Accepted Manuscript

From everyday emotions to aesthetic emotions: Towards a unified theory of musical emotions

Patrik N. Juslin

PII: S1571-0645(13)00063-8 DOI: 10.1016/j.plrev.2013.05.008 Reference: PLREV 364

To appear in: Physics of Life Reviews

Received date: 24 April 2013 Accepted date: 15 May 2013

Please cite this article in press as: Juslin PN. From everyday emotions to aesthetic emotions: Towards a unified theory of musical emotions. Phys Life Rev (2013), http://dx.doi.org/10.1016/j.plrev.2013.05.008

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Highlights

• A multi-mechanism framework for emotional responses to music is presented. • The framework can explain both ‘everyday emotions’ and ‘aesthetic emotions’. • Emotions may occur in parallel with aesthetic judgments or as a result of them. • Aesthetic judgments are assumed to involve individual sets of weighted criteria. • The framework can reconcile previously distinct approaches to musical emotions. From Everyday Emotions to Aesthetic Emotions: Towards a Unified Theory of Musical Emotions

Patrik N. Juslin Uppsala University Sweden

Abstract

The sound of music may arouse profound emotions in listeners. But such experiences seem to involve a ‘paradox’, namely that music - an abstract form of art, which appears removed from our concerns in everyday life - can arouse emotions - biologically evolved reactions related to human survival. How are these (seemingly) non-commensurable phenomena linked together? Key is to understand the processes through which sounds are imbued with meaning. It can be argued that the survival of our ancient ancestors depended on their ability to detect patterns in sounds, derive meaning from them, and adjust their behavior accordingly. Such an ecological perspective on sound and emotion forms the basis of a recent multi-level framework that aims to explain emotional responses to music in terms of a large set of psychological mechanisms. The goal of this review is to offer an updated and expanded version of the framework that can explain both ‘everyday emotions’ and ‘aesthetic emotions’. The revised framework – referred to as BRECVEMA – includes eight mechanisms: Brain Stem Reflex, Rhythmic Entrainment, Evaluative Conditioning, Contagion, Visual Imagery, Episodic Memory, Musical Expectancy, and Aesthetic Judgment. In this review, it is argued that all of the above mechanisms may be directed at information that occurs in a ‘musical event’ (i.e., a specific constellation of music, listener, and context). Of particular significance is the addition of a mechanism corresponding to aesthetic judgments of the music, to better account for typical ‘appreciation emotions’ such as admiration and awe. Relationships between aesthetic judgments and other mechanisms are reviewed based on the revised framework. It is suggested that the framework may contribute to a long-needed reconciliation between previous approaches that have conceptualized music listeners’ responses in terms of either ‘everyday emotions’ or ‘aesthetic emotions’.

Key words: Music, Listening, Emotion, Arousal, Aesthetics, Theory

Oh music In your depths we deposit our hearts and souls Thou has taught us to see with our ears And hear with our hearts

Kahlil Gibran

1. Introduction Most people in the Western world hear music every day. In a significant proportion of these episodes, the music arouses an emotion (Juslin et al., 2008), sometimes at a profound level. Yet, emotions in music appear to involve a paradox: On the one hand, we have ‘music’, an abstract form of art that seems distant from our concerns in everyday life and is commonly regarded as a harmless form of leisure activity (Pinker, 1997). On the other hand, we have ‘emotions’, evolved mechanisms which have served important functions in human survival throughout evolution (Plutchik, 1994, Ch. 9). How can these seemingly non-commensurable phenomena be linked together?

A Unified Theoretical Framework 2

At the center of a scientific enterprise to understand emotions to music must lie an attempt to capture the process through which musical sounds are imbued with meaning. As Patel (2008) has pointed out, human beings are unparalleled in their ability to make sense out of sound. An ecological perspective on sound perception suggests that the survival of our ancient ancestors depended on their ability to detect patterns in sounds, derive meaning from them, and adjust their behavior accordingly. Such a perspective forms the basis of a recent framework that aims to explain emotional responses to music in terms of a large set of psychological mechanisms (e.g., Juslin, 2005a; Juslin & Västfjäll, 2008a; Juslin et al., 2010). This review aims to provide an updated and expanded version of the framework that can explain both ‘everyday emotions’ and ‘aesthetic emotions’ (as defined further in Section 1.2).

The reminder of this article is structured as follows. Firstly, I provide ‘working definitions’ of important terms used throughout the review. Secondly, I briefly summarize previous findings with regard to the occurrence of musical emotions to give some sense of the phenomena that a theory of musical emotions must be able to explain. Thirdly, I outline the original BRECVEM framework and its associated predictions. Finally, I propose an additional mechanism in terms of aesthetic judgments, and consider the role of this component vis-à-vis the other components in the framework. I will attempt to show that the revised framework can explain some peculiar findings in the field and also help to reconcile different approaches to music and emotion.

1.1. Terminology The study of musical emotion has suffered from conceptual confusion (e.g., Juslin & Zentner, 2002). To promote clear communication and fruitful debate, it may be heuristic to adopt some conceptual distinctions, based on an increasing consensus in the affective sciences (Davidson, Scherer, & Goldsmith, 2003, p. xiii; Oatley, Keltner, & Jenkins, 2006, pp. 29-31). Thus, Table 1 offers a set of ‘working definitions’ of key terms, from the Handbook of Music and Emotion (Juslin & Sloboda, 2010, Ch. 1), which will be adopted in the present review.

(Insert Table 1 about here)

As seen in Table 1, the term affect is used as an ‘umbrella term’ that covers different affective phenomena. Musical affect may comprise anything from preference and mood to emotion and aesthetic judgment (the latter concept will be more carefully defined in a sub-sequent section). What all these have in common is valence: an evaluative feeling of an object, person, or event as being positive or negative. In addition, many scholars also require some degree of arousal, to distinguish affect from purely cognitive judgments (Frijda & Scherer, 2009). However, the arousal may range from barely noticeable (preference) to rather pronounced (emotion).

The following sections of this review will mainly be concerned with emotions. In this context, one needs to make a distinction between perception and arousal of emotions. We may simply perceive an emotion in the music or we may actually feel an emotion in response to the music (e.g., Gabrielsson, 2002). This review focuses in particular on the emotions aroused by music, which are usually inferred from a combination of verbal self-reports, physiological responses, and emotional expressions (e.g., Lundqvist, Carlsson, Hilmersson, & Juslin, 2009). However, it is important to note that the emotions aroused by a piece of music represent only one out of several aspects of music experience (e.g., physical, perceptual, cognitive, personal, social, and existential aspects; see Gabrielsson, 2011, p. 462) that may or may not occur in any individual instance of music listening. The term prevalence refers to the relative frequency of occurrence of emotional reactions to music in the population of interest (Juslin et al., 2008).

A Unified Theoretical Framework 3

1.2. What to explain: The prevalence of musical emotions A prerequisite for developing a satisfactory theory of musical emotions is to have a sufficient understanding of the phenomena that need to be explained: Which emotions, then, does music arouse in listeners? Although this question might appear simple enough, this is where the plot thickens. Throughout history, there have been two distinct approaches to describe the types of emotions aroused by works of art, including music. These may be termed ‘aesthetic emotions’ and ‘everyday emotions’, respectively.

Many scholars use the term ‘aesthetic emotions’ simply to refer to all emotional responses to art objects (e.g., music, theatre, painting), without implying that the emotions themselves are of a peculiar kind. However, the term has also been used more narrowly to refer to a ‘special’ emotion thought to be aroused when perceivers are engaged with artworks (Bell, 1914). This idea reflects the long-lasting influence of Kant (1790), who distinguished ‘aesthetic pleasure’ from other types of pleasure, the former type being the result of ‘disinterested’ perception and a concern with pure form. Are there emotions had when and only when a work is apprehended aesthetically? Levinson (1997a) states that “such a posit has not lately found favor, nor does it appear to respond to any pressing concern” (p. 31). Yet, the notion of ‘aesthetic emotions’ has re-appeared in recent research on music and emotion – including the claim that music arouses ‘unique’ or ‘music-specific’ emotions (Scherer & Zentner, 2008).

Other theorists have disputed the idea that art evokes some special type of emotion, unrelated to normal emotions (Dewey, 1959). On this view, art mainly arouses ‘everyday emotions’ that are similar to other emotions experienced in daily life and that include discrete emotions such as happiness, sadness, interest and surprise. “Garden-variety emotions, in fact, are what keep audiences engaged with artworks” (Carroll, 1997, p. 191). This view, then, argues that, while the circumstances surrounding art perception might be special, the emotions felt are a subset of the ordinary emotions experienced in life (Levenson, 1997a). Music listening may involve a characteristic frequency-distribution of emotions that is skewed towards positive emotions, but this may be just an example of a more general tendency: that there are different frequency distributions across the spectrum of emotions as a function of the domain sampled, whether it is, say, music, sports, or politics (Juslin, 2011).

Which of these views is correct? Or could it be that there is some truth to both accounts? Until recently, there was hardly any evidence that could resolve this issue. But when the psychology of music abandoned its previous fixation with experimental work, and began to explore music experiences in real-world settings (Hargreaves & North, 1997), researchers were able to obtain preliminary estimates of the prevalence of specific emotions.

First, it needs to be acknowledged that we frequently hear music without actually feeling any emotion at all, at least not one aroused by the music. According to recent estimates, the music arouses emotions only in about 55-65% of the episodes on average, and there seem to be wide individual differences in overall prevalence (e.g., Juslin & Laukka, 2004; Juslin et al., 2008). Second, when music does arouse emotions, survey studies suggest that it arouses quite a wide range of states. Among the most frequently reported emotions to date are the following broad categories: calm-relaxation, happiness-joy, nostalgia-longing, interest-expectancy, pleasure- enjoyment, sadness-melancholy, arousal-energy, love-tenderness, pride-confidence as well as various synonymous terms (Juslin & Laukka, 2004; Juslin et al., 2008, 2011; Sloboda, 1992; Wells & Hakanen, 1991; Zentner, Grandjean, & Scherer, 2008). ‘Mixed’ emotions (e.g., both joy and sadness) also occur, but in a minority of the episodes (13% in Gabrielsson, 2010; 11% in Juslin et al., 2011). A Unified Theoretical Framework 4

What evidence is there of music-unique emotions? Careful consideration of the data obtained so far reveals that the emotions claimed by some scholars to be ‘aesthetic emotions’ unique to music (e.g., wonder, tenderness, nostalgia, tension) may also occur in other everyday contexts that do not involve music, or even works of art more generally. Thus, the notion that there are emotions had when and only when a work of art is apprehended aesthetically is not supported. However, although ‘aesthetic emotions’ may not involve unique states, they might differ with respect to their underlying causal process. That is, we might construe ‘aesthetic emotions’ as emotions aroused by specifically aesthetic properties of a work of art (e.g., admiration for the skill of the artist or the beauty of the composition); that way the concept of ‘aesthetic emotion’ may retain its usefulness in decomposing listeners’ responses to music. For example, Koneþni (2005) suggested that a particular type of musical event will arouse aesthetic awe in listeners.

How common are such emotions? In a recent survey study, featuring a representative sample of the Swedish population (Juslin et al., 2011), ‘aesthetic emotions’ such as wonder and awe were rarely reported by music listeners, perhaps because the focus was on everyday listening. Surprisingly, however, a similar tendency was reported by Gabrielsson (2010) in his study of Strong Experiences with Music (SEM). Gabrielsson observed that, “none of our participants referred to aesthetic experience”, and that Koneþni’s (2005) ‘aesthetic trinity’ (aesthetic awe, being moved, thrills) “would only apply to a minority of the experiences in our collection of SEM reports” (p. 571). For example, a mere 3% of the listeners in the study reported having experienced any of the emotions wonder, admiration, reverence, or respect, though listeners also reported feeling grateful and insignificant, which might perhaps be linked to an aesthetic response. Zentner et al. (2008) reported more frequent occurrence of emotions such as wonder in their sample of self-reports, which involved mainly classical music.

At the current stage of data collection, it appears that ‘everyday emotions’ are more common than ‘aesthetic emotions’ in music listening, particularly if the prevalence estimates are based on representative samples of participants (Juslin et al., 2011) or situations (Juslin et al., 2008). However, the balance between the two types of emotions is likely to vary a lot across various contexts, and is still the focus of on-going research. The most prudent conclusion, which most researchers would presumably accept, is that both types of emotions do occur to some extent.

In sum, the findings from studies so far suggest that music listeners may experience anything from mere arousal, ‘chills’, and ‘basic’ emotions (e.g., happiness, sadness) to more ‘complex’ emotions (e.g., nostalgia, pride), and even ‘mixed’ emotions. ‘Aesthetic’ emotions (e.g., awe, admiration, wonder) also occur, but may be more rare (e.g., Gabrielsson, 2010, Huron, 2006). To the extent that ‘aesthetic emotions’ occur rarely in everyday life, we could be content that a theory of music and emotion that can explain only ‘everyday emotions’ will account for the majority of responses (cf. Juslin et al., 2010). However, a more fruitful approach is to attempt to augment the theory, so that we may be able to account for the whole range of emotions that can occur across different listening contexts.

1.3. How to explain it: The psychological mechanisms

How emotions to music are caused has usually been addressed in one of two ways. One way has been to map the contributing factors in the listener, the music, and the situation that may influence emotions (Gabrielsson, 2010). Another way has been to develop theories about the precise mechanism ‘mediating’ between musical events and experienced emotions. Although the first approach is useful to specify the conditions under which musical emotions will tend to occur, only the second approach can ultimately explain why these emotions occur. A Unified Theoretical Framework 5

Understanding the crucial role of underlying mechanisms in solving the mystery of emotions to music requires a broader consideration of the issue of how, precisely, emotions are evoked. General research on emotions and stress soon discovered that it was difficult to find objective situation predictors that would (invariably) affect different persons in the same way: different persons tend to react in different ways to the ‘same’ stimulus. This realization forms the basis of theories of emotion causation (for a recent, excellent overview, see Moors, 2009). A theory of emotion causation should explain both why a given event arouses an emotion (elicitation) and why the aroused emotion is of a certain kind (differentiation). The psychological process through which this is achieved is referred to as ‘the underlying mechanism’. A description of the psychological process should not be confused with its neural implementation in the brain or with the phenomenological experience it seeks to explain (Dennett, 1987).

