Philosophy of art assignment

Towards a neural basis of music-evoked emotions Stefan Koelsch

Cluster of Excellence ‘‘Languages of Emotion’’, Freie Universität Berlin, Habelschwerdter Allee 45, 14195 Berlin, Germany

Review

Music is capable of evoking exceptionally strong emotions and of reliably affecting the mood of individ- uals. Functional neuroimaging and lesion studies show that music-evoked emotions can modulate activity in virtually all limbic and paralimbic brain structures. These structures are crucially involved in the initiation, generation, detection, maintenance, regulation and termination of emotions that have survival value for the individual and the species. Therefore, at least some music-evoked emotions involve the very core of evolutionarily adaptive neuroaffective mechanisms. Because dysfunctions in these structures are related to emotional disorders, a better understanding of music-evoked emotions and their neural correlates can lead to a more systematic and effective use of music in therapy.

The benefits of investigating emotion with music In most humans, music can strongly affect emotion and mood, and such effects are among the main reasons to produce, and listen to, music [1,2]. However, a common misconception is that music-evoked emotions only involve aesthetic experiences, lacking motivational components and goal relevance (Box 1; for reviews see Refs. [2,3]). This view implies that music is not capable of evoking ‘‘every- day emotions’’, and therefore is not well suited to inves- tigate the neural basis of real emotions. Challenging that view, this article provides an overview of neuroscience studies on music and emotion, showing that activity in each and every so-called limbic and paralimbic brain structures can be modulated by listening to music, in both musically trained and untrained individuals. Therefore, music is a well-suited tool to investigate the neural corre- lates of emotion. A particular advantage of music is that it enables researchers to study a range of positive emotions (such as fun, joy and ‘‘chills’’), some of which are otherwise difficult to evoke in experimental settings. Moreover, studying the neural correlates of emotions with music also has direct relevance for music-therapeutic appli- cations (Box 2).

The following sections outline how neuroscience studies on music and emotion have advanced our understanding of the functional significance of different limbic and paralim- bic structures, and thus our understanding of emotion in general. This overview will mainly deal with functional neuroimaging and lesion studies (for EEG studies see Refs. [4–7]).

Corresponding author: Koelsch, S. ([email protected]).

1364-6613/$ – see front matter � 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.tics.2010.0

Limbic and paralimbic correlates of music-evoked emotions Although not well defined, ‘‘limbic’’ and ‘‘paralimbic’’ struc- tures are considered as core structures of emotional pro- cessing, because their lesion or dysfunction is associated with emotional impairment [8]. How limbic (e.g. amygdala and hippocampus) and paralimbic structures (e.g. orbito- frontal cortex, parahippocampal gyrus and temporal poles) interact, and which functional networks they form is still not well understood.

A central structure within the limbic/paralimbic neural circuitry is the amygdala, which has been implicated in the initiation, generation, detection, maintenance and termin- ation of emotions that are assumed to be important for the survival of the individual [9]. Several functional neuroima- ging [10–16] and lesion studies [17–19] have shown invol- vement of the amygdala in emotional responses to music (Figure 1). The first neuroimaging study showing activity changes in the amygdala was a positron emission tom- ography (PET) experiment by Blood and Zatorre [10], in which changes in regional cerebral blood flow (rCBF) were measured during ‘‘chills’’ (i.e. intense emotional experi- ences involving sensations such as goose bumps or shivers down the spine). Each participant listened to a piece of their own favorite music to which they usually had a chill experience. Increasing chill intensity correlated with rCBF decrease in the amygdala as well as the anterior hippo- campal formation. An increase in rCBF correlating with increasing chill intensity was observed in the ventral striatum, the midbrain, the anterior insula, the anterior cingulate cortex and the orbitofrontal cortex (the func- tional significance of these structures is discussed in the following sections; see Refs. [20–22] for patient studies on music-evoked pleasure).

Even if individuals do not have intense ‘‘chill’’ experi- ences,musiccan evokeactivitychangesintheamygdala,the ventral striatum and the hippocampus. Investigating the emotional valence dimension with music, Koelsch et al. [11] compared brain responses to joyful instrumental tunes (played by professional musicians) to those evoked by elec- tronically manipulated, permanently dissonant counter- parts of these tunes (for other studies using consonant and dissonant music see Refs. [7,12,23–25]). During the presentation of pleasant music, increases in blood–oxygen level dependent (BOLD) signals were observed in the ven- tralstriatum(presumablythenucleusaccumbens,NAc)and the anterior insula (among other structures). Dissonant music, by contrast, elicited increases in BOLD signals in the amygdala, the hippocampus, the parahippocampal

