expectation,” Psychology of
Music, 36/), 367-382.
Adrian, T. 2002. Penderecki, Krzysztof. In A. Latham (Ed.), The Oxford Com-
panion to Music: Oxford Music Online.
Anonymous. 2001. Sul ponticello. In Stanley Sadie and John Tyrrell (Eds.),
The New Grove Dictionary of Music and Musicians [2nd edition]. Oxford
Music Online.
Arnsten, A. F. T., and Goldman-Rakic, P. S. 1998. “Noise stress impairs pre-
frontal cortical cognitive function in monkeys: Evidence for a hyperdop-
aminergic mechanism,” Archives of General Psychiatry, 55(4), 362-368.
Balkwill, L. L., and Thompson, W. F. 1999. “A cross-cultural investigation of
the perception of emotion in music: Psychophysical and cultural cues,” Mu-
sic Perception, 17(1), 43-64.
Barthet, M., and Dixon, S. eds. 2012. Music and Emotions: Proceedings of the
9th International Symposium on Computer Music Modelling and Retrieval.
London, UK: Queen Mary, University of London. Retrieved April 17, 2013
Behrens, G. A., and Green, S. B. 1993. “The ability to identify emotional content
of solo improvisations performed vocally and on three different instruments,”
Psychology of Music, 21(1), 20-33. doi: 10.1177/030573569302100102.
Blood, A. J., Zatorre, R. J., Bermudez, P., and Evans, A. C. 1999. “Emotional
responses to pleasant and unpleasant music correlate with activity in paral-
imbic brain regions,” Nature Neuroscience, 2(4), 382-387.
Breitenstein, C., Lancker, D. V., and Daum, I. 2001. “The contribution of speech
rate and pitch variation to the perception of vocal emotions in a German and
an American sample,” Cognition & Emotion, 15(1), 57-79.
Bunt, L., and Pavlicevic, M. 2001. “Music and emotion: Perspectives from mu-
sic therapy.” In P. N. Juslin and J. A. Sloboda eds., Music and emotion:
Theory and research. New York, NY: Oxford University Press, 181-201
Cohen, A. J. 2010. “Music as a source of emotion in lm.” In P. N. Juslin and
J. A. Sloboda eds., Handbook of music and emotion: Theory, research, ap-
plications, 879-908. New York, NY: Oxford University Press.
Cuddy, L. L. 2008. “Tributes to Leonard B. Meyer: Editor’s introduction,” Mu-
sic Perception, 25(5), 477-477.
Evers, S., and Suhr, B. 2000. “Changes of the neurotransmitter serotonin but
not of hormones during short time music perception,” European Archives
of Psychiatry and Clinical Neuroscience, 250(3), 144-147. doi: 10.1007/
Fairbanks, G., and Pronovost, W. 1939. “An experimental study of the pitch
characteristics of the voice during the expression of emotion,” Communica-
tions Monographs, 6(1), 87-104.
Fischoff, S. 2005. “The evolution of music in lm and its psychological im-
pact on audiences.” Retrieved from
tvf454/5 lmmusic.pdf
Frankenhaeuser, M., and Lundberg, U. 1977. “The in uence of cognitive set
on performance and arousal under different noise loads,” Motivation and
Emotion, 1(2), 139-149. doi: 10.1007/bf00998516.
Gabrielsson, A., and Juslin, P. N. 1996. “Emotional expression in music perfor-
mance: Between the performer’s intention and the listener’s experience,”
Psychology of Music, 24(1), 68-91.
Gabrielsson, A., and Lindström, E. 2001. “The in uence of musical structure
on emotional expression.” In P. N. Juslin and J. A. Sloboda, eds., Music and
Emotion: Theory and Research, 223-248. New York, NY: Oxford Univer-
sity Press.
Gienow-Hecht, J. C. E. 2009. Sound diplomacy: Music and emotions in trans-
atlantic relations, 1850-1920. Chicago, IL: University of Chicago Press.
Glass, D. C., Singer, J. E. 1972. Urban stress: Experiments on noise and social
stressors. New York, NY: Academic Press.