Given the crucial role of emotion in music listening, it is a primary goal of music psychology to explain these mechanisms. Still, few articles in the literature so far have proposed or tested any theory about mechanisms (Juslin & Västfjäll, 2008a, p. 560). In fact, apart from Meyer’s (1956) and Berlyne’s (1971) books, most theories focus on the expressive properties of music that enable listeners to perceive emotions in music (Clynes, 1977; Cooke, 1959; Juslin, 2001;

Langer, 1957). Most studies of emotions aroused by music have been untheoretical in nature. Scholars have aimed to obtain direct links between surface features of the music and aroused emotions, but such correlations do not constitute an explanation: they just move the burden of explanation from one level (Why does the 2nd movement of Beethoven’s ‘Eroica’ symphony arouse sadness?) to another level (Why does slow tempo arouse sadness?). Only a description of the process that mediates between surface features and aroused emotions (the mechanism) constitutes a proper explanation.1 The neglect of mechanisms has prevented researchers from explaining individual differences, like, for example, why the same piece of music may arouse different emotions in different listeners (cf. Gowensmith & Bloom, 1997). The solution to this problem is a more theory-driven approach that takes the underlying mechanism into account.

Careful consideration of emotion theories in research outside the musical domain reveals that the most commonly discussed mechanism in the past is cognitive appraisal (Scherer, 1999). Cognitive appraisal refers to a process whereby an emotion is evoked in a person because an event is interpreted as having important implications for the person’s goals (in terms of goal congruence, coping potential, or compatibility with social norms). The problem is that music as such rarely has implications for life goals.2 Indeed, preliminary data suggest that cognitive appraisal is rarely the cause of musical emotions in everyday life (Juslin et al., 2008). Hence, it can be argued that the strong focus on cognitive appraisal in previous emotion research has impeded our understanding of musical emotions. This is reflected in a common though faulty assumption about musical emotions.

1.4. The assumption of realism

Recall the ‘paradox’ that I mentioned at the beginning of this article, namely that music – an abstract form of art, which appears removed from our concerns in everyday life – can arouse emotions – biologically evolved reactions related to human survival. According to the theory given in this review, failure to resolve this paradox is partly due to an assumption that seems implicit in much reasoning about musical emotion. We can call it ‘the assumption of realism’.

1 The belief that objects and events directly cause our emotions is referred to as the nativist fallacy (Silvia, 2012). 2 Of course, music is often used to achieve different goals (e.g., relaxation). To avoid confusion, it must be noted that there is a distinction between using music to achieve a goal (e.g., to get distracted) and a goal being involved in the underlying process through which the music produces its emotional effect. The focus here is on the latter. A Unified Theoretical Framework 6

The assumption is that, because we are aware that ‘music’ is something of no consequence to our goals, our emotion mechanisms should respond (or rather not respond) accordingly. Thus, for example, Kivy (1990) suggests that “music provides neither the objects nor, therefore, the belief-opportunities that would make it possible for musical works to arouse such emotions as anger, sadness, joy” (p. 165). In fact, because such responses to music seem so implausible, it has been concluded that music cannot arouse such emotions. This conclusion runs into trouble however, if we consider prevalence data, which clearly show that music can arouse these very emotions, as well as several others (Juslin et al., 2011). This problem might be resolved, if we follow the lead of Blaise Pascal, who argued that ‘The heart has its reasons … of which reason knows nothing’; that is, emotions have a logic or rationality of their own, apart from conscious thought. To explain musical emotions, we need to uncover this underlying logic.

Rejecting the assumption of realism, one may proceed from the notion that there are multiple mechanisms, and that some of these do not actually have access to, or take into consideration, information about whether the object of attention is ‘music’ or not – the mechanisms respond to certain information, wherever it occurs. As Nesse (2009) notes: “the mechanisms selection shapes to regulate emotion may use any cue correlated with the relevant situation, even if it is not reliably or causally connected” (p. 160). This leads to the intriguing scenario that you may know that what you hear is ‘just music’, but the mechanisms that evoke your emotions do not, which can explain why musical events may arouse even emotions that do not appear to ‘make sense’ in a musical context. In other words, the assumption that at least some mechanisms do not necessarily treat musical stimuli as different from other stimuli opens the door for a novel type of framework that can explain a number of phenomena in responses to music.

2. The BRECVEM Framework: A Recapitulation Although mechanisms have been mostly neglected in previous research, several scholars have discussed possible mechanisms, typically focusing on a few possibilities (e.g., Berlyne, 1971; Dowling & Harwood, 1986; Meyer, 1956; Sloboda, 1998). Scherer and Zentner (2001) offered a useful review of mechanisms and moderators that may be involved in the process, but made no attempt to integrate these or to develop a set of hypotheses about their characteristics.

Juslin and associates (Juslin, 2005a; Juslin & Västfjäll, 2008a) outlined a more comprehensive framework that featured six psychological mechanisms (besides cognitive appraisal), through which music might arouse emotions, singularly or in combination. Juslin et al. (2010) updated this framework, featuring an additional mechanism (suggested in several commentaries on the original version; Juslin & Västfjäll, 2008b, p. 605). Thus, the updated version included seven mechanisms (Brain stem reflexes, Rhythmic entrainment, Evaluative conditioning, Contagion, Visual imagery, Episodic memory, and Musical expectancy), which were collectively referred to as BRECVEM. It was argued that these seven mechanisms, along with cognitive appraisal, could account for the lion’s share of emotions aroused by musical events in everyday life. To pave the way for a later discussion of interactions between mechanisms, I will summarize this framework, before making the case that it needs to be augmented by an additional mechanism (Aesthetic judgment) to account for ‘aesthetic emotions’.

2.1. Emotion architecture: Basic assumptions The point of departure is an evolutionary perspective. I assume that emotions are adaptive because of their ability to quickly evaluate what is happening in the environment, so that the organism can choose a suitable course of action under conditions of limited time, knowledge or computational capacity, based in part at least on what types of activity have been adaptive in the past (e.g., Ekman, 1992; cf. Simon, 1956). A Unified Theoretical Framework 7

Mechanisms of emotion induction are regarded as information-processing devices at different levels of the brain, which utilize distinct types of information to guide future behavior (Juslin, 2011). It is further hypothesized that sound perception has survival value partly because of its capacity to activate these mechanisms. In a sense, the emotions represent an extension of the perceptual process that enables us to infer not only the identity and location of an object, but also its potential consequences or ‘affordances’ (Gibson, 1979; Griffiths, 2004). The emotion induction mechanisms are conceived of as based on a number of (more or less) distinct brain functions that have developed gradually and in a specific order during evolution, from simple sensations and conditioning to complex syntactic processing (Gärdenfors, 2003). Because the mechanisms depend on functions of different evolutionary origins, each mechanism will have unique characteristics that influence its functioning (see predictions in Section 2.3).

Some of the characteristics of the mechanisms may be linked to their mental representations. A representation is a physical state that conveys some meaning or information about the state of the world within a specific processing system (cf. Fox, 2008). Representations have both a specific form and content. Cognitive theorists assume that information may be represented in the brain in several different ways, for instance associative, analogical, sensori-motoric, and propositional representations (e.g., Dalgleish, 2004; Teasdale, 1999). Hence, a basic premise of BRECVEM is that different mechanisms involve different modes of mental representation.

Each mode of representation may have both strengths and weaknesses with regard to its utility in tracking significant aspects of the environment. For instance, simple representations enable mechanisms to accomplish ‘quick and dirty’ processing in response to urgent threats, whereas more complex representations enable mechanisms to accomplish more sophisticated, although relatively slow, processing of environmental cues. An emotional system that employs multiple modes of information brings the advantage that, if one mechanism fails to respond to a critical event, another mechanism can pick up the slack. This is an example of functional redundancy, or the fact that the same behavioral function can be achieved through many different processes in the central nervous system (Norris, Coan, & Johnstone, 2007; see also Brunswik, 1956). On the other hand, a system that features several mechanisms at different levels may also produce conflicting outputs under certain conditions (Griffiths, 2004), hence the occurrence of ‘mixed’ emotions (Larsen & McGraw, 2011).

The above focus on mental representations may seem to suggest that the present framework is incompatible with the perspective of embodied cognition (Shapiro, 2011), although this is not necessarily the case. Consistent with embodied cognition, the present framework assumes (a) that emotions per definition are embodied phenomena that serve to guide action, (b) that there is a continuous interaction between the perceiver and the ecology; and (c) that sensori-motoric links play an essential role in some of the mechanisms. However, following Clark and Toribio (1994), I assume that some mechanisms (e.g., visual imagery) involve adaptive problems that are more ‘representation-hungry’, and that therefore require more elaborate representations of a traditional type (‘strong’ as opposed to ‘weak’ representations; Shapiro, 2011, pp. 152-153).

It is important to note that the emotion-induction mechanisms are not ‘turned off’ just because 3 we hear music. This is because many mechanisms involve low-level processes that developed before music even existed, and that consequently don’t treat ‘music’ as a distinct type of event. I assume that each mechanism may respond in its own manner to information in the music, the

3 Hence, whereas Budd (1985) defined music as “the art of uninterpreted sounds” (p. ix), the present framework rather assumes that music is constantly interpreted. A Unified Theoretical Framework 8 listener, and the situation, collectively referred to here as the ‘musical event’. All mechanisms in the BRECVEM framework (and aesthetic judgments, discussed later) have in common that they involve psychophysical relationships between ‘external’ features of the environment (i.e., the music and the context) and ‘internal’ features of the perceiver. A particular piece of music could feature certain types of information, but whether these bits of information will activate a mechanism depends on characteristics of the listener and the context also. The music ‘affords’ a certain response, but does not guarantee it. Attention is a crucial factor: What aspects of the music are most ‘perceptually salient’? What are the listener’s motives and goals? The fact that responses to music reflect psychophysical relationships may explain both why listeners could respond differently to the same piece of music, and why there could also be some similarities. Key is the psychological mechanism that mediates between the musical event and the listener experiencing the music. The following section offers a summary of the mechanisms included in the original BRECVEM framework (for a more in-depth discussion, see Juslin et al., 2010).

2.2. The BRECVEM mechanisms: A closer look

Brain stem reflex refers to a process whereby an emotion is induced by music because one or more fundamental acoustic characteristics of the music are taken by the brain stem to signal a potentially important and urgent event that needs attention (Davis, 1984). In music, this could involve sounds that are sudden, loud, or dissonant, or that feature accelerating patterns. Brain stem reflexes are quick, automatic, and unlearned. A famous example that may evoke a brain stem reflex in a listener given a certain sound level is the kettledrum stroke in Joseph Haydn’s Symphony No. 94 (Simons, 1996, p. 81). More commonly perhaps, such reactions result from dynamic changes in the music (e.g., a full orchestra or band coming in after a short solo intro). Brain stem reflexes may increase arousal and evoke feelings of surprise in the listener (Juslin, Harmat, & Eerola, in press).

Rhythmic entrainment refers to a process whereby an emotion is evoked by a piece of music because a powerful, external rhythm in the music influences some internal bodily rhythm of the listener (e.g., heart rate), such that the latter rhythm adjusts toward and eventually ‘locks in’ to a common periodicity (Clayton, Sager, & Will, 2005). The adjusted heart rate can then spread to other components of emotions, such as feelings, through ‘proprioceptive feedback’. Harrer and Harrer (1977) observed that listeners tended to synchronize either their heart rate or their breathing to the music, and that one could ‘drive’ their pulse with appropriate music. Such entrainment, stimulated by a strong pulse, as evident in techno music, march music, and certain types of film music, can increase arousal, but may also arouse feelings of communion (e.g., ‘feeling connected’; Demos et al., 2012; and ‘emotional bonding’, Levitin, 2010, p. 56).

Evaluative conditioning (EC) refers to a process whereby an emotion is induced by a piece of music simply because this stimulus has often been paired with other positive or negative stimuli. For example, a particular piece of music may have occurred repeatedly together in time with a specific event that always makes you happy, such as meeting your friends. Over time, through repeated pairing, the music itself will eventually arouse happiness, even in the absence of the friendly interaction. The use of melodic themes to evoke emotions associated with certain characters or events was part of Wagner’s Leitmotif strategy. Such conditioning has so far been investigated only to a limited extent in marketing (Blair & Shimp, 1992; see also Bolders, Band, & Stallen, 2012). Evaluative conditioning is interesting since it involves subconscious, unintentional, and effortless processes that can be subtly affected by mundane musical events (e.g., Razran, 1954).

A Unified Theoretical Framework 9

Emotional contagion refers to a process whereby an emotion is induced by a piece of music because the listener perceives the emotional expression of the music, and then ‘mimics’ this expression internally. The idea is that we become aroused by voice-like features of the music because a ‘brain module’ responds automatically to certain stimulus features as if they were coming from a human voice expressing emotions (Juslin, 2001), possibly through some kind of ‘mirror-neuron system’ (Rizzolatti & Craighero, 2004). Preliminary support for the notion that responses to music may involve an inner imitation of voice characteristics comes from a brain imaging study by Koelsch et al. (2006) that indicated that listening to expressive music activated brain regions associated with pre-motor representations for vocal sound production. Robinson (2005) argued that this mechanism has limited application, since “it applies only to music that sounds like vocal expression” (p. 388). She seems to forget that most music heard today is vocal music. Moreover, recent findings suggest that voice-like features of a violin or cello can arouse basic emotions such as sadness in listeners, if the music matches the features of a corresponding emotional expression (Juslin et al., in press).