1.002 131

Box 1. Social functions of music: The seven Cs

Humans have a need to engage in social activities; emotional effects

of such engagement include fun, joy and happiness, whereas

exclusion from this engagement represents an emotional stressor

and has deleterious effects on health [59,60]. Making music is an

activity that involves several social functions: (1) when we make

music, we make contact with other individuals (preventing social

isolation); (2) music automatically engages social cognition [61]; (3)

it engages co-pathy in the sense that interindividual emotional

states become more homogeneous (e.g. reducing anger in one

individual and depression or anxiety in another), thus promoting

interindividual understanding and decreasing conflicts [62]; (4)

music involves communication (notably, for infants and young

children, musical communication during parent–child singing of

lullabies and play songs is important for social and emotional

regulation, as well as for social, emotional and cognitive develop-

ment [63,64]); (5) music making also involves coordination of

movements (requiring the capability to synchronize movements to

an external beat) [65–67]. The coordination of movements in a group

of individuals appears to be associated with pleasure (e.g. when

dancing together), even in the absence of an explicit shared goal; (6)

performing music also requires cooperation (involving a shared goal

and increasing interindividual trust); notably, engaging in cooperative

behavior is an important potential source of pleasure [68,69] and; (7)

as an effect, music leads to increased social cohesion of a group [70],

fulfilling the ‘‘need to belong’’ [71], and the motivation to form and

maintain interpersonal attachments [60,72]. Social cohesion also

strengthens the confidence in reciprocal care (see Ref. [64] for the

caregiver hypothesis) and the confidence that opportunities to

engage with others in the cited social functions will also emerge in

the future. Music seems to be capable of engaging all of the ‘‘Seven

Cs’’ at the same time, which is presumably part of the emotional

power of music. In this regard, music does serve a goal, namely the

goal to fulfill social needs that are of vital importance for the

individual. Therefore, the notion that music evokes only aesthetic

experiences without goal relevance is doubtful.

Box 2. Relevance for therapy

Music therapy (MT) can have effects that improve the psychological

and physiological health of individuals. A heuristic working factor

model for music therapy [72] assumes five factors which contribute

to the effects of MT. These factors refer to the modulation of

emotion, attention, cognition, behavior and communication.

Given that music can change activity in brain structures that

function abnormally in patients with depression (such as amygdala,

hippocampus and nucleus accumbens; see main text), it seems

plausible that music can be used to stimulate and regulate activity in

these structures (either by listening to or by making music), and

thus ameliorate symptoms of depression. However, so far the

scientific evidence for effectiveness of MT on depression is

surprisingly weak, because of the lack of high-quality studies, and

the small number of studies with randomized, controlled trials [73].

Studies on neurological applications of MT have so far mainly

dealt with the therapy of stroke patients. Recent evidence suggests

that playing melodies either with the hand on a piano, or with the

arm on electronic drum pads that emit piano tones, helps stroke

patients to train fine as well as gross motor skills with regard to

speed, precision and smoothness of movements [74]; it seems likely

that an emotional component contributes at least partly to these

effects, because this treatment was more effective than a standard

rehabilitation. Electrophysiological data suggest that these effects

are due to enhanced cortical connectivity and stronger activation of

the motor cortex as a result of music-supported movement training

[75].

Other studies showed that isometric musical stimuli have the

capability of regulating gait and arm control in patients with stroke

and Parkinson’s disease, presumably as a result of music-evoked

arousal and priming of the motor system via auditory stimulation,

as well as a result of entrainment of the motor system to the beat of

the music [75,76]. Moreover, positive emotions elicited by preferred

music can decrease visual neglect (possibly by increasing atten-

tional resources) [77], and listening to self-selected music after

stroke appears to improve recovery in the domains of verbal

memory and focused attention (along with less depressed and

confused mood) [78]. The neural mechanisms for such effects,

however, remain to be specified.

Review Trends in Cognitive Sciences Vol.14 No.3

gyrus and the temporal poles (and decreases of BOLD signals were observed in these structures in response to the pleasant music). Notably, patients with unilateral resec- tion of the medial temporal lobe including the parahippo- campal cortex show diminished emotional sensitivity to dissonant music [24,25], consistent with activity changes within the parahippocampal gyrus observed in functional neuroimaging studies using stimuli with varying degrees of dissonance [11,23]. The results of these studies [11,23–25] suggest a specific role of the mid-portion of the parahippo- campalgyrusfortheprocessingofacousticroughness,which is perhaps also relevant for the decoding of the affective content of vocal signals.

In an attempt to investigate the neural correlates of sadness, fear and joy, Baumgartner et al. [14] observed that auditory information interacts with visual infor- mation in several limbic and paralimbic structures, in- cluding the amygdala and the hippocampus (for other studies using joyful and sad music see Refs. [24–26]): activity changes in these structures were stronger during the combined presentation of fearful or sad photographs with fearful or sad music, compared to when only visual information was presented. The combined presentation of music and photographs also elicited stronger activation in the parahippocampal gyrus, and the temporal poles. Activity changes in the amygdala, hippocampal formation, parahippocampal gyrus and temporal poles were also found in two other functional magnetic resonance imaging (fMRI) studies [11,27], suggesting that these structures

132

form a network which plays a prominent role in emotional processing (blue-colored structures in Figure 2).