Graham, F. K. and Slaby, D. A. 1973. “Differential heart rate changes to equal-
ly intense white noise and tone,” Psychophysiology, 10(4), 347-362. doi:
Grewe, O., Nagel, F., Kopiez, R., and Altenmüler, E. 2005. “How does music
arouse ‘chills’”? Annals of the New York Academy of Sciences, 1060(1),
446-449. doi: 10.1196/annals.1360.041.
Hansen, N. C. 2011. “The legacy of Lerdahl and Jackendoff’s A Generative
Theory of Tonal Music: Bridging a signi cant event in the history of music
theory and recent developments in cognitive music research,” Danish Year-
book of Musicology, 38, 33-55.
Hansen, N. C., and Pearce, M. (submitted). Predictive uncertainty re ects sta-
tistical learning of sensory input.
Hevner, K. 1937. “The affective value of pitch and tempo in music,” The Amer-
ican Journal of Psychology, 49, 621-630. doi: 10.2307/1416385.
Huron, D. 2001. “Tone and voice: A derivation of the rules of voice-leading
from perceptual principles,” Music Perception, 19(1), 1-1-64.
_______. 2006. Sweet anticipation: Music and the psychology of expectation.
Cambridge, MA, US: The MIT Press.
_______. 2008. “A comparison of average pitch height and interval size in ma-
jor- and minor-key themes: Evidence consistent with affect-related pitch
prosody,” Empirical Musicology Review, 3(2), 59.
_______. 2011. “Why is sad music pleasurable? A possible role for prolactin.”
Musicae Scientiae, 15(2), 146-158.
Huron, D., and Sellmer, P. 1992. “Critical bands and the spelling of vertical
sonorities,” Music Perception, 10(2), 129-150.
Huron, D., Yim, G., and Chordia, P. 2010. “The effect of pitch exposure on
sadness judgments: An association between sadness and lower than normal
pitch.” Paper presented at the The 11th International Conference on Music
Perception and Cognition, Seattle, WA.
Husain, G., Thompson, W. F., and Schellenberg, E. G. 2002. “Effects of musical
tempo and mode on arousal, mood, and spatial abilities,” Music Perception,
20(2), 151-171. doi: 10.1525/mp.2002.20.2.151.
Juslin, P. and Laukka, P. 2004. “Expression, perception, and induction
of musical emotions: A review and a questionnaire study of every-
day listening,” Journal of New Music Research, 33(3), 217-238. doi:
Juslin, P. N. 2005. “From mimesis to catharsis: Expression, perception, and
induction of emotion in music.” In D. Miell, R. A. R. MacDonald and D. J.
Hargreaves (Eds.), Musical Communication, 85-115. Oxford, UK: Oxford
University Press.
Juslin, P. N., and Vastfjall, D. 2008. “Emotional responses to music: The need
to consider underlying mechanisms,” Behavioral and Brain Sciences, 31,
Koelsch, S. 2010. “Towards a neural basis of music-evoked emotions,” Trends
in Cognitive Sciences, 14(3), 131-137.
Konečni, V. J. 2008. “A skeptical position on ‘musical emotions’ and an alterna-
tive proposal,” Behavioral and Brain Sciences, 31(05), 582-584.
Krumhansl, C. L. 1997. “An exploratory study of musical emotions and psy-
chophysiology,” Canadian Journal of Experimental Psychology/Revue ca-
nadienne de psychologie expérimentale, 51(4), 336-353. doi: 10.1037/1196-
Krumhansl, C. L., and Agres, K. R. 2008. “Musical expectancy: The in uence
of musical structure on emotional response,” Behavioral and Brain Sci-
ences, 31(05), 584-585. doi: 10.1017/S0140525X08005384.
Lane, R. D., Reiman, E. M., Bradley, M. M., Lang, P. J., Ahern, G. L., Da-
vidson, R. J., and Schwartz, G. E. (1997). “Neuroanatomical correlates of
pleasant and unpleasant emotion,” Neuropsychologia, 35(11), 1437-1444.
doi: 10.1016/s0028-3932(97)00070-5.