Visual imagery refers to a process whereby an emotion is evoked in the listener because he or she conjures up inner images (e.g., of a beautiful landscape) while listening to the music. The listener is conceptualizing the musical structure in terms of a nonverbal mapping between the metaphorical ‘affordances’ of the music and ‘image-schemata’, grounded in bodily experience

(Kövecses, 2000; Lakoff & Johnson, 1980). Music can be quite effective in stimulating visual imagery (Osborne, 1980), though there are wide individual differences among listeners: Some experience it regularly, whereas others hardly experience it at all. Still, when it occurs, it may arouse feelings of pleasure (Juslin et al., 2008) and deep relaxation (McKinney et al., 1997), with far-reaching implications for clinical applications.

Episodic memory refers to a process whereby an emotion is induced in a listener because the music evokes a personal memory of a specific event in the listener’s life (Baumgartner, 1992). This has been referred to as the ‘Darling, they are playing our tune’ phenomenon (see Davies, 1978). When the memory is evoked, so is also the emotion associated with the memory. Such emotions may be intense - perhaps because the physiological response patterns to the original events are stored in memory, along with the content (Lang, 1979). Studies show that episodic memories linked to music often arouse emotions such as nostalgia (Janata et al., 2007; Juslin et al., 2008).4 Furthermore, listeners may take pride in music which is consistent with a sense of self-identity and belonging (Huron, 2009; see also DeNora, 2000). The bitter-sweet longing for something valuable that is gone, but might one day return, characteristic of the Portuguese Fado genre and its feelings of saudade, is also strongly linked to episodic memories.

Musical expectancy refers to a process whereby an emotion is induced in a listener because a specific feature of the music violates, delays, or confirms the listener’s expectations about the continuation of the music. The expectations are based on the listener’s previous experience of the same musical style (Pearce et al., 2010), as argued by Meyer (1956). His influential theory received support from an investigation by Steinbeis, Koelsch, and Sloboda (2006), which used subjective and physiological measures to capture emotional reactions to unexpected harmonic progressions in a Bach chorale (see also Sloboda, 1991). Musical emotions related to violation of expectancies might include anxiety (Meyer, 1956, p. 27) and surprise (Huron, 2006, p. 348, p. 356), as well as thrills (Sloboda, 1991).

4 “Why did Odysseus weep” in response to the song of the great bard, Demodocus, asks Bicknell (2009), and answers “because the music mattered to him...it connected him with his past and reminded him of significant events in his personal history” (p. 149). A Unified Theoretical Framework 10

2.3.Theoretical predictions: An update By synthesizing research in various domains outside music, Juslin and Västfjäll (2008a) were able to offer the first set of hypotheses that could help music researchers to distinguish among the various mechanisms. Table 2 presents an updated version of these hypotheses that features the added mechanism rhythmic entrainment, as well as some new dimensions along which the seven mechanisms can be differentiated. The hypotheses can serve to guide future studies and will also be revised in the light of findings from such studies.

(Insert Table 2 about here)

Evolutionary order refers to the approximate order in which the seven brain functions may be hypothesized to have appeared during evolution (see Gärdenfors, 2003; Joseph, 2000; see also Donald, 1991, 2001; Reber, 1993; Tulving, 1983). First came simple sensations and an ability to direct attention (as reflected in brain stem reflexes), along with physical coupling processes (as reflected in rhythmic entrainment) – both of which occur in some form even in very simple organisms. Then developed perception (i.e., interpreted sensory impressions), which is crucial for all remaining mechanisms, followed by a simple capacity for learning associations among events (as reflected in conditioning). Perception and conditioning have been observed even in flies, worms, snails, and fish (LeDoux, 2000). However, the ability to respond empathically to vocal emotion expressions of conspecifics (as reflected in emotional contagion) is a somewhat 5 later development associated with social mammals and the ‘the limbic system’ (Ploog, 1992). Later still is the development of an inner imagination (as reflected in imagery), which requires mental representations that are ‘detached’ from direct sensory stimulation (Gärdenfors, 2003), and that appeared along with development of the neocortex of the brain. Such representations enable ‘time travel’ in an organism. This might involve the ability to imagine future events or the ability to recall specific memories of previous events in one’s life (as reflected in episodic memory). Episodic memories require detached representations, as well as self-consciousness (i.e., perceptions of an inner world and a sense of self that is separate from the external world). A conception of time and an inner world are essential to the ability to plan (e.g., sequences of action) and narrate (e.g., to order elements into a coherent story), which paves the way for the development of a complex language. Language presupposes the previous brain functions, and also requires the capability for syntactic processing (as reflected in musical expectancy) to be able to handle a complicated grammar (Schoenemann, 1999).

Survival value of brain function describes the most important benefit that each brain function originally brought to those organisms that possessed this function. Thus, for example, visual imagery allowed humans to ‘simulate’ important events internally by means of self-conjured mental images in the absence of direct sensory input, which meant that overt and potentially dangerous actions could be evaluated internally before they were implemented in the external world (Gärdenfors, 2003).

Information focus specifies broadly the kind of information that each induction mechanism is processing. Thus, for instance, evaluative conditioning focuses on covariation between events (Reber, 1993), whereas contagion focuses on motor expression (Preston & de Waal, 2002).

Mental representation describes the mode of mental representation utilized by the respective mechanism. Each psychological mechanism represents a unique form of input that may guide

5 The term empathy refers in this context simply to a process through which an organism comes to feel the same emotion as another organism (e.g., by means of motor mimicry). A Unified Theoretical Framework 11

6 decision making regarding future behavior. Thus, for instance, contagion involves a sensori- motoric representation, whereas musical expectancy involves a schematic representation.

Key brain regions describes the brain regions that have been most consistently associated with each mechanism in neuroimaging studies. These hypotheses were primarily based on imaging studies of each mechanism in non-musical contexts, on the assumption that these studies have used more ‘process-pure’ tasks and stimuli (Cabeza & Nyberg, 2000) than most music studies conducted so far (Koelsch, 2012; but see Janata, 2009). In principle, musical emotions may be expected to involve three classes of brain regions: (1) regions involved simply because music is perceived (e.g., the primary auditory cortex); (2) regions usually involved in the conscious experience of emotions, regardless of the exact cause of the emotion (e.g., the rostral anterior cingulate and the medial prefrontal cortex; Lane, 2000, pp. 356-358; PAG, Panksepp, 1998); and (3) regions involved in emotional information-processing that might differ depending on the mechanism inducing the emotion. Thus, although musical emotions involve several brain regions, the hypotheses in Table 1 focus only on the last type of regions - those that may help researchers to discriminate between mechanisms. The hypotheses should not be interpreted as implying that each mechanism is located in a separate region. Rather, each mechanism might involve a particular ‘network’ of dynamically interacting brain regions (cf. Bressler & Menon, 2010), some of which might overlap for different mechanisms.

Cultural impact and learning refers to the relative extent to which each mechanism is affected differently by music that differs from one culture to another. For example, brain stem reflexes reflect mainly ‘hard-wired’ reactions to simple features that are not affected much by learning, whereas musical expectancy reflects learned schemata for specific styles of music that differ from one culture to another and that make listeners from different cultures react differently to the same piece of music (Juslin, 2012).

Ontogenetic development concerns the approximate time in the development when respective mechanism might begin to have a noticeable effect on musical emotions. For example, brain stem reflexes to music seem to be functional even prior to birth, whereas responses involving musical expectancy do not develop fully until between the ages of 5 and 11 (Sloboda, 1989).

Induced affect specifies which affective states might be expected to be aroused depending on the mechanism. For example, whereas emotional contagion might be expected to arouse only ‘basic’ emotions that have more or less distinct nonverbal expressions, visual imagery might be expected to arouse all possible human emotions (e.g., Lane, 2000, pp. 362-363).

Temporal focus of affect specifies whether the emotions aroused by respective mechanism are oriented mainly toward the present (e.g., surprise for brain stem reflex), toward the past (e.g., nostalgia for episodic memory), or toward to the future (e.g., anxiety for musical expectancy). This is related to the extent of ‘time travel’ enabled by the mechanism (Evolutionary order).

Induction speed refers to how much processing time the mechanism requires, relative to other mechanisms, for an emotion to occur in a specific situation. For example, brain stem reflexes can induce emotions very quickly (in less than a second), whereas musical expectancy can be expected to require more time (a number of seconds), since some of the musical structure has to unfold in order for a musical expectation to occur that can be confirmed or violated.

6 The idea that the mechanisms bring distinct types of information to the table could explain why they all remain in the behaviour repertoire despite the development of more complex forms of reasoning at a propositional level. A Unified Theoretical Framework 12

Degree of volitional influence refers to the extent to which the listener him- or herself could actively influence the induction process (e.g., through focus of attention, active recall, self- activation). For instance, emotional reactions that involve evaluative conditioning might be involuntary and ‘automatic’, whereas reactions that involve visual imagery may be strongly influenced by the way a listener actively chooses to entertain some inner images and themes rather than others (Larsen, 1995).

Availability to consciousness states the extent to which at least some aspects of the induction process are available to the listener’s consciousness, so that the listener may be able to explain his or her response. For example, if a piece of music evokes an episodic memory, the listener will have a conscious recollection of the previous event and some inkling of the reasons (e.g., the appraisal) that made this event evoke the emotion that is now re-experienced. Conversely, evaluative conditioning responses to music can be both learned and aroused outside conscious awareness (Öhman & Soares, 1994).

Modularity refers to the extent to which the induction process of the mechanism can function as an independent and information-encapsulated ‘module’, which may be activated in parallel with other psychological processes (Coltheart, 1999). Emotional contagion may be described as highly ‘modular’, because it may be activated independently of other processes, and is not influenced by the information of other modules (e.g., we respond to the expressive features of the music as if they came from a human being expressing emotions through the voice, even if we know, at some cognitive level, that the music is not actually a voice). Conversely, imagery is highly dependent on attentional resources, which are easily distracted by competing stimuli that may interfere with the imagery process (Kosslyn, Thompson, & Ganis, 2006).

Dependence on musical structure refers to the extent to which the induction process depends on the precise structure or style of the music heard. At one extreme, the structure of the music is not important as such: it simply functions as a ‘retrieval cue’. This is the case for evaluative conditioning and episodic memory. At the other extreme, the precise unfolding pattern of the musical structure strongly determines the nature of the induced response. This is the case for musical expectancy. However, all mechanisms engage in one way or another with the musical structure, whether it is a melodic theme, a voice-like timbre, a rhythm, or an unusual chord.

2.4. Implications of the BRECVEM framework One implication of the original framework is that different approaches might be reconciled by observing that they focus on different mechanisms. Hence, the framework may help to resolve disagreements about which emotions music can arouse, how early musical emotions develop, whether listeners are active or passive in the causal process, how much time it takes to arouse an emotion through music, and whether emotions to music are innate or learned responses - it all depends on the mechanism concerned (Juslin et al., 2010).

This line of reasoning has in turn a major implication for empirical research: In order for data to contribute in a cumulative fashion to our knowledge, researchers must try to specify, as far as possible, the mechanism involved in each case. Otherwise, studies will produce results that are inconsistent, or that cannot be given a clear interpretation. While a neglect of mechanisms has been the rule rather than the exception, the problem has been particularly salient in studies that involve psychophysiological indices and brain imaging (for overviews, see Hodges, 2010, and Koelsch, 2012, respectively), which have yielded notoriously inconsistent findings so far. Preliminary results suggest that activation of specific target mechanisms through experimental manipulation could yield more predictable patterns (Juslin et al., in press). A Unified Theoretical Framework 13

As noted earlier (Section 2.3), a major assumption of the BRECVEM framework is that each mechanism has a tendency to arouse some emotions rather than others, but between them, the mechanisms may account for a wide range of emotions to music – both ‘basic’ and ‘complex’. However, one limitation of the BRECVEM framework, as described above, is that it does not seem able to account for emotions that are caused by aesthetic judgments. Cognitive appraisal theory is also at loss to explain this type of emotions, since the emotions do not involve goals, plans, or motives. Hence, a novel mechanism has been added to the framework to account for ‘aesthetic emotions’.

3. Extending the Framework: The Case of Aesthetic Judgments It is commonly assumed that art and emotion are inextricably linked (Hjort & Laver, 1997, p. 3), as implied by the concept of ‘aesthetic emotions’. However, as noted above, it is possible to discern at least two different uses of this term: First, some scholars use the term simply to refer to all emotional reactions to ‘art objects’ – music, theatre, painting, literature – without implying that the emotions themselves are of a peculiar kind. (On this view, all mechanisms outlined above produce ‘aesthetic emotions’, regardless of the underlying causal process.)

However, a view that considers all responses to music to be ‘aesthetic’ excludes nothing, and thus renders the term ‘aesthetic’ largely meaningless. Certainly it does not help researchers to decompose the wide range of responses that might occur during music listening (Section 1.2). A more useful approach is to consider the possibility that certain responses could be somehow different to a sufficient extent to motivate the use of the term ‘aesthetic’ for this sub-set. In the present review, the term is reserved for responses to music that involve some kind of aesthetic judgment. The point is not that the associated emotion, if any, is ‘unique to art’, but rather that the origin of the response – the causal process – is of a particular kind.

What does psychology has to offer when it comes to explaining aesthetic judgments? Though ‘empirical aesthetics’ has been regarded as one of the oldest subdomains in psychology (e.g., Fechner, 1876), psychologists have generally avoided to study aesthetic responses (Jacobsen, 2006). Reber (2008) claims that there is virtually no exchange between empirical psychology and art theory, and several researchers have noted the lack of a comprehensive psychological theory that may explain aesthetic responses (Dahlstedt, 1990; Leder et al., 1994), particularly in the case of music (Juslin, 2009). The field of music and emotion, furthermore, makes little reference to aesthetics. Thus, for example, a recent overview did not even mention the words ‘aesthetic’ or ‘art’ (Lamont & Eerola, 2011).

Psychologists have instead focused on more ‘mundane’ processes, such as ‘music preference’ (Hargreaves & North, 2010). However, it has been argued that, “equating aesthetic emotions with mild positive states [preference] underestimates the emotional power of the arts” (Silvia, 2009, p. 9). That is, emotions cannot be reduced to preferences. Juslin et al. (2010) argue that ‘emotion’, ‘preference’, and ‘aesthetic judgment’ are (partly) independent phenomena, which, however, may influence each other in several ways. The implication is that each phenomenon needs to be investigated in its own right. Quite clearly, aesthetic judgments have received the least attention in psychological research to date. Hence, the following sections aim to provide some preliminary inroads into this complex and difficult topic.