The findings of Baumgartner et al. [14] received support from a study by Eldar et al. [13], who showed that when eitherpositive(joyful) ornegative(fearful) musicwasplayed simultaneously with an emotionally neutral film clip, it evoked stronger signal changes in the amygdala, and in areas of the ventrolateral frontal cortex, compared to when only music or only film clips were presented. Moreover, the combination of negative (but not positive) music and neutral film clips evoked stronger signal changes in the anterior hippocampal formation compared to when only music or only film clips were presented. Subjective ratings showed that the music plus film conditions were not perceived as significantly more positive or negativethan when music was presentedalone.Therefore,thefunctionalsignificanceofthe increase in signal change (in the amygdala and hippocampal formation) remains unclear. However, the findings that the visualsystemmodulatessignalchangesintheamygdalaare corroborated by data showing that simply closing the eyes during listening to fearful music also leads to increased amygdalar activity [16].

Another important finding by Eldar et al. [13] was that activity changes in the amygdala were observed in response to both positive and negative stimulus combi- nations. This supports the view that the amygdala is not

Figure 1. Illustration of some structures belonging to the limbic/paralimbic

system. The diamonds represent music-evoked activity changes in these

structures (see figure legend for references). Note the repeatedly reported

activations of amygdala, nucleus accumbens and hippocampus, reflecting that

music is capable of modulating activity in core structures of emotion. Top left: view

of the right hemisphere; top right: medial view; bottom left: anterior view; bottom

right: bottom view.

Figure 2. Schematic representation of anatomical connections of some limbic and

paralimbic structures involved in the emotional processing of music (Figure 1 and

main text). ACC: anterior cingulate cortex; ant Ins: anterior insula; Am (BL):

basolateral amygdala; Am (CM) corticomedial amygdala (including the central

nucleus), Hipp: hippocampal formation; NAc: nucleus accumbens; OFC:

orbitofrontal cortex; PH: parahippocampal gyrus; Temp P: temporal pole.

Connectivity is depicted based on Refs. [37,79–81].

Review Trends in Cognitive Sciences Vol.14 No.3

only involved in negative but also in positive emotions [28], clearly challenging the rather simplistic view that the amygdala is primarily a ‘‘fear center’’ in the human brain. Notably, the amygdala is not an anatomical unity: it is composed of several distinct nuclei (the lateral, basal, accessory basal, central, medial and cortical nuclei), and although the amygdala has become one of the most inten- sely studied brain structures, the functional significance of these nuclei, as well as their interaction with other struc- tures, is not well understood [29]. A music study by Ball et al. [12] was the first to provide insight into different functional properties of different subregions of the human amygdala in response to auditory stimulation. This study used original (mainly consonant) piano pieces as pleasant stimuli, and permanently dissonant versions of these stimuli as unpleasant stimuli (similar to other studies investigating the valence dimension with music [7,11,23]). The authors investigated signal changes in the amygdala in response to both consonant and dissonant music. A BOLD signal increase was observed in the baso- lateral amygdala (to both types of music), and signal decrease in a superior region of the amygdala. Also using consonant and dissonant music, Fritz and Koelsch [27] reported BOLD signal decreases with increasing emotional valence in a central aspect of the amygdala (presumably

lateral and/or basal nuclei), whereas BOLD signals increased with increasing valence in a superior aspect of the amygdala (including the substantia innominata). Importantly, the central aspect of the amygdala was found to be functionally connected to the temporal pole, the hippocampus and the parahippocampal gyrus, whereas the superior aspect of the amygdala (presumably the cor- ticomedial amygdala) was functionally connected with the ventral striatum and the orbitofrontal cortex. This suggests that different nuclei of the amygdala are involved in modulating activity of different emotion networks.

As mentioned above, the amygdala and related limbic structures play a critical role for emotions that are assumed by some to have survival value for the individual and the species [9]. The studies cited in this section provide compel- ling evidence that music can evoke activity changes in these brain structures, suggesting that at least some music- evoked emotions involve the very core of evolutionarily adaptive neuroaffective mechanisms. This challenges the notion that music-evoked emotions are merely illusions, rather than real emotions (reviewed in Refs. [2,3]; Box 1). The next section provides further support for the view that music is capable of evoking real emotions by illustrating neural correlates of music-evoked pleasure.