Lerdahl, F., and Jackendoff, R. 1983. A Generative Theory of Tonal Music.
Cambridge, MA: MIT Press.
Madison, G. 2008. “What about the music? Music-speci c functions must be
considered in order to explain reactions to music,” Behavioral and Brain
Sciences, 31(05), 587-587.
Meyer, L. B. 1956. Emotion and Meaning in Music. Chicago, IL: Chicago Uni-
versity Press.
Meyer, L. B., Narmour, E., and Solie, R. A. 1988. Explorations in music, the
arts, and ideas: essays in honor of Leonard B. Meyer. Hillsdale, NY: Pen-
dragon Press.
Mirka, D. 2000. “Texture in Penderecki’s sonoristic style,” Music Theory On-
line, 6(1).
_______. 2001. “To cut the Gordian knot: The timbre system of Krzysztof Pen-
derecki,” Journal of Music Theory, 45(2), 435-456.
Morton, E. S. 1994. “Sound symbolism and its role in non-human vertebrate
communication.” In L. Hinton, J. Nichols and J. Ohala, eds., Sounds Sym-
bolism, 348-365. Cambridge, UK: Cambridge University Press.
Nagel, F., Kopiez, R., Grewe, O., and Altenmüller, E. 2008. “Psychoacoustical
correlates of musically induced chills.” Musicae Scientiae, 12(1), 101-113.
Narmour, E. 2011. “Our varying histories and future potential: Models and
maps in science, the humanities, and in music theory,” Music Perception,
29(1), 1-21.
Ockelford, A. 2008. Review: “D. Huron, Sweet Anticipation: Music and the
Psychology of Expectation,” Psychology of Music, 36(3), 367-382.
Panksepp, J. 1995. “The emotional sources of ‘chills’ induced by music,” Music
Perception, 13(2), 171-207.
Rigg, M. 1937. “Musical expression: an investigation of the theories of Er-
ich Sorantin,” Journal of Experimental Psychology, 21(4), 442-455. doi:
Rigg, M. G. 1964. “The mood effects of music: A comparison of data from four
investigators,” Journal of Psychology, 58, 427-438.
Robinson, J. 2005. Deeper than reason: Emotion and its role in literature, mu-
sic, and art. Oxford, UK: Clarendon Press.
Saffran, J. R., Johnson, E. K., Aslin, R. N., and Newport, E. L. 1999. “Statistical
learning of tone sequences by human infants and adults,” Cognition, 70(1),
27-52, doi: 10.1016/S0010-0277(98)00075-4.”
Scherer, K. R., and Oshinsky, J. S. 1977. “Cue utilization in emotion attribu-
tion from auditory stimuli,” Motivation and Emotion, 1(4), 331-346. doi:
Schubert, E. 2004. “Modeling perceived emotion with continuous musical fea-
tures,” Music Perception, 21(4), 561-585. doi: 10.1525/mp.2004.21.4.561.
Schutz, M., Huron, D., Keeton, K. and Loewer, G. 2008. “The happy xylo-
phone: Acoustics affordances restrict an emotional palate,” Empirical Mu-
sicology Review, 3(3), 126-135.
Siegman, A. W., and Boyle, S. 1993. “Voices of fear and anxiety and sadness
and depression: The effects of speech rate and loudness on fear and anxi-
ety and sadness and depression,” Journal of Abnormal Psychology, 102(3),
Sloboda, J. A. 1991. “Music structure and emotional response: Some empirical
ndings,” Psychology of Music, 19(2), 110-120.
Stephens, C. L. 2007. Autonomic Differentiation of Emotions: A Cluster Analy-
sis Approach. (MSc dissertation). Virginia, US: Virginia Polytechnic Insti-
tute and State University.
Thompson, W. F. 2007. Review: “David Huron, Sweet Anticipation: Music and
the Psychology of Expectation, Empirical Musicology Review, 2(2), 67-
Thompson, W. F. 2009. Music, thought, and feeling: understanding the psy-
chology of music. New York, NY: Oxford University Press.