3.1. The defining feature of aesthetic responses A first assumption of the present account is that not all music experiences should be properly regarded as ‘aesthetic’, as theoretically defined here. I will argue that an aesthetic experience A Unified Theoretical Framework 14 occurs to the extent that an aesthetic judgment has been made – which is to say that aesthetic criteria have been brought to bear on the music – or, expressed differently, that ‘an aesthetic attitude’ has been adopted by the listener. Once a piece of music is treated as art, the process of aesthetic judgment may begin.

What, exactly, are such aesthetic judgments like? The nature of aesthetic judgments has been discussed within the field of aesthetics, the branch of philosophy devoted to conceptual and theoretical inquiry into art and aesthetic experience (Levinson, 2003, p. 3). Several hundred years of philosophical work has so far failed to yield a consensus that could guide empirical research. Part of the problem is defining ‘art’, with some scholars suggesting that the concept of art is inherently open, and so resistant to definition. Despite these problems of definition, common conceptions of ‘aesthetic experience’ emphasize its focus on an object’s form (e.g., elements of art, design principles, and physical materials) and content (e.g., semantic content, including what the artist may have intended or what the audience actually perceived).

However, as should be apparent by now, all of the BRECVEM mechanisms discussed earlier engage, in some way or another, with the form and/or content of the music. Hence, this is not unique to an aesthetic appreciation. Levinson (2003) points out that, “aesthetic properties are perceptual properties relevant to the aesthetic value of the object that possesses them” (p. 6); not every perceivable feature of an art object is relevant to its aesthetic status (Gopnik, 2012, p. 133). However, when certain aspects of the music’s form and content are evaluated by the listener with regard to their value as art, it would appear appropriate to speak of an ‘aesthetic experience’. This does not suggest that listeners cannot have intense, emotional, and valuable experiences of music without making an aesthetic judgment. It is only stating that responses that involve aesthetic judgments may be different in some respects. Aesthetic responses, then, are responses that include, but are not necessarily limited to, some type of aesthetic judgment. Also included are the regular aspects of music experience (Gabrielsson, 2011; cf. Section 1.1).

3.2. The aesthetic judgment process: A rough sketch First, it may be useful to situate aesthetic judgments within the present theoretical framework. I assume that the aesthetic judgment mechanism is a much more recent addition to the human behavior repertoire than the seven BRECVEM mechanisms. Whereas these latter mechanisms have a long evolutionary history, current notions of ‘art’ and ‘the artist’ are relatively modern concepts that would not have been recognized by our human ancestors (Dahlstedt, 1990). As compared to the other mechanisms, an aesthetic judgment is assumed to rely more on ‘higher’ cognitive functions, domain-relevant knowledge, and a fluid, individualized process that may change across time and context. That is to say that aesthetic judgments are strongly (although not exclusively) influenced by cultural variables. Although there may be some universals that set constraints on what we initially find beautiful (e.g., symmetry in faces), “the arts are about particularity, and the particularity of art history is inherently unpredictable in terms of broad universals” (Sheridan & Gardner, 2012, p. 281). Whereas the basic operations of the emotion- induction mechanisms discussed earlier are consistent over time, aesthetic and stylistic norms and ideas change over time in society (e.g., Jacobsen, 2006), making it much more difficult to pin down a definitive account of music listeners’ aesthetic judgments for all times and places. However, though it may be difficult in practice to characterize aesthetic judgments, it may not be impossible in principle. Thus, in the following, I will offer some preliminaries with respect to what a psychological theory of aesthetic judgment in music experience might look like.7

7 Some art theorists, such as Sharpe (2004), argues that, ‘the work, not the experience, is the proper focus of attention’, but since the artwork cannot be understood except through experience, the notion of the art object ‘itself’ lacks plausibility. A Unified Theoretical Framework 15

3.2.1. The aesthetic attitude Figure 1 presents a schematic outline of the aesthetic judgment process with regard to music. Consistent with Leder et al.’s (1994) theoretical model of aesthetic appreciation of visual art (particularly modern art), I assume that the aesthetic judgment process begins with an initial classification of the music as ‘art’, which will lead the listener to adopt an aesthetic attitude. This means that the listener’s attention is focused on the music, and that aesthetic criteria are brought to bear on the music.

(Insert Figure 1 about here)

Because, as Leder et al. (1994) suggest, “the borders between what was considered an artwork once and what is called art today are constantly changing” (p. 490), it is not self-evident when an aesthetic attitude is ‘called for’. The impetus can come from various sources. An ‘aesthetic framing’ (e.g., that the music is performed in a prestigious concert hall) could serve as a ‘cue’ that the music ‘deserves’ an aesthetic attitude or the music itself could have some perceptually salient feature (e.g., novelty, extraordinary beauty), which draws the listener’s attention to its ‘aesthetic potential’. Or the music could arouse an intense emotion, which, for some listeners at least, is suggestive of aesthetic value (Juslin & Isaksson, 2013).

In general, it can be expected that some pieces of music will ‘invite’ an aesthetic attitude to a greater extent than other pieces (because of certain formal features), but there is no guarantee that an aesthetic judgment will be initiated. This will depend on several factors, including the listener’s attention and knowledge, the characteristics of the situation etc.8 In fact, it might be the case that when two listeners hear the same piece of music, only one of these listeners will have an aesthetic experience, since a different set of features may trigger an aesthetic attitude in different listeners (Section 3.2.3).

Does the listener have to be consciously aware that an ‘aesthetic judgment’ is being made, in order for it to ‘count’ as a true instance of such a judgment? Although I believe that aesthetic judgments are often made explicitly, in particular amongst ‘music experts’, I propose that the judgment might also be implicit under certain circumstances. The defining feature is whether the listener’s aesthetic criteria have been brought to bear on the music, not whether the initial classification as art, or the subsequent evaluation, was made on a conscious level. It should be noted that implicit judgments can be inferred from data in carefully designed experiments that manipulate relevant perceptual properties in a systematic manner (Cooksey, 1996).

3.2.2. Inputs to the judgment process Once an ‘aesthetic attitude’ has been adopted, perceptual and cognitive analyses of the music will proceed, which provide inputs to the aesthetic judgment process (see Figure 1). This may be construed as a continuously on-going process. Similar to emotional responses (Juslin et al., 2008), aesthetic processing is influenced by a number of factors in the artwork, the perceiver, and the situation (Jacobsen, 2006). Information related to these factors is ‘channelled’ through the perception, cognition, and emotion of the listener (see also Shimamura, 2012, pp. 23-24):

8 This suggests the testable hypothesis that the very same piece of music can lead to vastly different reactions in a listener (positive vs. negative), depending on whether it was judged in terms of ‘art’ or not. We may like a pop song as such, but once we apply criteria to evaluate the song’s aesthetic merits, our liking for it may well wane. A Unified Theoretical Framework 16

- Perceptual inputs are sensory impressions, based on ‘low-level features’ of the music, which are regarded as mandatory in the process (Jacobsen, 2006; Shimamura, 2012).9 These occur early, both in the judgment process and in ontogenesis, and are assumed to be mainly biologically based and fairly uniform among listeners, since they reflect basic principles of perceptual organization (Bregman, 1990). Examples of perceptual features of relevance may be symmetry, order, proportion, and figure-ground contrast (e.g., Arnheim, 1974; Ramachandran & Hirstein, 1999; Reber, 2008).

- Cognitive inputs involve ‘higher-order’ concepts that are more knowledge-based than the perceptual-sensory inputs. The knowledge is organized in terms of ‘schemata’, or socially determined memory representations, which store domain-specific knowledge about music (including information concerning the musical style, the composer or the performer, and music history). Both the nature and extent of cognitive input will vary widely depending on the level of expertise of the listener. However, even provision of information at the moment of reception may strongly influence aesthetic judgments by listeners (Crozier & Chapman, 1981). It is commonly assumed that explicit education in the arts may encourage a more advanced appreciation of art by providing depth and novel perspectives beyond one’s own immediate reaction (Sheridan & Gardner, 2012).

- Emotional inputs may or may not occur in any individual instance. If the music evokes an emotion, through one of the BRECVEM mechanisms outlined earlier, this response will inform the aesthetic judgment process to the extent that the listener includes in her set of aesthetic criteria at least one criterion that could be informed by the reaction (see examples below). However, emotional input is not needed for an aesthetic judgment to take place or to produce an outcome (discussed in Section 3.3).

3.2.3. The role of subjective criteria Perceptual, cognitive, and emotional inputs provide information about the musical event, but whether this information will have a decisive effect on the aesthetic judgment depends on the listener’s subjective criteria for aesthetic value, which represents the next stage (see Figure 1). What are the dimensions or criteria that may underlie listeners’ aesthetic judgments of music? It should be noted from the outset that there is a paucity of research on these matters, and that the following review will offer only some suggestions. As a point of departure, we could look at different definitions or conceptions of art put forward from ancient Greece to modern times. The following views seem to have been particularly influential in the philosophy of aesthetics:

Ɣ art as representation of nature (‘mimesis’; Plato, 1923) Ɣ art as possessing certain properties such as beauty (Kant, 1790), complexity (Berlyne, 1971), or the sublime (Lyotard, 1994) Ɣ art as expression (e.g., ‘clarifying an emotion’; Collingwood, 1938) Ɣ art as originality (e.g., ‘modernism’; Greenberg, 1971; Kraus, 1986) Ɣ art as good taste (e.g., ‘the absence of extremes’; Fechner, 1876) Ɣ art as artistic skill (e.g., ‘fashioning objects that are admired’; Sparshott, 1978) Ɣ art as reception (e.g., ‘feelings in the audience’; Tolstoy, 1960) Ɣ art as conveying a message (‘implicitly advancing some proposition’; Levinson, 2001) Ɣ art as intention (‘art is something produced with the intention of giving the capacity to satisfy aesthetic interest’; Beardsley, 1982)

9 Although one could envisage making aesthetic judgments of merely imagined music, the process of auditory imagery would still involve many of the same brain regions for sensory processing as would actual music (e.g., Zatorre et al., 1996). A Unified Theoretical Framework 17

Ɣ art as defined by institutions (‘art is what art schools, museums, and artists define as arts’; Dickie, 1974)

As seen, different art theorists have advocated distinct criteria for aesthetic value. However, to explain listeners’ responses to music, a psychological theory must take their criteria as a point of departure rather than the criteria embraced by art theorists, even though the former are most certainly influenced by the latter to some extent:

“What counts as the salient features and effects of works of art are in fact determined by the judgments of experts and elites. A scientist, that is, may need to pay attention to what art experts say because the nature of our entire artistic experience trickles down from them.” (Gopnik, 2012, p. 130)

Which conception of ‘art’, then, best applies to the ‘average’ music listener? In a recent survey study, Juslin and Isaksson (2013) required listeners, both musicians and non-musicians, to rate the relative importance of various criteria (primarily based on the above conceptions) for their consideration of music as ‘art’. Figure 2 presents some results in terms of mean ratings across listeners, and some examples of ratings by individual listeners. Three things are apparent from these results: First, there appears to be no single criterion that accounts for listeners’ aesthetic judgments of music. Second, certain criteria (e.g., Expression, Emotional arousal, Originality, Skill, Message, Beauty) were judged as more important than others, overall. Finally, different listeners tended to focus on different criteria.10 It should be acknowledged that explicit ratings of the relative importance of criteria in general is not really the same as proven causal effects of these criteria on actual judgments of music, but it seems plausible that the two are related – perhaps particularly for ‘engaged’ music listeners (Lamont & Greasley, 2009), who are better able to articulate their subjective criteria.

(Insert Figure 2 a/b about here)

It may be tentatively concluded that aesthetic judgments by music listeners generally embrace an art definition in terms of multiple attributes or clusters, rather than a single attribute that is ‘a necessary and sufficient condition’ (for a discussion, see Stecker, 2003); that some criteria for aesthetic value are generally regarded as much more important than others; and that there are individual differences in how these attributes are weighted. Thus, when two listeners hear the same piece of music, their aesthetic judgments could produce highly different outcomes – hence the common expression that there is “no accounting for taste” (Shimamura, 2012, p. 3).

Given such individual differences with respect to aesthetic criteria, it appears pertinent to ask whether one could ever hope to predict a listener’s aesthetic judgments of pieces of music. In fact, by applying Brunswik’s (1956) lens model paradigm (e.g., Juslin & Bänziger, 2009), we might be able to map individual listeners’ aesthetic criteria, as well as their relative weighting functions, such that we can predict their judgments of new pieces of music with a fair level of success. The mapping could take the form of listening tests, where music examples are varied in a systematic way with regard to different dimensions, and listeners are required to rate each example with respect to both overall aesthetic quality and individual criteria. Their judgments

10 The relative role of different aesthetic criteria here can be compared to the findings in the study by Istok et al. (2009), which relied on (high-speed) recall memory of aesthetic aspects of music. The results revealed ‘beauty’ as a primary factor, but one can also discern responses related to criteria such as skill, emotion arousal, novelty and wittiness. The problem, it seems, was that listeners tended to confuse criteria and reaction (cause vs. effect) in their responses. Furthermore, the reliance on recall memory meant that some criteria of potential importance may never have crossed the participants’ minds. A Unified Theoretical Framework 18 can then be modeled using Multiple Regression Analysis (Cooksey, 1996) – similarly to what has been done with judgments of emotional expression in music performances (Juslin, 2000). The feasibility of an ‘idiographic’ approach (Brunswik, 1956) was shown by Jacobsen (2004), in a study where participants made aesthetic judgments of visual patterns that were varied in a systematic manner. Results revealed that individual models provided much better precision in predicting perceivers’ judgments than did aggregated (averaged) models. Höfel and Jacobsen (2003) explored the temporal stability of such judgments and found that the stability tended to decrease with larger time-intervals between the measurements, which ranged from three days to 14 months. Still, the judgments did show stability in the short-term.