133

Review Trends in Cognitive Sciences Vol.14 No.3

Music affects dopaminergic neural activity Several studies have shown that listening to pleasant music activates brain structures implicated in reward and experiences of pleasure (perceived pleasantness evoked by the engagement in social functions during mak- ing, and listening to, music is addressed in Box 1). Blood and Zatorre [10] reported that the ventral striatum (pre- sumably the NAc; Figure 1) is involved in intensely plea- surable ‘‘chill’’ responses to music. Similarly, another PET study by Brown et al. [30] reported activation of the ventral striatum (in addition to the subcallosal cingulate cortex, the anterior insula and the posterior part of the hippo- campus) during listening to two unfamiliar, pleasant pieces contrasted with a resting condition. Activation of the ventral striatum in response to pleasant music was also observed in three studies using fMRI (one investigated the valence dimension [11], another examined differences in pleasantness as a result of the predictability of music [31], and the third investigated music-evoked memories [32]). One of these studies [31] reported that activation of the ventral striatum was connected to activity in the ventral tegmental area (VTA) and the hypothalamus. This suggests that the hemodynamic changes observed in the ventral striatum reflected dopaminergic activity: the NAc is innervated in part by dopaminergic brainstem neurons (located mainly in the VTA as well as in the substantia nigra) and is part of the so-called ‘reward circuit’ [33]. This circuit includes projections from the lateral hypothalamus via the medial forebrain bundle to the mesolimbic dopa- mine pathway involving the VTA with projections to the NAc [34]. Further support for the assumption that the hemodynamic changes in the ventral striatum reported in Refs. [10,11,30–32] involved dopaminergic neural activity stems from a recent PET study [35] showing that strong music-evoked pleasure (including ‘‘chill’’ experi- ences) lead to increased dopamine binding in the NAc.

Importantly, activity in the NAc (as well as activity in the ventral pallidum [33]) correlates with motivation- and reward-related experiences of pleasure, for instance during the process of obtaining a goal, when an unexpected reach- able incentive is encountered, or when individuals are presented with a reward cue (reviewed in Refs. [33,36]). In humans, NAc activity has been reported for sexual activity, intake of drugs, eating of chocolate and drinking water when dehydrated [33,36]. It has, therefore, pre- viously been suggested that NAc activity correlates with the subjective experience of fun [3], but more detailed information about the functional significance of the NAc is needed to determine the role that the NAc possibly plays for other emotions as well.

The NAc also appears to play a role in invigorating, and perhaps even selecting and directing, behavior in response to stimuli with incentive value, as well as in motivating and rewarding such behavior [36]. The NAc is considered as a ‘limbic motor interface’ [37], because (i) the NAc receives input from limbic structures such as amygdala and hippocampus; (ii) injecting dopamine in the NAc causes an increase in locomotion; and (iii) the NAc projects to other compartments of the basal ganglia, which play an important role for the learning, selection and execution of actions. This motor-related function of the NAc puts it in a

134

key position for the generation of a drive to move to, join in and dance to pleasant music, although the neural basis for this drive needs to be specified.

It is important to note that in three of the cited studies [11,30,31] participants did not report chill responses during music listening, suggesting that dopaminergic pathways including the NAc can be activated by music as soon as it is perceived as pleasant (i.e. even in the absence of extreme emotional experiences involving chills). Results from the reviewed studies indicate that music can easily evoke experiences of pleasure, or fun, associated with the activity of a reward pathway involving the hypo- thalamus, the VTA and the NAc. This emotional power of music needs to be explored to provide more systematic knowledge that could be used in support of the therapy of affective disorders related to anhedonia (such as depress- ive disorders or Parkinson’s disease; Box 2). It has pre- viously been argued that music cannot only evoke subjective experiences of fun (involving the NAc) but also experiences of joy and happiness [3]. The next section puts forward the hypothesis that the latter experiences involve different neural systems than those involved in experi- ences of fun.

Music and the hippocampus Compared to studies investigating emotion with stimuli such as emotional faces, affective pictures, pain stimuli or reward stimuli, the number of studies reporting activity changes within the (anterior) hippocampal formation in response to music [10,11,13,14,26,27,30] is remarkably high (Figure 1). It is well established that the hippocampus plays an important role for learning and memory [38], as well as for novelty and expectedness [39] (for relations between music-evoked emotions and memory processes see Refs. [32,40,41]). However, at least in some of the func- tional neuroimaging studies that used music to investigate emotion, it is unlikely that the hippocampal activations were simply a result of memory processes. For example, in the fMRI study by Mitterschiffthaler et al. [26], in which sad (as compared to neutral) music elicited changes in the anterior hippocampal formation, participants were prob- ably comparably familiar with neutral and sad pieces. Similarly, participants were presumably equally unfami- liar with the happy and fearful musical pieces used in the study by Eldar et al. [13].

Therefore, studies on music and emotion remind us of James W. Papez’s view that the hippocampus also plays an important role for emotional processes [42], a notion which is at least around 70 years old, but has unfortunately fallen into abeyance. The hippocampus has dense reciprocal connections with structures involved in the regulation of behaviors essential for survival (such as ingestive, repro- ductive and defensive behaviors), and with structures involved in the regulation of autonomic, hormonal and immune system activity [37]. Such structures include the amygdala, hypothalamus, thalamic nuclei, the sep- tal-diagonal band complex, the cingulate gyrus, the insula and autonomic brain stem nuclei (Figure 2). Efferent con- nections project to the NAc, other parts of the striatum, as well as to numerous other limbic, paralimbic and non- limbic structures [37]. The functional significance of these

Box 3. Outstanding questions

� How can the emotion-evoking power of music be used in the therapy of affective disorders related to anhedonia, such as

depressive disorders, or of Parkinson’s disease?