Tirovolas, A. K., and Levitin, D. J. 2011. “Music perception and cognition re-
search from 1983 to 2010: A categorical and bibliometric analysis of em-
pirical articles in Music Perception,” Music Perception, 29(1), 23-36.
Turner, B., and Huron, D. 2008. “A comparison of dynamics in major-and mi-
nor-key works,” Empirical Musicology Review, 3(2), 64-68.
Van Strien, J. W., and Boon, C. A. 1997. “Altered visual eld asymmetry for
lexical decision as a result of concurrent presentation of music fragments of
different emotional valences,” Journal of the International Neuropsycho-
logical Society, 3(05), 473-479.
VanderArk, S. D., and Ely, D. (1992). Biochemical and galvanic skin responses
to music stimuli by college students in biology and music, Perceptual and
Motor Skills, 74(3, Pt 2), 1079-1090. doi: 10.2466/pms.74.4.1079-1090
VanderArk, S. D., and Ely, D. 1993. “Cortisol, biochemical, and galvanic skin
responses to music stimuli of different preference values by college stu-
dents in biology and music,” Perceptual and Motor Skills, 77(1), 227-234.
Vuust, P., and Frith, C. D. 2008. “Anticipation is the key to understanding mu-
sic and the effects of music on emotion,” Behavioral and Brain Sciences,
31(05), 599-600. doi: 10.1017/S0140525X08005542.
Wedin, L. 1972. “A multidimensional study of perceptual-emotional qualities
in music,” Scandinavian Journal of Psychology, 13, 241.
Zander, M. F. 2006. “Musical in uences in advertising: How music modi es
rst impressions of product endorsers and brands,” Psychology of Music,
34(4), 465-480.
... While the authors mention Gabrielsson and Lindström's (2001) seminal article, they fail to derive a priori hypotheses from this review. As outlined by Hansen (2013), the rich research literature on emotions in music spans back to at least Hevner (1937) and Rigg (1937) with a multitude of other papers published during the intervening years before Gabrielsson and Lindström's overview. Although their 17year-old article provides an important contribution to the field, it is also not fair to claim that it still represents "the current state of research on the perceived emotional content of individual musical elements" (p. ...
Full-text available
This commentary discusses Sun and Cuthbert's (2018) exploratory analysis of emotional word painting in a corpus of English-language popular and folk songs. The authors are complimented for their application of computational tools to an impressively large sample of a somewhat understudied musical genre, and for their detailed level of analysis mapping musical features to the semantic content of individual words. This work, however, suffers from a lack of a priori predictions which causes multiple comparison issues leading to a dramatic reduction in statistical power. The selection of musical features and analytical strategies also seems arbitrary at times due to the absence of motivating hypotheses. It is argued that the ethological literature on affective vocal communication in animals might offer an avenue for future hypothesis-driven research on this topic.
Full-text available
An analysis of 9,788 instrumental themes shows that minor-key themes are, on average, slightly lower in pitch than major-key themes. The lower pitch is not merely an artifact of structural differences in the scales. In addition, instrumental themes in minor keys show a weak though significant tendency to use smaller pitch intervals. Both results are consistent with observations in speech prosody, where sad speakers exhibit a lower F0 and narrower pitch fluctuation compared with normal or happy speakers.
Full-text available
In many ways, the structure of music resembles that of language, including the acoustic cues used to communicate emotion. In speech, sadness is imparted through a combination of low fundamental frequency, dark timbre, and a slow rate of articulation. As the acoustic properties of the xylophone are not conducive to mimicking these cues, it seems to follow that composers would avoid attempts to write “sad” music for it. We investigated this idea by comparing the repertoire of the xylophone with that of the marimba – a similar instrument whose acoustic structure permits a greater variety of timbres, pitch heights, and tone durations. An analysis of repertoire drawn from the Percussive Arts Society database of recital programs reveals that 60% of the tonal marimba examples surveyed were written in minor (nominally “sad”) keys. In contrast, a parallel analysis of xylophone literature found minor keys used in only 6% of the examples surveyed. Further investigation revealed that the only examples of minor-key xylophone compositions included in this survey are in fact typically performed on the marimba. The avoidance of minor-key works on xylophone by both composers and performers is consistent with the idea that instruments restricted to producing tones with short durations, bright timbres, and high pitch heights are unable to mimic the speech cues used to convey sadness and/or depression.