In summary, a basic premise of this approach is that aesthetic judgments of music are multi- dimensional: They involve individual sets of subjective criteria for aesthetic value, as well as a relative weighting of these criteria. The criteria are assumed to have moderate stability, such that listeners’ judgments can be predicted to some extent, at least in a short-term perspective. The listener’s aesthetic criteria function as ‘filters’, which determine which of the perceptual, cognitive, and emotional inputs will actually be allowed to influence the aesthetic judgment. To take a simple example, a particular piece of music may ‘afford’ a listener the possibility of aesthetic appreciation in terms objective features (e.g., order, symmetry) that have been found to correlate with perceived beauty (Reber, 2008). But whether this impression is rated highly depends on the individual listener’s criteria for aesthetic value: some listeners may regard the perception of beauty as essential to the value of an art object, whereas others may regard it as peripheral at best (Figure 2b). It is not that some listeners fail to detect the perceptual features, it is simply that the features are ‘discounted’ in the aesthetic judgment.

3.2.4. A preliminary set of criteria Further research is urgently needed to map the criteria that are most influential on judgments. I propose that listeners’ aesthetic criteria for music will tend to include – but not be limited to – broad dimensions related to the following concepts (most of which received high ratings in the survey study by Juslin & Isaksson, 2013):

Beauty: Older theories posited that ‘all art aims at beauty’, but “there is a marked tendency to abandon the old concepts of beauty as the sole criterion of good art” (Leder et al., 1994, p. 490). Lorand (2007) admits that beauty has receded, or even disappeared, from contemporary aesthetic theory, but claims that it is “as relevant now as it was at the time of Plato and Kant, simply because it has never ceased to be of interest in everyday life”. This may be especially true for music. Indeed, Juslin and Isaksson (2013) observed in their survey study that listeners regarded ‘beauty’ as one of the most important criteria for their choice of music, and this was particularly true for musicians. ‘Beauty’ was also a primary factor in a survey study of verbal associations to music (Istok et al., 2009). Gabrielsson’s study (2011) of peak experiences with music included several references to percepts of musical beauty. But what, exactly, is beauty, and how can it be explained? Most modern analyses of beauty adopt an interactionist view on beauty, compatible with the psychophysical principle outlined earlier, which states that beauty emerges from a relationship between the perceiver and the object, rather than from ‘objective’ features of the object or ‘subjective’ features of the perceiver.11 Perception of beauty has been linked to numerous characteristics of an object such as symmetry, balance, proportion, clarity, order, figure-ground contrast, and the absence of extremes (e.g., Eysenck, 1981; Reber, 2008). The most influential attempt to explain the effects of these features on perceived beauty is the perceptual fluency hypothesis (e.g., Reber, Schwarz, & Wienkielman, 2004), which states that

11 This account rejects the claims by some theorists (e.g., Bicknell, 2009, pp. 141-142) that aesthetic value could be ‘objectively defined’. A Unified Theoretical Framework 19 the more fluently the perceiver can process an object, the more positive is the response. All of the features mentioned above have in common that they enhance the perceptual fluency. Their effects reflect innate principles for stimulus organization, but also cultural influences in terms of repetition and implicit learning. However, perceptual fluency might not be the whole story. Indeed, another notion is that listeners “are responding to the beauty in the music’s expressive character” (Bicknell , 2009, p. 108), which suggests that the criterion of ‘beauty’ is sometimes conflated with another criterion – ‘expression’. “The authentic expression of essentic form is experienced as ‘sincerity’” (Clynes, 1977, p. 60), and such purity of expression “also turns out to be beautiful” (p. 54). Still, even this aspect may be related to perceptual fluency: The closer to the ‘ideal’ form an expression is, the more fluently the emotion recognition will proceed.

Expression: A recurrent idea in the literature is that art is expressive (e.g., Levinson, 2003). In other words, it can ‘express’, ‘represent’, ‘signify’ or ‘refer to’ something else beyond itself; and that ‘something’ – the ‘content’ or ‘meaning’ – is regarded by many as a crucial aspect of aesthetic value. The nature of the content has been the focus of much debate – even if we limit ourselves only to expression in music. Indeed, throughout history, music has been regarded as expressive of motion, tension, human character, identity, religious faith, and social conditions; though the most common hypothesis is that listeners perceive music as expressive of emotions (Gabrielsson & Juslin, 2003). Research confirms this notion. For instance, in a study by Juslin and Laukka (2004), 141 participants were asked what, if anything, music expresses. They were required to tick any item that seemed reasonable from a list of alternatives, and could also add their own alternatives. (The list was based on a thorough survey of the literature on expression in music.) The results indicated that, although there were individual differences with regard to most alternatives, there was complete agreement (100%) among listeners that music expresses emotions. Further, 76% of the listeners reported that they ‘often’ perceive music as expressive of emotions (for similar data in a study of performers, see Lindström et al., 2003). Expression in music reflects a large number of musical features, shaped by both composer and performer, which have been extensively mapped in previous research (Juslin, 2009, Table 1). Interviews suggest that audiences prefer the most expressive musicians (Boyd & George-Warren, 1992), and expression is one of the main criteria by which music performances are formally assessed (McPherson & Schubert, 2004). In the survey study by Juslin and Isaksson (2013), expression was rated as the most important criterion in judging the aesthetic value of music amongst both musicians and non-musicians (see Figure 2).

Novelty: Artists often strive to create something new in their work. Indeed, most definitions of creativity involve some reference to ‘novelty’ or ‘originality’, frequently combined with the requirement that it should be ‘valuable’ or ‘useful’ (Simonton & Damian, 2012). While some degree of novelty is probably expected in most accounts of art, the desire to create something new – even at the expense of other criteria represented here – arguably reached a peak during the 1950s, with the modernistic avantgarde concept and its obsession with anything novel and original (Dahlstedt, 1990). Berlyne (1960) argued that a novel artwork is interesting because it falls between the perceiver’s existing categories, creating ‘uncertainty’ and ‘conflict’ between competing categorizations and interpretations. He distinguished between ‘short-term novelty’ (contrasts with recent experience), ‘long-term novelty’ (novelty within a longer timespan) and ‘complete novelty’ (e.g., encountering a new object). All three types may readily be observed in music: novelty within a piece; novelty from a music-historical perspective; or entirely novel styles or instruments. Survey data suggest that originality retains an important role in aesthetic judgments among today’s music listeners (Juslin & Isaksson, 2013), but it does not seem to be the most important criterion (see Figure 2). And, as observed by Clarke (2012a), not all music performance takes creativity as its goal. It is more plausible to describe music performance in A Unified Theoretical Framework 20 terms of a continuum, from ‘interpretation’ to ‘innovation’. The same applies to composition, but here the demands for novelty are arguably far greater. One aspect of novelty that has been operationalized and tested is ‘melodic originality’ (Simonton, 2001). Melodies that score high in melodic originality sound more novel, unpredictable, and interesting. However, because of ‘habituation’, “each generation of composers feels compelled to produce work that overcomes habituation to a given level of originality” (p. 214). Indeed, even individual composers tend to steadily increase their use of originality, as their careers progress. However, as far as aesthetic judgment in music experience is concerned, it is the listener’s subjective sense of novelty that is decisive for the outcome (cf. Heyduk, 1975), which in turn is influenced to a great extent by knowledge and experience (Hargreaves & North, 2010).

Emotion: It has often been suggested that art should arouse emotions in the perceiver (e.g., Shimamura, 2012, p. 9; Tolstoy, 1960), particularly in music (the renowned violinist, Yehudi Menuhin, settled for no less a goal than “to move an audience”; Menuhin, 1996, p. 413). This view is implicit in the Ancient idea that art may bring ‘catharsis’, “the purification of the soul through affective experience” (Cook & Dibben, 2010, p. 47), and also in the assertion that art involves ‘aesthetic pleasure’ or ‘enjoyment for its own sake’ (Huron, 2009). Robinson (2005) claimed that a good criterion for successful Romantic expression is the arousal of “appropriate emotions” in the listener (p. 380), and emotional arousal was an important aspect of Berlyne’s (1971) New Empirical Aesthetics, which involved the explicit assumption that an art object is designed to ‘arouse’ the perceiver somehow. However, it needs to be acknowledged that some listeners regard the emotions aroused by a piece of music (if any) as irrelevant to the aesthetic value of the music. And Hanslick (1986) famously denied that music’s ability to evoke strong emotions in listeners (which, incidentally, he did not deny!) was indicative of its artistic value. It seems safe to conclude, therefore, that the relationship between aesthetic judgment of music and emotion is a complex and controversial one, and this relationship is, of course, at the heart of the present discussion (see Sections 3.3 and 4.1). Survey data indicate that many ‘ordinary’ listeners do consider the music’s ability to arouse emotion to be a very important criterion for its aesthetic value (Figure 2), though the precise nature of these emotions has been a matter of some debate (see Section 1.2).

Skill: Humans cultivate, recognize, and admire technical artistic skills (Dutton, 2009), and this form of appreciation applies to musical skills as much as to any other art form. Thus, for instance, several respondents in Gabrielsson’s (2010) SEM study referred to “the qualities of the performance, such as the performer’s skill” in explaining their response (p. 568). Sloboda (1999) discussed the case of virtuosity as a source of emotion. For instance, we may ‘marvel’ with amazement or awe at the exceptional speed with which a pianist is able to play a difficult passage, or we may admire a brilliant composer’s use of the medium. The emphasis on skill in artistic endeavours is so ingrained that many people tend to discount works of art that weren’t difficult to create (i.e., that did not require special skills): ‘I could have done that myself’. In a musical context, the notion of ‘skill’ has broadened from its original focus on composing and physically performing music to include also production skills and ‘recycled’ music played by ‘machines’. However, any piece of music considered ‘a great work of art’ is likely to ‘invite’ an attribution of great skill to its creator. Musical skills may be discernible in separation from other criteria (e.g., technical abilities), or may be more or less ‘inferred’ from other criteria in hearing the music (e.g., the ability to perform expressively or to write beautiful melodies).

Message: Art usually implies the conveying or communicating of an idea (Levinson, 2001). For instance, the artist might intend to convey a certain message or not; or the perceiver might perceive a certain message or not. Some scholars would suggest that the value of a work of art A Unified Theoretical Framework 21 depends on whether the intended message is detected by a perceiver (Reber, 2008), but others would probably dispute this. In any case, it is important to note that an art work’s potential to convey referential information to the perceiver is separate from the question of whether the art work was made with a particular intention. Hence, a listener could judge the aesthetic value of a piece of music in terms of it having a certain ‘message’, regardless of whether the music was in fact composed with that message in mind. Whenever we search for the underlying meaning of the music (for an example, see Cook & Dibben, 2010, p. 49), we are effectively applying a conceptual approach to art (Shimamura, 2012). A listener’s subjective sense of understanding of the musical ‘message’ may lead to a feeling of ‘cognitive mastering’ (Leder et al., 2004) or an appreciation of the ‘wittiness’ of the work (Levenson, 2003). The notion of communication has been fairly controversial amongst music theorists (for a discussion, see Juslin, 2005b), but evidence from survey studies strongly suggest that many performers and listeners do conceive of music as a form of communication (Juslin & Laukka, 2004; Lindström et al., 2003), though the contents of this communication (i.e., the message) remains a matter of debate. The point is that some listeners will judge the aesthetic value of music in terms whether they can discern a clear ‘idea’, ‘message’, or ‘concept’ in the piece. In many cases, this process will be aided by knowledge about the piece or artist involved. If the music in question contains lyrics, this may further contribute to the assessment of the message criterion (for examples of the role of lyrics in strong experiences with music, see Gabrielsson, 2011, pp. 432-435).

Style: Style refers to the distinct manner in which an act is performed or an artifact made, or ought to be performed and made (Gombrich, 1998). Artistic objects and performances satisfy ‘rules of composition’, which place them in a recognizable style. An artist often tries to create something novel (see the Novelty criterion). In doing so, the artist will implicitly or explicitly establish a new set of ‘rules’. If other artists see these ‘rules’ as valid for themselves, they will also apply these in their works. The resulting works of art then take on that ‘style’. The notion of style could also be applied at the individual level, however. Leder et al. (1994) go so far as to suggest that, “nowadays, an artist’s success is mainly due to a recognizable and distinctive artistic style” (p. 491). Styles are, however, necessarily ‘fuzzy’ categories. The classification is slightly arbitrary, style categories often overlap, and categories may come and go over time (Hargreaves & North, 2010). Even so, there is clearly some substance to the notion of artistic style in music. Composing music within the boundaries of a specific style creates coherence; it places the music within a specific historical and social context; and it gives listeners a frame of reference for understanding the music. Musical understanding may contribute to a sense of ‘mastery’ in the listener, and may explain the preference-for-prototypes tendency observed in previous studies. More specifically, it has been reported that listeners will appreciate the most those musical themes that are ‘prototypical’ of a particular style (Martindale & Moore, 1989). Indeed, Simonton (2001) suggested that listeners with lots of experience with classical music will eventually learn to identify even the distinctive styles of specific composers, presumably by relying on minor encoding habits (e.g., distinctive ways of constructing thematic material; Paisley, 1964). Needless to say, style appreciation depends heavily on knowledge, which may be implicit or explicit in nature. Furthermore, Leder et al. (1994) argue that style may be one aspect where preferences could affect the aesthetic judgment. Specifically, it is suggested that the explicit classification of a work of art as belonging to a non-preferred style may terminate further aesthetic processing. Hargreaves and North (2010) note that the idea that some styles can be considered as somehow more or less ‘serious’ than others is “highly dubious” (p. 535), but it remains plausible that the aesthetic judgments of some listeners are influenced not only by the perceived degree of adherence to a given style, but also by whether the style itself is a preferred one or not.