� What are the neural correlates of the mechanisms underlying the evocation of emotions by music?

� How (and why) does the interaction of music and visual information impact on activity of the amygdala and the hippo-

campus?

� What is the neural basis of the drive to move and dance to (pleasant) music?

� Can modulation of hippocampal activity (and possible upregula- tion of neurogenesis in the hippocampus) with music be used for

the therapy of depressed patients and patients with PTSD?

� What role does the ACC play for the synchronization of biological subsystems?

� What is the neural basis of beneficial effects of music listening in stroke patients?

Review Trends in Cognitive Sciences Vol.14 No.3

connections places the hippocampus (along with the amyg- dala and the orbitofrontal cortex) in a pivotal position for emotional processing, and it has previously been noted that the key to understanding the function of the hippo- campus lies in the fact that it has major projections not only to cortical association areas but also to subcortical limbic structures [37].

The notion that the hippocampus is involved in emotional processes (in addition to its more cognitive functions such as memory and spatial representation) is supported by significant empirical evidence. First, individ- uals with depression show structural as well as functional abnormality of the hippocampus (reviewed in Refs. [43,44]). Second, the hippocampus is unique in its vulner- ability to emotional stressors. In animals, chronic stress related to helplessness and despair leads to death of hippocampal neurons and related hippocampal atrophy [44], consistent with studies on humans that show reduced hippocampal volume in individuals suffering from child- hood sexual abuse [45] and post-traumatic stress disorder (PTSD) [46]. The loss of hippocampal volume during and after emotional traumatization, or during depression, is assumed to be partly due to both a downregulation of neurogenesis in the hippocampus and death of hippo- campal neurons [44]. Third, activity changes in the anterior hippocampal formation (as well as in the amyg- dala) in response to pleasant and unpleasant music are reduced in individuals with reduced ‘‘tender positive emo- tionality’’ (i.e. with reduced capability of producing tender positive feelings that can be described as soft, loving, warm and happy) compared to individuals of a normal control group [47].

Although only little specific information about the invol- vement of the hippocampus in the processing of emotions is yet available, the results of the study by Koelsch et al. [47] (as well as of studies showing hippocampal dysfunction and structural damage in depressive individuals) encourage the hypothesis that the hippocampus is a critical structure for the generation of tender, positive emotions, such as joy and happiness. This hypothesis relates feelings of happi- ness (supposedly involving the hippocampus, presumably in connection with the amygdala, parahippocampal gyrus and temporal poles) to a different neural network than the experiences of fun (involving the NAc, the VTA, and the hypothalamus; see previous section). Future neuroimaging studies of emotion should carefully control for familiarity, novelty and memory processes elicited by different stimulus categories to rule out the possibility that hippo- campal activations are due to such factors.

Notably, owing to the capability of music to evoke activity changes in the hippocampus, it is conceivable that music therapy with depressed patients and with PTSD patients has positive effects on the upregulation of neuro- genesis in the hippocampus, but this is still an open ques- tion (Box 3).

It is also worth noting that because of its particular sensitivity to emotional stressors, inhibition of neural pathways projecting to the hippocampus during the per- ception of unpleasant stimuli could represent a sensitive neural mechanism that serves the prevention of potential damage of hippocampal neurons [3,11]. Thus, it is possible

that activity changes observed in the amygdala and the hippocampus during the presentation of unpleasant (or threatening) stimuli is not necessarily a result of the generation of fear (or other unpleasant emotions) but could well reflect inhibitory processes activated to prevent the hippocampus from traumatization during exposure to potentially harmful stimuli.

Effects of music on insular and anterior cingulate cortex activity Current theories of emotion emphasize the association between emotion and changes in physiological arousal (mainly involving changes in autonomic and hormonal activity). Changes in autonomic activity have been reported to be associated with activity changes in the anterior cingulate cortex (ACC) and the insular cortex [48–50], and music studies using PET or fMRI have observed activity changes in both of these structures (during music-evoked chills [10], as well as during experi- ences of fear and sadness [14]). Note, however, that activity changes in the ACC or insular cortex are not necessarily related to emotional processing (for the role of the ACC in performance monitoring and motor activity see Ref. [51]; for movement-related functions of the insula and the involvement of the insula in the perception of speech and music see Ref. [3,52]).

It has recently been proposed [3] that the ACC is involved in the synchronization of biological subsystems (such as physiological arousal, motor expression, motiva- tional processes, monitoring processes and cognitive appraisal; [53]). The synchronization of these subsystems is likely to occur as an effect of every emotional instance and could even be indispensable for subjective emotional experiences (usually referred to as feelings). The ACC is in a unique position to accomplish such synchronization, owing to its involvement in cognition, autonomic nervous system activity, motor activity, motivation and monitoring.