Full-text available
An analysis of dynamic markings in 24 paired preludes in major and minor keys shows that the minor mode is associated with a generally lower dynamic level than the major mode. The results are consistent with observations in speech prosody, where sad or depressed speakers show reduced acoustic energy compared with normal or happy speakers.
In his so-called "sonoristic" period of the early 1960s--represented by pieces such as Threnody, Fluorescences, Polymorphia, and others--Penderecki employed a compositional system whose axiomatic concept was not a single sound, but the sound matter in its totality. Distinct states of this sound matter were governed by two relatively independent systems: (1) a basic system which ruled the texture of sound masses and (2) a timbre system governing their sound color. Categories of the basic system are a few binary oppositions concerning pitch, time, and loudness: spatial mobility vs. immobility, temporal mobility vs. immobility, spatial continuity vs. discontinuity, temporal continuity vs. discontinuity, high vs. low register, loud vs. soft dynamics. These categories account for the morphology of the basic system because a combination of terms chosen from individual categories generates an inventory of units in Penderecki's sonoristic style. The same set of categories also determines syntax, as the temporal order of units in the course of musical narration is ruled by the internal logic of individual binary oppositions. Categories of the timbre system are in turn metal, wood, and leather--materials of which the sound sources of traditional musical instruments are most often made--forming a ternary opposition. The timbre system underlies the wealth of new musical tools as well as eccentric playing techniques on traditional instruments called for by the composer.
This book takes the insights of modern psychological and neuroscientific research on the emotions and brings them to bear on questions about our emotional involvement with the arts. It begins by laying out a theory of emotion, one that is supported by the best evidence from current empirical work on emotions, and then in the light of this theory examines some of the ways in which the emotions function in the arts. Written in a clear and engaging style, the book will make fascinating reading for anyone who is interested in the emotions and how they work, as well as anyone engaged with the arts and aesthetics, especially with questions about emotional expression in the arts, emotional experience of art forms, and, more generally, artistic interpretation. Part One develops a theory of emotions as processes, having at their core non-cognitive 'instinctive' appraisals, 'deeper than reason', which automatically induce physiological changes and action tendencies, and which then give way to cognitive monitoring of the situation. Part Two examines the role of the emotions in understanding literature, especially the great realistic novels of the 19th century. It is argued that such works need to be experienced emotionally if they are to be properly understood. A detailed reading of Edith Wharton's novel The Reef demonstrates how a great novel can educate us emotionally by first evoking instinctive emotional responses, and then getting us to cognitively monitor and reflect upon them. Part Three puts forward a new Romantic theory of emotional expression in the arts. Part Four deals with music, both the emotional expression of emotion in music, whether vocal or instrumental, and the arousal of emotion by music. The way music arouses emotion lends indirect support to the theory of emotion outlined in Part One.
A basic issue about musical emotions concerns whether music elicits emotional responses in listeners (the 'emotivist' position) or simply expresses emotions that listeners recognize in the music (the 'cognitivist' position). To address this, psychophysiological measures were recorded while listeners heard two excerpts chosen to represent each of three emotions: sad, fear, and happy. The measures covered a fairly wide spectrum of cardiac, vascular, electrodermal, and respiratory functions. Other subjects indicated dynamic changes in emotions they experienced while listening to the music on one of four scales: sad, fear, happy, and tension. Both physiological and emotion judgments were made on a second-by-second basis. The physiological measures all showed a significant effect of music compared to the pre-music interval. A number of analyses, including correlations between physiology and emotion judgments, found significant differences among the excerpts. The sad excerpts produced the largest changes in heart rate, blood pressure, skin conductance and temperature. The fear excerpts produced the largest changes in blood transit time and amplitude. The happy excerpts produced the largest changes in the measures of respiration. These emotion-specific physiological changes only partially replicated those found for non-musical emotions. The physiological effects of music observed generally support the emotivist view of musical emotions.