A Unified Theoretical Framework 22

The sublime: The meaning of the term ‘sublime’ is complex and has also changed over time (cf. Bicknell, 2009). It has sometimes been used more or less interchangeably with beauty, but certain authors have suggested that the sublime and the beautiful are distinct qualities, even if they often occur together (Burke, 1958). The sublime refers to a certain aesthetic quality that evokes ‘awe’ in the perceiver, a sense of being overwhelmed or overpowered by an object. It is aroused by “a particularly magnificent, large, or powerful object or event”, which might be ‘natural’ (e.g., mountains, thunderstorms, the raging sea) or made by humans (e.g., buildings, works of art) (Bicknell, 2009, p. 9), usually linked with ‘feelings of insignificance before this great work of art’. Many authors link the quality of the sublime to fear (Huron, 2006), as long as the terror is ‘not too proximate’. Thus, Burke (1958) considers it an overwhelming emotion evoked by a glimpse of infinite perfection of the divine, ‘a sort of delightful horror’. Haidt and Seder (2009) define awe as “reverential or respectful fear” in response to “vastness” – usually something physically vast, but also things with “vast power, genius or complexity”. Not every listener appreciates being overwhelmed by music, however. Hence, ‘the sublime’ is not rated highly as a criterion by all music listeners (Figure 2). Some examples of the sublime in music experiences are given in Gabrielsson (2011). The sublime might be linked to musical features such as complexity, loudness, and beauty. It appears generally agreed that experiences of awe to sublime artworks are rare (cf. Huron, 2006; Juslin, 2011; Koneþni, 2005), but such unusual experiences may still help to motivate people to seek out new aesthetic experiences.

An aesthetic judgment might be initiated by a salient percept in any of the above dimensions, but once the evaluation of one criterion has started, other criteria might be evaluated as well. It is not assumed that the aesthetic criteria are ‘orthogonal’ to each other, but rather that they (like most variables in the ‘real’ world; Brunswik, 1956) are more or less correlated with one another – positively or negatively. For example, conformity to style rules might be positively correlated with beauty – as ‘prototypicality’ could influence perceptual fluency – but may be negatively correlated with novelty. The obvious implication is that a work of art cannot be all things to all people all the time. However, because music is considered to be the temporal art par excellence, music may be better able to alternate between satisfying different demands at different moments in time, than, say, a visual work of art. In addition, all of the criteria above are likely to involve some unique variance that can be captured in music-judgment studies.

The kind of criteria that a listener applies to a piece of music might depend on expertise, such as music education. Cupchik and Laszlo (1992) proposed that ‘naïve’ perceivers adopt a more ‘direct’ mode of reception, whereas ‘experts’ adopt a more ‘cognitive’ reception. I take this to mean that non-experts rely to a greater degree on sensory and perceptually-based criteria (e.g., ‘beauty’), whereas experts rely to a greater degree on knowledge-based criteria (e.g., ‘style’ or ‘message’). Knowledge enables the perceiver to form expectations. This helps to interpret the artwork and direct attention to its salient features (Gombrich, 1960). Aesthetic sensitivity may require some information about the art-historical context (Bullot & Reber, 2013).

Based on studies of human judgment that involve multiple cues (for a review, see Brehmer & Brehmer, 1988), we might expect judges to generally use a relatively small number of criteria (due to limited working memory capacity). Although the precise number of criteria used may vary among individuals (Figure 2), a tentative estimate is that it ranges from 1 to 10, typically hovering around a handful. Given the knowledge-based nature of many criteria, one may also hypothesize that ‘experts’ tend to use a larger number of criteria, and, in particular, are better able to articulate their criteria, than are ‘novices’. They might also be more consistent in their use of the criteria (Brehmer & Brehmer, 1988).

A Unified Theoretical Framework 23

The present approach assumes that aesthetic judgments of music are mostly determined by the outcome of evaluations of specific criteria. However, as McPherson and Schubert (2004) have noted, any music evaluation will be affected by a number of additional factors that impact on the reliability or consistency of the evaluation (p. 62). For instance, various contextual factors such as visual impressions, audience support, order effects, and stereotyping can all influence aesthetic judgments of music. Such extraneous variables may affect which criteria become the focus of attention, among other things.

An interesting question is whether music involves a unique set of criteria, as compared to the other arts. The present framework presumes that there is a great deal of ‘overlap’ in aesthetic judgments across various media, even though the relative weighting of the criteria may differ (e.g., the ‘mimetic’ approach to art is arguably less common in music, than in the visual arts). However, some of the variables studied in the New Empirical Aesthetics (Berlyne, 1971) are not regarded as part of an aesthetic judgment in the present account. For instance, familiarity and the so-called ‘mere exposure effect’ can be accounted for by the Evaluative Conditioning mechanism discussed earlier (Zajonc, 2001), and do not really address music ‘as art’. Tempo (or speed) is similarly not regarded as a proper basis for aesthetic evaluation, although tempo is of course indirectly implicated in the ‘Expression’ dimension, as involved in the expression of emotion in music (Juslin & Lindström, 2010). Nonetheless, an object’s speed or familiarity are not by themselves sufficient grounds for inviting an aesthetic judgment, as defined here. It remains an open question whether earlier studies in the field have insufficiently distinguished between ‘aesthetic judgment’ and (mere) ‘preference’ (Table 1) when exploring the effects of various stimulus variables (see Section 3.2.5).

3.2.5. From aesthetic judgment to emotion What is the relationship then between aesthetic judgments and emotions? The present account assumes that aesthetic judgments will not necessarily arouse an emotion, but that they will do so when (at least) one of the evaluative criteria reaches a certain level. (This is consistent with

Fechner’s (1876) notion of ‘aesthetic threshold’.) Thus, for example, a piece of music judged as merely ‘pleasing’ to the ears may simply be registered as that, without evoking an emotion. However, a piece judged as extraordinarily beautiful may indeed evoke an emotional reaction. Research by Höfel (2003) suggests that it is feasible to operationalize the concept of ‘aesthetic threshold’ for individual perceivers of works of art.

Figure 1 illustrates, in a highly simplified and schematic fashion, different ‘overall’ outcomes of the aesthetic judgment process, as it relates to ‘preference’ and ‘emotion’. If the judgment process indicates that, on balance, the music is ‘good’, this will result in liking (preference). Conversely, if the process indicates that the music is not so good, it will result in dislike. For 12 both of these cases, no emotion is necessarily aroused , unless one of the other mechanisms has been activated during the listening process. If, however, the result of the judgment is that the music is judged as extraordinarily good (or bad), overall or on at least one of the criteria, an emotion will be aroused in addition to the preference (with the valence determined by the outcome of the aesthetic judgment). Assessments with respect to particular criteria occur in a continuous fashion, but judgment outcomes will occur as ‘read-out’ episodes at certain points in time (e.g., ‘cued’ by particularly significant moments in the music, such as the ending of a piece or an improvisation, or the moment when a decision has to be made about buying a CD or leaving a concert). These judgment outcomes may be subject to updating as new ‘evidence’

12 Because emotion is not mandatory in this judgment process, the present approach differs from previous views that have defined aesthetic experiences as ‘intense’ (Koneþni, 1982; Price, 1986). According to the present view, an aesthetic experience can also be relatively calm and contemplative, if the judgment outcome is merely liking. A Unified Theoretical Framework 24 comes along during continued listening to the music. Research in event perception may offer clues to how the segmentation of events culminates with judgment outcomes at crucial event boundaries (Tversky & Zacks, 2013).

In the absence of any emotions aroused by the BRECVEM mechanisms, preference, emotion and aesthetic judgment will be consistent with each other. Consider, however, the possibility that a piece that is judged as being of low aesthetic value by the listener does arouse a positive emotion, through a psychological mechanism such as Evaluative conditioning. In such a case, emotion and aesthetic judgment will end up contradictory, with ambivalent feelings as a result (discussed further in Section 4.1).

I submit that, for ‘engaged’ music listeners (Lamont & Greasley, 2009), including musicians, music preferences will largely reflect aesthetic judgments of music. For less ‘engaged’ music listeners, however, it seems plausible that music preferences will commonly be based on other factors (including positive emotions evoked through the BRECVEM mechanisms), effectively by-passing any aesthetic judgment, as defined here (e.g., ‘I prefer this music because it makes me feel good’). Music preferences may be based on any number of reasons, of which emotion and aesthetic judgment are two distinct possibilities. But music preference requires neither of these processes. In this approach, then, aesthetic judgments always lead to preference (like or dislike), whereas preferences are not always based on aesthetic judgments.

The above account presumes that preference, emotion, and aesthetic judgment are independent to some degree (for a discussion, see Juslin et al., 2010, pp. 634-637). Empirical evidence of a dissociation between (aesthetic) quality judgments of music and preferences were reported by Hargreaves, Messerschmidt, and Rubert (1982). In their investigation, both musically trained and untrained listeners gave higher ratings of quality to classical pieces than to popular pieces, but this pattern was not found for ratings of liking (i.e., preference). Evidence of a dissociation between aesthetic judgments and arousal of emotion was obtained in a questionnaire study by Thompson (2006), which explored listeners’ responses during a concert. Results indicated that listeners were able to differentiate between the perceived quality of the performance (aesthetic judgment) and their own emotional reaction. (Their enjoyment of the musical event was better predicted by their emotional engagement, than by the perceived quality of the performance.)

Thus, the present framework postulates that an aesthetic judgment can lead to both liking (i.e., preference) and emotion. Liking (or disliking) is a mandatory outcome, whereas emotion is a possible additional outcome. Previous studies of aesthetic judgments in music have primarily focused on liking (e.g., Hargreaves & North, 2010), but here the focus is on the emotions that may occur as a result of an aesthetic judgment. One can assume that partly different emotions are aroused by aesthetic judgments based on different criteria. A few examples may suffice to illustrate this point. Novelty in the music may surprise the listener (Berlyne 1971), which may be followed by interest if the music is comprehensible (Silvia, 2006). ‘Cognitive mastering’ of a complex musical structure leads to satisfaction (Leder et al., 1994). Beautiful objects ‘invite’ aesthetic contemplation and reward it with pleasure (Bicknell, 2009, p. 40). Skill in a musical performance could arouse admiration in the listener (Hodges & Sebald, 2011, p. 199; see also Huron, 2006), or, if the performer’s skill is exceptional, it might arouse wonder. If music with both great beauty and physical grandeur is heard in a ‘colossal’ performance setting, such as a medieval cathedral, it could arouse awe (Koneþni, 2005, p. 37). More generally, we will enjoy pieces of music that are highly consistent with our aesthetic values (Silvia, 2012).

A Unified Theoretical Framework 25

Not all aesthetic judgments will evoke positive emotions, however. Scruton (1997) argues that an artwork can fail in two ways: it may fail to interest us (which I would regard as a failure to invite an aesthetic attitude) or it may invite an interest of which we disapprove (which I would interpret as a completed aesthetic judgment process resulting in a primarily negative outcome: see Figure 1).

3.3. Interactions with BRECVEM mechanisms The present account assumes that emotions may (a) simply co-occur with aesthetic judgments (being activated independently through one of the BRECVEM mechanisms) or (b) follow as a direct result of an aesthetic judgment. The former emotions have been referred to as ‘everyday emotions’, the latter as ‘aesthetic emotions’ (see Figure 3). The context of music listening will often help to determine which of these types of emotions will occur (see Section 3.1). It needs to be reiterated that the distinction is based on the underlying causal process, rather than on the resulting emotions per se. For example, interest can be an ‘everyday emotion’ or an ‘aesthetic emotion’, simply depending on how it was aroused. Because several processes involved in the emotion-induction are implicit in nature, it’s possible that listeners occasionally ‘misattribute’ an emotion aroused by one of the BRECVEM mechanisms (i.e., an ‘everyday emotion’) to an aesthetic judgment on a conscious level that happened to occur at the same time, thus creating the (false) impression of an aesthetically-motivated emotion.

(Insert Figure 3 about here)

If this account is largely correct (i.e., that an aesthetic judgment might occur independently of other emotion-induction mechanisms), it seems pertinent to ask whether the different types of mechanism might influence each other. Spinoza (1677/1989) regarded emotions as ‘states that make the mind inclined to think one thing rather than another’. Thus, emotions aroused by the BRECVEM mechanisms outlined earlier may influence or ‘bias’ aesthetic judgments by their highlighting of some features of the music rather than others. This would be one example of a more general principle of ‘affect-as-information’ – the notion that people often use feelings to guide beliefs and judgments (Forgas, 1995). Therefore an important principle of the expanded framework is that the activation of a BRECVEM mechanism, and its associated emotion, may ‘inform’ an aesthetic judgment indirectly related to the mechanism (see Figure 3).

Activation of the Contagion mechanism, which reflects the mainly iconically-coded emotional expression of the music, can inform aesthetic judgments in terms of the ‘expression’ criterion, highlighting to the listener that the music is expressive. Activation of the Musical expectancy mechanism, which reflects deviations from stylistical expectations, may inform the listener’s judgment of the music on the ‘novelty’ criterion (and also, perhaps, the ‘style’ criterion). The brain stem reflex mechanism, which responds to ‘extreme’ acoustic features such as loudness, may inform the listener about the power and vastness of the music, as linked to the ‘sublime’ criterion. And perhaps the Visual imagery mechanism may inform the listener’s judgment of the underlying ‘message’ of the music, by means of analogy and metaphor.

Moreover, the activation of any of the seven BRECVEM mechanisms, along with the ensuing emotion, might inform the ‘emotion’ dimension; that is, indicating an aesthetic value in terms of being able to arouse emotions, at least if emotion is part of the listener’s subjective criteria. Another possible effect of emotion on aesthetic processing is that positive emotions can lead a listener to process the music more ‘holistically’, whereas negative emotions will him or her to process it more ‘analytically’ (cf. Leder et al., 1994). This could have an effect on the number or type of criteria that are brought to bear on a piece of music. A Unified Theoretical Framework 26

Conversely, if the listener adopts an ‘aesthetic attitude’ to a piece of music, his or her criteria for aesthetic value and their relative weighting may have a bearing on whether the music will arouse emotions through any of the other mechanisms, by influencing the listener’s attention during the actual listening to the music. (For example, the aesthetic framing of a concert with classical music might invite an aesthetic attitude that makes the listener focus on the criterion ‘expression’, which then will increase the chance that the Contagion mechanism is activated.)