Emotions are usually not only accompanied by auto- nomic but also by endocrine (i.e. hormonal) effects, which, in turn, have effects on immune system function [54,55]. With regard to music, such effects are particularly relevant when they are related to a reduction of stress or ameliora- tion of depression and anxiety [56] (Box 2). However, more

135

Review Trends in Cognitive Sciences Vol.14 No.3

evidence of the emotional effects of music on autonomic, hormonal and immune system activity is needed for music therapy to be systematically employed in the treatment of diseases related to endocrine, autonomic or immune sys- tem dysfunction.

Concluding remarks Despite active research in the area of affective neuro- science, the different roles of various brain regions involved in emotion are still not well understood. This review illustrates that music is an important, perhaps even indis- pensable, tool to gain such knowledge. Future work with music can contribute to the investigation of the neural networks underlying different emotions, with the particu- lar advantage that music can be used to study a range of positive as well as negative emotions.

As yet only little is known about the neural correlates of different psychological processes underlying the evocation of emotion with music (such as emotional contagion [11], musical expectancy [15,57,58] or musical memories [2]). Specific knowledge can be gained by systematically manip- ulating different such processes to identify their neural correlates.

Finally, with regard to music therapy, future work needs to determine which types of music (taking into account individual experiences and preferences) are best suited to stimulate specific limbic and paralimbic brain structures (e.g. the hippocampus in depressive patients or dopaminergic system activity in patients with Parkinson’s disease). Therefore, although several outstanding ques- tions remain, better insight into the neural basis of emotions evoked by music will lead to a better understand- ing of how to employ music in the therapy of affective disorders.

Acknowledgements This work was supported by the German Research Foundation (Deutsche Forschungsgemeinschaft) through grant KO 2266/2-1. Comments from Thomas Fritz, Ulrike Altmann, Sabine Aust, and from three anonymous reviewers, helped to improve this article considerably. Ulrike Altmann considerably helped to design Figure 2.

References 1 Sloboda, J.A. (1992) Empirical studies of emotional response to

music. In Cognitive Bases of Musical Communication (Riess-Jones, M. and Holleran, S., eds), pp. 33–45, American Psychological Association

2 Juslin, P.N. and Västfjäll, D. (2008) Emotional responses to music: the need to consider underlying mechanisms. Behav. Brain Sci. 31, 559– 575

3 Koelsch, S. et al. (2010) Functional neuroimaging. In Music and Emotion (Juslin, P. and Sloboda, J.A., eds), p. 975, Oxford University Press

4 Schmidt, L.A. and Trainor, L.J. (2001) Frontal brain electrical activity (EEG) distinguishes valence and intensity of musical emotions. Cogn. Emot. 15, 487–500

5 Altenmüller, E. et al. (2002) Hits to the left, flops to the right: different emotions during listening to music are reflected in cortical lateralisation patterns. Neuropsychologia 40, 2242–2256

6 Baumgartner, T. et al. (2006) From emotion perception to emotion experience: emotions evoked by pictures and classical music. Int. J. Psychophysiol. 60, 34–43

7 Sammler, D. et al. (2007) Music and emotion: electrophysiological correlates of the processing of pleasant and unpleasant music. Psychophysiology 44, 293–304

136

8 Dalgleish, T. (2004) The emotional brain. Nat. Rev. Neurosci. 5, 583– 589

9 Price, J.L. (2005) Free will versus survival: brain systems that underlie intrinsic constraints on behavior. J. Comp. Neurol. 493, 132–139

10 Blood, A. and Zatorre, R.J. (2001) Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion. Proc. Natl. Acad. Sci. U. S. A. 98, 11818–11823

11 Koelsch, S. et al. (2006) Investigating emotion with music: an fMRI study. Hum. Brain Mapp. 27, 239–250

12 Ball, T. et al. (2007) Response properties of human amygdala subregions: evidence based on functional MRI combined with probabilistic anatomical maps. PLoS One 3, e307

13 Eldar, E. et al. (2007) Feeling the real world: limbic response to music depends on related content. Cereb. Cortex 17, 2828–2840

14 Baumgartner, T. et al. (2006) The emotional power of music: how music enhances the feeling of affective pictures. Brain Res. 1075, 151–164

15 Koelsch, S. et al. (2008) Amygdala activity can be modulated by unexpected chord functions during music listening. Neuroreport 19, 1815–1819

16 Lerner, Y. et al. (2009) Eyes wide shut: amygdala mediates eyes-closed effect on emotional experience with music. PLoS One 4, e6230

17 Gosselin, N. et al. (2005) Impaired recognition of scary music following unilateral temporal lobe excision. Brain 128, 628–640

18 Gosselin, N. et al. (2007) Amygdala damage impairs emotion recognition from music. Neuropsychologia 45, 236–244

19 Dellacherie, D. et al. (2008) Is the neutral condition relevant to study musical emotion in patients? Music Percept. 25, 285–294

20 Griffiths, T.D. et al. (2004) ‘When the feeling’s gone’: a selective loss of musical emotion. J. Neurol. Neurosurg. Psychiatry 75, 344–345

21 Stewart, L. et al. (2006) Music and the brain: disorders of musical listening. Brain 129, 2533–2553

22 Matthews, B.R. et al. (2009) Pleasurable emotional response to music: a case of neurodegenerative generalized auditory agnosia. Neurocase 15, 248–259

23 Blood, A.J. et al. (1999) Emotional responses to pleasant and unpleasant music correlate with activity in paralimbic brain regions. Nat. Neurosci. 2, 382–387

24 Gosselin, N. et al. (2006) Emotional responses to unpleasant music correlates with damage to the parahippocampal cortex. Brain 129, 2585–2592

25 Khalfa, S. et al. (2008) Evidence of lateralized anteromedial temporal structures involvement in musical emotion processing. Neuropsychologia 46, 2485–2493

26 Mitterschiffthaler, M.T. et al. (2007) A functional MRI study of happy and sad affective states evoked by classical music. Hum. Brain Mapp. 28, 1150–1162

27 Fritz, T. and Koelsch, S. (2005) Initial response to pleasant and unpleasant music: an fMRI study. Neuroimage 26 (Suppl.), T-AM 271

28 Murray, E.A. (2007) The amygdala, reward and emotion. Trends Cogn. Sci. 11, 489–497

29 LeDoux, J. (2007) The amygdala. Curr. Biol. 17, R868–R874 30 Brown, S. et al. (2004) Passive music listening spontaneously engages

limbic and paralimbic systems. Neuroreport 15, 2033–2037 31 Menon, V. and Levitin, D.J. (2005) The rewards of music listening:

response and physiological connectivity of the mesolimbic system. Neuroimage 28, 175–184

32 Janata, P. (2009) The neural architecture of music-evoked autobiographical memories. Cereb. Cortex 19, 2579–2594

33 Berridge, K.C. et al. (2009) Dissecting components of reward: ‘liking’, ‘wanting’, and learning. Curr. Opin. Pharmacol. 9, 65–73

34 Björklund, A. and Dunnett, S.B. (2007) Dopamine neuron systems in the brain: an update. Trends Neurosci. 30, 194–202

35 Salimpoor, V.N. et al. (2008) The rewarding aspects of music listening involve the dopaminergic striatal reward systems of the brain: an investigation with [C11]raclopride PET and fMRI. Neuroimage 47 (Suppl. 1), S39–S41

36 Nicola, S.M. (2007) The nucleus accumbens as part of a basal ganglia action selection circuit. Psychopharmacology 191, 521–550

37 Nieuwenhuys, R. et al. (2007) The Human Central Nervous System: A Synopsis and Atlas, (4th edn), Springer

38 Moscovitch, M. et al. (2006) The cognitive neuroscience of remote episodic, semantic and spatial memory. Curr. Opin. Neurobiol. 16, 179–190

Review Trends in Cognitive Sciences Vol.14 No.3

39 Nadel, L. (2008) The hippocampus and context revisited. In Hippocampal Place Fields: Relevance to Learning and Memory (Mizumori, S., ed.), pp. 3–15, Oxford University Press

40 Eschrich, S. et al. (2008) Unforgettable film music: the role of emotion in episodic long-term memory for music. BMC Neurosci. 9:48

41 Jäncke, L. (2008) Music, memory and emotion. J. Biol. 7:21 42 Papez, J. (1995) A proposed mechanism of emotion. 1937 [classical

article]. J. Neuropsychiatry Clin. Neurosci. 7, 103–112 43 Videbech, P. and Ravnkilde, B. (2004) Hippocampal volume and

depression: a meta-analysis of MRI studies. Am. J. Psychiatry 161, 1957–1966

44 Warner-Schmidt, J.L. and Duman, R.S. (2006) Hippocampal neurogenesis: opposing effects of stress and antidepressant treatment. Hippocampus 16, 239–249

45 Stein, M.B. et al. (1997) Hippocampal volume in women victimized by childhood sexual abuse. Psychol. Med. 27, 951–959

46 Bremner, J.D. (1999) Does stress damage the brain? Biol. Psychiatry 45, 797–805

47 Koelsch, S. et al. (2007) A cardiac signature of emotionality. Eur. J. Neurosci. 26, 3328–3338

48 Critchley, H.D. (2005) Neural mechanisms of autonomic, affective, and cognitive integration. J. Comp. Neurol. 493, 154–166

49 Critchley, H.D. et al. (2000) Cerebral correlates of autonomic cardiovascular arousal: a functional neuroimaging investigation in humans. J. Physiol. 523, 259–270