Finally, one can envisage ‘everyday emotions’ and ‘aesthetic emotions’ to occur together and ‘reinforce’ each other in some circumstances, but to more or less ‘interfere’ with each other in others. Thus, for instance, an extremely negative aesthetic judgment of a piece of music might ‘over-rule’ a potential (positive) emotional response aroused by other induction-mechanisms, primarily for mechanisms involving ‘higher’ cognitive processes. Yet, some other ‘low-level’ mechanisms (e.g., conditioned responses) might remain ‘immune’ to such quality judgments, due to their ‘modular’ nature (Table 2: see also Section 4.1).

4. Discussion 4.1. Reaping the benefits: Implications of an expanded framework This review has presented a multi-level framework of emotions that might help to explain the occurrence of ‘mixed’ or ‘conflicting’ emotions such as a bitter-sweet combination of joy and melancholy (e.g., a piece could make a listener happy through emotional contagion, but at the same time make the listener sad due to episodic memories associated with the piece.) Indeed, interactions between multiple mechanisms, and the complex emotions they produce together, could explain why some scholars have been inclined to postulate ‘unique’ or ‘music-specific’ emotions, as pointed out by Patel (2010; see also Section 4.2).13

A rule of thumb is that any new theory that hopes to supplant what came before must explain the same phenomena that the old theory did – as well as new ones. The extended framework retains all the BRECVEM mechanisms and can therefore account for the same phenomena as the previous framework could. But what is gained by adding aesthetic judgment as a separate and distinct type of emotion-induction mechanism in the framework? Most clearly, the novel framework may account for ‘aesthetic emotions’, which reflect aesthetic judgments of music. In addition, I will argue that the extended framework can account for several other interesting phenomena in emotional responses to music. One phenomenon that has been much discussed in the field is the so-called ‘pleasurable sadness’ experienced with regard to music (Levinson, 1997b). More specifically, it has been noted that people tend to avoid experiencing sadness in real life, but that they somehow find the experience of sadness pleasurable in music listening. This (apparent) paradox can be addressed in several different ways.

Kivy (1990) would presumably claim that music listeners do not experience any sadness at all – they simply perceive sadness expressed in the music, and confuse emotions perceived with emotions felt. If they experience an emotion, it is not sadness. Indeed, recent findings suggest that music with a sad expression14 can arouse several other emotions in the listener, including nostalgia, peacefulness and wonder (see Juslin et al., in press; Vuoskoski et al., 2012). In such cases, there is no paradox to explain. (It’s not odd that music that arouses positive emotions is experienced as pleasurable – even it features a sad expression, e.g., minor mode, slow tempo,

13 The wide range of mechanisms and their interactions that might be involved during a musical experience is consistent with Perlovsky’s (2012) theory that musical emotions may serve the function of restoring a balance between conceptual differentiation and synthesis (‘wholeness’) in human consciousness. 14 The frequently used term ‘sad music’ is far too imprecise: Does it refer to (a) music with a sad expression, (b) music that arouses sadness in the listener, or (c) both? A Unified Theoretical Framework 27 low pitch, and descending contours). However, while Kivy’s hypothesis can perhaps account for some cases of reported ‘pleasurable sadness’, there is now evidence that music with a sad expression can arouse genuine sadness in listeners (Juslin et al., in press; Vuoskoski & Eerola, 2012), and that some sadness-arousing music may be experienced as pleasurable nevertheless (Huron, 2011; Vuoskoski et al., 2012). Thus, it seems, the paradox remains.

A different account, offered by Huron (2011), is that the aroused sadness will trigger adaptive physiological responses (e.g., the release of prolactin) that will counter-act the painful nature of the sad emotion. The pleasure experienced is due to the consoling and comforting effect of hormones. While there may be some truth to this notion, the account seems inadequate in two ways. First, the physiological effect is actually an ‘after-effect’, and so there is no ‘pleasurable sadness’, there is only pleasure following sadness. (If that turns out to be the case, reducing the time lag between a brief exposure of the music and the self-report would allow only the ‘pure’ sadness occur, before any ‘after-effect’ has occurred.) Like with Kivy’s hypothesis, there’s no real ‘paradox’ here, only a series of events. But, second, if that’s the case, then why do we not attempt to experience sadness to any other event or object in everyday life, in order to obtain a similar ‘after-effect’? Huron’s account arguably fails to address how music is special in this regard, which is what many scholars have claimed (Levenson, 1997b).

Schubert (1996) offered a theory that is better able to account for the special nature of music (or art, more generally). He follows Berlyne’s (1971) suggestion that, “in stimulus situations classifiable as art, there are cues that inhibit the aversion system” (p. 93). Based on a neural model, Schubert suggests that an aesthetic context will activate a node in the neural network that inhibits the displeasure center of the brain. Pleasure then arises from the mere process of activation. (If this were true, one wonders, how would the listener know that what he or she is feeling is ‘sadness’?) However, this account appears inconsistent with recent findings from a study by Vuoskoski et al. (2012), which indicated that experienced sadness to pieces of music was rated as pleasurable to some degree, whereas responses to scary music were rated as only unpleasant. If dissociation from displeasure takes place in an aesthetic context, Vuoskoski and coworkers ask, why is not ‘scary’ music pleasurable also?

The present framework offers another account, which may be more compatible with the above findings. The apparent paradox can be explained in terms of an interaction between two of the mechanisms, the Contagion mechanism and the Aesthetic judgment mechanism. On this view, listeners do indeed experience ‘genuine’ sadness – resulting from the Contagion mechanism – but they also experience pleasure which results from the perceived beauty of the music. If this hypothesis is correct, there is actually a ‘mixed’ emotion of sadness and pleasure. It is not that the sadness per se is a source of pleasure, it only happens to occur together with a percept of beauty. Levinson (2004) notes that, because life itself has a ‘mixed’ character with happiness and sadness, good things and bad, music that reflects this bitter-sweet or poignant quality will move us the most.15

It follows from this hypothesis that the ‘pleasurable sadness’ postulated by researchers should ‘dissolve’, once a listener is exposed to a piece of music that is sadness-arousing, although not aesthetically rewarding for the listener. Sadness-arousing capacity and aesthetic attractiveness need to be manipulated independently in a factorial design, in order to dissociate the aesthetic judgment from the aroused sadness. If the music evokes sadness, but is otherwise perceived as ‘ugly’ or ‘terrible’, the apparent ‘pleasure’ associated with the ‘sadness’ should disappear.

15 Part of the beauty may reside in the expression of the music. There is some indication that music with a sad expression is perceived as more expressive than music with other emotional expressions (Juslin, 1997, p. 246). A Unified Theoretical Framework 28

One attractive feature of this account is that it might help to explain the ‘unique’ context of music vis-a-vis other sadness-arousing phenomena in everyday life (‘appreciating beauty’), and that it makes predictions about when sadness will be combined with pleasure as well as when it will not. In addition, by explaining the so-called ‘pleasurable sadness’ in terms of a ‘mixture’ of two emotions, it avoids the rather problematic assertion that an emotion that is per definition ‘negative’ in valence (i.e., sadness) is experienced as ‘positive’, in which case the experienced state arguably fails to meet a basic criterion for sadness.

The present framework can also account for many other examples of ‘mixed’ or ‘conflicting’ emotions to music. One intriguing example is what we could call ‘guilty pleasures’ in music listening, which may occur when one experiences a positive emotion to a piece of music one judges to be of a poor aesthetic quality. It may also account for a type of experience reported by Bicknell (2009) and others, where one is ‘seeing through’ a manipulative attempt to evoke emotions in ‘cheap art’, and still cannot help but being affected. In both cases, an explanation can be sought in the interaction between multiple mechanisms. Simply put, mechanisms of an ‘implicit’ nature (e.g., evaluative conditioning) could be ‘immune’ to any negative judgments concerning aesthetic value.

Finally, the framework may perhaps explain the prevalent attitude among philosophers of art that an aesthetic response should be a ‘detached’ or ‘distanced’ consideration of an art object, which does not let emotions ‘come in the way’ (Juslin et al., 2010). This common view can be understood in terms of the present framework as a consequence of the fact that any emotions aroused by the seven BRECVEM mechanisms (see Section 2.2) could influence the aesthetic judgment, and so potentially ‘bias’ that judgment: “A listener who is often deeply moved by music is not necessarily a good judge of that music’s merit” (Bicknell, 2009, p. 123). Even if the aesthetic judgment itself would result in an emotion, that emotion would be a more ‘true’ reflection of the listener’s aesthetic criteria, than a judgment ‘inflected’ by emotions via other mechanisms – at least if the listener does not view ‘emotion’ as one of the key criteria for his or her aesthetic judgment (e.g., Section 3.2). Huron (2009) suggested that “the disembodied, non-utilitarian notion of aesthetic pleasure posited by Kant cannot easily be reconciled with biology” (p. 157). However, in the present approach, the ‘disinterested perception’ discussed by Kant and other philosophers simply refers to an aesthetic judgment unaffected by emotions aroused by the BRECVEM mechanisms.

4.2. The hurdles ahead: Problems for a multi-level theory The BRECVEM model is an example of a multi-representational approach to the information- processing involved in emotional reactions to music. Such an approach has both strengths and weaknesses. As noted by Dalgleish (2004), a multi-level theory buys the researcher increased explanatory power, though the theory also becomes more complex and difficult to understand. This is clearly the case with the present framework. In defense, however, I quote Brunswik’s (1952) insistence that we deal with the phenomenon under investigation “on an adequate level of complexity” (p. 1). If musical emotions are multiply determined, then this is what we shall have to deal with as researchers.

One crucial implication of the present framework for future research is that, because many of the processes and mechanisms are ‘implicit’ in nature and could occur in parallel, researchers cannot rely merely on phenomenological report or introspection to explain musical emotions. (The music experience is the thing that needs explaining, rather than being that explanation.) Most of what goes on in the causal process might, in fact, not be consciously available. Silvia A Unified Theoretical Framework 29

(2006) notes that “perceived causality and true causality diverge”, since “processes irrelevant to causality influence the attributions that people make”: “people tend to attribute causality to salient stimuli, even when salience is unrelated to the effect” (p. 134). Hence, reports of music experience may sometimes lead us astray. A case in point is the recurrent notion that there are ‘unique’ emotions had when, and only when, we hear music. Researchers supporting this idea have so far been unable to come up with emotional states that occur only in reaction to music. But if the emotions are not ‘unique’, why do we experience music, or art more generally, as so special? The short answer is that the special nature of aesthetic experience comes from all the other aspects of our conscious experience, apart from any emotion. The descriptive system for music experience developed by Gabrielsson (2010, 2011) illustrates that such experiences can comprise physical, behavioural, perceptual, cognitive, existential, and developmental aspects, in addition to emotional ones. Hence, Juslin (2011) suggested that we should be more open to the possibility that, much of what makes musical experiences unique are truly non-emotional aspects such as the conscious perception of musical structure or form and its dynamic changes over time, or conscious thoughts and reflections on the artwork and one’s own response to it. However, the underlying causes of any emotion that might occur are not necessarily related to these conscious impressions. This implies that to study underlying mechanisms, we will have to rely on experimental methods where specific mechanisms are manipulated so as to produce immediate behavioral effects in listeners. Being able to predict and control aroused emotions in terms of specific mechanisms is the ultimate evidence of a valid process-description.

The above reasoning suggests that we must create experimental paradigms that reliably arouse specific emotions in listeners, through each of the mechanisms described in previous sections. To separate the effects of the different mechanisms, one needs to be able to activate as well as suppress specific mechanisms in each case, by manipulating various aspects of the music, the listening situation, and the listener. This may be done in at least three principal ways: Firstly, one might select or manipulate pieces of music in such a manner as to provide or withhold the information required for a certain mechanism to be activated, while leaving or removing other information (the principle of information selection). Secondly, one can design the specific test procedure in such a way that it will prevent the type of ‘information-processing’ required for a mechanism to be activated (the principle of interference). One can for instance give listeners a task that recruits attentional resources to such an extent that visual imagery, also dependent on these resources, will be made impossible. Another possibility could be to use a neurochemical interference strategy; for example, it has been shown that blocking of a specific class of amino acid receptors (NMDA) in the lateral amygdala can interfere with the acquisition of evaluative conditioning (e.g., Miserendino et al., 1990). Yet another form of interference involves the use of transcranial magnetic stimulation (Pascual-Leone et al., 2002). By disrupting brain activity at crucial times and locations, one may prevent specific mechanisms from becoming activated by a musical stimulus. Thirdly, one can manipulate the listener, by creating specific memories during the test procedure, prior to presenting the ‘target’ stimulus (the principle of procedural history) to enable study of mechanisms such as evaluative conditioning and episodic memory.

In addition, converging evidence from multiple sources is required to draw conclusions about specific mechanisms being involved in particular instances. That is, in addition to self-reports of evoked emotions (which one should be able to predict), one may need post-hoc self-reports regarding mechanisms (diagnostic questions), physiological measures and facial expressions, and the use of systematic control conditions that can help to rule out alternative explanations. Preliminary findings suggest that it may indeed be possible to activate particular mechanisms through experimental paradigms (Juslin et al., in press), but interactions between BRECVEM mechanisms and aesthetic judgments have yet to be investigated. A Unified Theoretical Framework 30

When it comes to empirically exploring candidate-criteria for aesthetic value (Section 3.2.4), we may need to break down the domain in more manageable parts (Chatterjee, 2012). That is, it may be necessary to ‘map’ the characteristics of separate aesthetic criteria, similarly to what has been done with various BRECVEM mechanisms (e.g., Janata, 2009; Juslin et al., in press; Steinbeis et al., 2006). Empirical research so far has focused primarily on the beauty criterion, for instance investigating the perceptual (Reber et al., 2004) and neural (Jacobsen et al., 2005) correlates of beauty. However, as important as beauty might be, it does not exhaust the criteria that are relevant to an understanding of aesthetic judgments of music. Sooner or later it will be necessary to examine how various criteria jointly contribute to aesthetic response, for instance by means of judgment studies (cf. Section 3.2.3), using a combination of systematically varied and real performances of music. Which criteria are used by listeners? How are they weighted? What are their ‘function forms’ (e.g., linear, quadratic)? How reliable are the judgments? How might individual differences be explained in terms of background variables, such as expertise? An important issue is whether listeners’ self-reported aesthetic criteria will correspond to their judgment policies in actual judgments (given that some processes might be implicit in nature.) Table 2 offered hypotheses concerning the characteristics of each BRECVEM mechanism. It is not yet possible to present a detailed set of hypotheses for various aesthetic criteria, mainly due to a lack of relevant studies, but developing such hypotheses certainly appears feasible as a goal for future research. This may involve developmental trends for style appreciation (e.g., Gardner, 1973), neural correlates of perceived expressivity in music performances (Chapin et al., 2010), or perceptual correlates of novelty (Berlyne, 1971).