50 Craig, A.D. (2009) How do you feel — now? The anterior insula and human awareness. Nat. Rev. Neurosci. 10, 59–70

51 Cole, M.W. et al. (2009) Cingulate cortex: diverging data from humans and monkeys. Trends Neurosci. 32, 566–574

52 Mutschler, I. et al. (2007) A rapid sound-action association effect in human insular cortex. PLoS One 2, e259

53 Scherer, K.R. (2000) Emotions as episodes of subsystem synchronization driven by nonlinear appraisal processes. In Emotion, Development, and Self-organization: Dynamic Systems Approaches to Emotional Development (Lewis, M.D. and Granic, I., eds), pp. 70–99, Cambridge University Press

54 Dantzer, R. et al. (2008) From inflammation to sickness and depression: when the immune system subjugates the brain. Nat. Rev. Neurosci. 9, 46–56

55 Koelsch, S. and Siebel, W. (2005) Towards a neural basis of music perception. Trends Cogn. Sci. 9, 578–584

56 Koelsch, S. (2009) A neuroscientic perspective on music therapy. Ann. N. Y. Acad. Sci. 1169, 374–384

57 Tillmann, B. et al. (2006) Cognitive priming in sung and instrumental music: activation of inferior frontal cortex. Neuroimage 31, 1771–1782

58 Steinbeis, N. et al. (2006) The role of harmonic expectancy violations in musical emotions: Evidence from subjective, physiological, and neural responses, J. Cogn. Neurosci. 18, 1380–1393

59 Cacioppo, J.T. and Hawkley, L.C. (2003) Social isolation and health, with an emphasis on underlying mechanisms. Perspect. Biol. Med. 46, S39–S52

60 House, J.S. et al. (1988) Social relationships and health. Science 241, 540–545

61 Steinbeis, N. and Koelsch, S. (2009) Understanding the intentions behind manmade products elicits neural activity in areas dedicated to mental state attribution. Cereb. Cortex 19, 619–623

62 Alcorta, C.S. et al. (2008) Ritual harmony: toward an evolutionary theory of music. Behav. Brain Sci. 31, 576–577

63 Fitch, W.T. (2006) The biology and evolution of music: a comparative perspective. Cognition 100, 173–215

64 Trehub, S.E. (2003) The developmental origins of musicality. Nat. Neurosci. 6, 669–673

65 Overy, K. and Molnar-Szakacs, I. (2009) Being together in time: musical experience and the mirror neuron system. Music Percept. 26, 489–504

66 Patel, A.D. et al. (2009) Experimental evidence for synchronization to a musical beat in a nonhuman animal. Curr. Biol. 19, 827– 830

67 Kirschner, S. and Tomasello, M. (2009) Joint drumming: social context facilitates synchronization in preschool children. J. Exp. Child Psychol. 102, 299–314

68 Rilling, J. et al. (2002) A neural basis for social cooperation. Neuron 35, 395–405

69 Tomasello, M. et al. (2005) Understanding and sharing intentions: the origins of cultural cognition. Behav. Brain Sci. 28, 675–691 discussion 691–735

70 Cross, I. and Morley, I. (2008) The evolution of music: theories, definitions and the nature of the evidence. In Communicative Musicality (Malloch, S. and Trevarthen, C., eds), pp. 61–82, Oxford University Press

71 Baumeister, R. and Leary, M.R. (1995) The need to belong: desire for interpersonal attachments as a fundamental human motivation. Psychol. Bull. 117, 497–529

72 Hillecke, T. et al. (2005) Scientific perspectives on music therapy. Ann. N. Y. Acad. Sci. 1060, 271–282

73 Maratos AS. et al. Music therapy for depression. Cochrane Database of Systematic Reviews 2008, Issue 1. Art. No.: CD004517. DOI:10.1002/ 14651858.CD004517.pub2

74 Altenmüller, E. et al. (2009) Neural reorganization underlies improvement in stroke-induced motor dysfunction by music- supported therapy. Ann. N. Y. Acad. Sci. 1169, 395–405

75 Hoemberg, V. (2005) Evidence based medicine in neurological rehabilitation: a critical review. Acta Neurochir. 93, 3–14

76 Thaut, M.H. et al. (2009) Neurologic music therapy improves executive function and emotional adjustment in traumatic brain injury rehabilitation. Ann. N. Y. Acad. Sci. 1169, 406–416

77 Soto, D. et al. (2009) Pleasant music overcomes the loss of awareness in patients with visual neglect. Proc. Natl. Acad. Sci. U. S. A. 106, 6011– 6016

78 Särkämö, T. et al. (2008) Music listening enhances cognitive recovery and mood after middle cerebral artery stroke. Brain 131, 866–876

79 Ongür, D. and Price, J.L. (2000) The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans. Cereb. Cortex 10, 206–219

80 Barbas, H. et al. (1999) Medial prefrontal cortices are unified by common connections with superior temporal cortices and distinguished by input from memory-related areas in the rhesus monkey. J. Comp. Neurol. 410, 343–367

81 Augustine, J.R. (1996) Circuitry and functional aspects of the insular lobe in primates including humans. Brain Res. Rev. 22, 229– 244

137