4.3. Coda: Concluding remarks In the present review, I have presented an updated version of a multi-level theory of emotion- causation, which assumes that music arouses emotions through several induction mechanisms that music largely shares with other emotional stimuli in everyday life. It has been argued that the framework can account for a wide range of reactions to music: from mere arousal, ‘chills’, and ‘basic’ emotions (e.g., happiness, sadness, interest, surprise) to more ‘complex’ emotions (e.g., nostalgia, pride) and ‘mixed’ emotions. Moreover, through the addition of a mechanism in terms of aesthetic judgment, the framework may also account for ‘aesthetic emotions’ (e.g., awe, interest, admiration, pleasure).

I have further suggested that emotional reactions to music may occur in simultaneous parallel with a non-emotional aesthetic judgment process or as a direct result of that judgment process (depending on the outcome of the process). ‘Everyday emotions’ (aroused by the BRECVEM mechanisms) and aesthetic judgments may influence each other in continuous feedback loops, which can explain several peculiar findings in the field (e.g., ‘guilty pleasures’, ‘enjoyment of sadness’) and the recurrent idea that aesthetic response should involve ‘disinterested pleasure’ (“the dry appreciation of the beauty of form”, Harré, 1997 p. 116). Treatises on the aesthetics of music are presumably expected to address Hanslick’s (1986) influential ideas at some point, and based on the present framework, one could argue that Hanslick was both right and wrong: In agreement with Hanslick, I assume that aesthetic judgment precedes and leads to aesthetic feeling. However, contrary to Hanslick, listeners do regard emotional arousal as an important criterion for the aesthetic value of music. And it is the listener’s criteria that count, if the goal is to explain actual responses to music.

In this review, I have adopted the position that there are no emotions that are unique to music, in the sense that they only occur during music listening (although some emotions are certainly more common than others in music contexts). But for the sake of argument, what if there were A Unified Theoretical Framework 31 a music-specific emotion, what would that emotion be like? Such a posit would have to begin with a consideration of what is unique about the experience of music more generally, because as noted by Clarke (2012b, p. 209), many aspects of our conscious experience of music might not have a uniquely special significance, as compared to other forms of human consciousness. Music is ‘experienced in time’, we are being told, but which aspect of human consciousness is not?

I would like to submit that one unique aspect of the musical experience is that it (to the extent we can imagine ourselves this) resembles listening to a language for which we can understand the (semi)syntax, but not the semantics (except for some nonverbal inflections that we refer to as expression). In listening to music, there is a strong sense that something ‘highly structured’ and ‘meaningful’ is being said, but our brain cannot make out what it is. The resulting feeling might be construed as ‘mild excitement’, mixed with ‘confusion’, as our brain is continuously ‘hooked’ by this subtle, yet inexplicable, language-like structure that we call ‘music’. Imagine the allure of this strange language, if it is also able to mysteriously evoke ‘everyday emotions’ (through the BRECVEM mechanisms) or ‘aesthetic emotions’ (through an aesthetic judgment of, say, its beauty).

If these notions appear highly speculative to the reader, this is only in keeping with the overall tenor of this review, which has offered more preliminary ideas than ‘hard’ empirical evidence. This is consistent with my belief that, what we need most at the current stage, is an abundance of ideas that may be subjected to empirical testing in order to be supported or discarded. Only then can we begin our journey toward a unified theoretical framework which may reconciliate previously distinct approaches to the mystery of emotional reactions to music.

Acknowledgments This research was supported by the Swedish Research Council, through a grant to the author (nr. 421-2010-2129). I would like to thank Alf Gabrielsson, John Sloboda, Leonid Perlovsky, Goncalo Barradas, and the anonymous referees for helpful comments on an earlier version of this manuscript. Correspondence concerning this article may be addressed to: Patrik N. Juslin, Department of Psychology, Uppsala University, Box 1225, SE-751 42 Uppsala, SWEDEN.

References Arnheim, R. (1974). Art and visual perception. Berkeley, CA: University of California Press.

Baumgartner, H. (1992). Remembrance of things past: Music, autobiographical memory, and emotion. Advances in Consumer Research, 19, 613-620.

Beardsley, M. (1982). The aesthetic point of view: Selected essays. Ithaca, NY: Cornell University Press.

Bell, C. (1914). Art. London: Chatto & Windus.

Berlyne, D. E. (1960). Conflict, arousal and curiosity. New York: McGraw-Hill.

Berlyne, D. E. (1971). Aesthetics and psychobiology. New York: Appleton Century Crofts.

Bicknell, J. (2009). Why music moves us. New York: Palgrave-Macmillan.

A Unified Theoretical Framework 32

Blair, M. E., & Shimp, T. A. (1992). Consequences of an unpleasant experience with music: A second-order negative conditioning perspective. Journal of Advertising, 21, 35-43.

Bolders, A. C., Band, G. P. H, & Stallen, P. J. (2012). Evaluative conditioning induces changes in sound valence. Frontiers in Psychology: Emotion Science, 3 (106), 1-12.

Boyd, J., & George-Warren, H. (1992). Musicians in tune. Seventy-five contemporary musicians discuss the creative process. New York: Fireside.

Bregman, A. S. (1990). Auditory scene analysis: The perceptual organization of sound. Cambridge, MA: The MIT Press.

Brehmer, A., & Brehmer, B. (1988). What have we learned about human judgment from thirty years of policy capturing? In B. Brehmer & C. R. B. Joyce (Eds.), Human judgment: The SJT view (pp. 75-114). Amsterdam: Elsevier.

Bressler, S. L., & Menon, V. (2010). Large-scale brain networks in cognition: Emerging methods and principles. Trends in Cognitive Sciences, 14, 277-290.

Brunswik, E. (1952). The conceptual framework of psychology. Chicago: Chicago University Press.

Brunswik, E. (1956). Perception and the representative design of psychological experiments. Berkeley, CA: University of California Press.

Budd, M. (1985). Music and the emotions. The philosophical theories. London: Routledge.

Bullot, N. J., & Reber, R. (2013). The artful mind meets art history: Toward a psycho- historical framework for the science of art appreciation. Behavioral and Brain Sciences, 36, 123-137.

Burke, E. (1958). A philosophical enquiry into the origin of our ideas of the sublime and beautiful. London: Routledge & Kegan Paul.

Cabeza, R., & Nyberg, L. (2000). Imaging cognition II: An empirical review of 275 PET and fMRI studies. Journal of Cognitive Neuroscience, 12, 1-47.

Carroll, N. (1997). Art, narrative, and emotion. In M. Hjort & S. Laver (Eds.), Emotion and the arts (pp. 190-211). New York: Oxford University Press.

Chapin, H., Jantzen, K., Kelso, J. A. S., Steinberg, F., & Large, E. (2010). Dynamic emotional and neural responses to music depend on performance expression and listener experience. PloS One, 5, e13812.

Chatterjee, A. (2012). Neuroaesthetics: Growing pains of a new discipline. In A. P. Shimamura & S. E. Palmer (Eds.), Aesthetic science: Connecting minds, brains, and experience (pp. 299-317). New York: Oxford University Press.

Clark, A., & Toribio, J. (1994). Doing without representing? Synthese, 101, 401-431.

A Unified Theoretical Framework 33

Clarke, E. F. (2012a). Creativity in performance. In D. J. Hargreaves, D. Miell & R. A. R. MacDonald (Eds.), Musical imaginations: Multidisciplinary perspectives on creativity, performance, and oerception (pp. 17-30). New York: Oxford University Press.

Clarke, E. F. (2012b). Music perception and musical consciousness. In D. Clarke & E. Clarke (Eds.), Music and consciousness: Philosophical, psychological, and cultural perspectives (pp. 193-213). New York: Oxford University Press.

Clayton, M., Sager, R., & Will, U. (2005). In time with the music: the concept of entrainment and its significance for ethnomusicology. European Meetings in Ethnomusicology, 11, 3-75.

Clynes, M. (1977). Sentics: The touch of emotions. New York: Doubleday.

Collingwood, R. G. (1938). Principles of art. Oxford: Clarendon Press.

Coltheart, M. (1999). Modularity and cognition. Trends in Cognitive Sciences, 3, 115-120.

Cook, N., & Dibben, N. (2010). Emotion in culture and history: Perspectives from musicology. In P. N. Juslin & J. A. Sloboda (Eds.), Handbook of music and emotion: Theory, research, applications (pp. 45-72). New York: Oxford University Press.

Cooke, D. (1959). The language of music. London: Oxford University Press.

Cooksey, R. W. (1996). Judgment analysis. Theory, methods, and applications. New York: Academic Press.

Crozier, W. R., & Chapman, A. J. (1981). Aesthetic preferences, prestige, and social class. In D. O’Hare (Ed.), Psychology and the arts (pp. 242-278). Brighton, UK: Harvester.

Cupchik, G. C., & Laszlo, J. (1992). Emerging visions of the aesthetic process: Psychology, semiology, and philosophy. New York: Cambridge University Press.

Dahlstedt, S. (1990). Musikestetik [Music aesthetics]. Uppsala: Department of Musicology, Uppsala University.

Dalgleish, T. (2004). Cognitive theories of posttraumatic stress disorder: The evolution of multirepresentational theorizing. Psychological Bulletin, 130, 228-260.

Davidson, R. J., Scherer, K. R., & Goldsmith, H. H. (Eds.). (2003). Handbook of affective sciences. New York: Oxford University Press.

Davies, J. B. (1978). The psychology of music. London: Hutchinson.

Davis, M. (1984). The mammalian startle response. In R. C. Eaton (Ed.), Neural mechanisms of startle behavior (pp. 287-342). New York: Plenum Press.

Demos, A. P., Chaffin, R., Begosh, K. T., Daniels, J. R., & Marsh, K. L. (2012). Rocking to the beat: Effects of music and partner’s movements on spontaneous interpersonal coordination. Journal of Experimental Psychology: General, 141, 49-53.

A Unified Theoretical Framework 34

Dennett, D. C. (1987). The intentional stance. Cambridge, MA: MIT Press.

DeNora, T. (2000). Music in everyday life. Cambridge, UK: Cambridge University Press.

Dewey, J. (1959). Art as experience. New York: Capricorn Books.

Dickie, G. (1974). Art and the aesthetic: An institutional analysis. Ithaca, NY: Cornell University Press.

Donald, M. (1991). Origins of the modern mind: Three stages in the evolution of culture and cognition. Cambridge, MA: Harvard University Press.

Donald, M. (2001). A mind so rare: The evolution of human consciousness. New York: Norton.

Dowling W. J., & Harwood, D. L. (1986). Music cognition. New York: Academic Press.

Dutton, D. (2009). The art instinct: Beauty, pleasure and human evolution. New York: Bloomsbury Press.

Ekman, P. (1992). An argument for basic emotions. Cognition & Emotion, 6, 169-200.

Eysenck, H. J. (1981). Aesthetic preferences and individual differences. In D. O. Hare (Ed.), Psychology and the arts (pp. 76-101). Brighton, UK: Harvester.

Fechner, G. T. (1876). Vorschule der ästhetik. Leipzig: Breitkopf and Hartel.

Forgas, J. (1995). Mood and judgment: The affect infusion model (AIM). Psychological Bulletin, 117, 39-66.

Fox, E. (2008). Emotion science. Basingstoke, UK: Palgrave Macmillan.

Frijda, N. H., & Scherer, K. R. (2009). Affect (psychological perspectives). In D. Sander & K. R. Scherer (Eds.), The Oxford companion to emotion and the affective sciences (p. 10). New York: Oxford University Press.

Gabrielsson, A. (2002). Emotion perceived and emotion felt: Same or different? Musicae Scientiae, Special Issue 2001-2002, 123-147.

Gabrielsson, A. (2010). Strong experiences with music. In P. N. Juslin & J. A. Sloboda (Eds.), Handbook of music and emotion: Theory, research, applications (pp. 547-574). New York: Oxford University Press.

Gabrielsson, A. (2011). Strong experiences with music: Music is much more than just music. New York: Oxford University Press.

Gabrielsson, A., & Juslin, P. N. (2003). Emotional expression in music. In R. J. Davidson, K. R. Scherer, & H. H. Goldsmith (Eds.), Handbook of affective sciences (pp. 503-534). New York: Oxford University Press.

A Unified Theoretical Framework 35

Gardner, H. (1973). Children’s sensitivity to musical styles. Merrill Palmer Quarterly of Behavioral Development, 19, 67-77.

Gärdenfors, P. (2003). How homo became sapiens: On the evolution of thinking. New York: Oxford University Press.

Gibson J. J. (1979). The ecological approach to visual perception. Boston: Houghton Mifflin.

Gombrich, E. H. (1969). Art and illusion. Princeton, NJ: Princeton University Press.

Gombrich, E. H. (1998). Style. In D. Preziosi (Ed.), The art of art history: A critical anthology (p. 150). New York: Oxford University Press.

Gopnik, B. (2012). Aesthetic science and artistic knowledge. In A. P. Shimamura & S. E. Palmer (Eds.), Aesthetic science: Connecting minds, brains, and experience (pp. 129-159). New York: Oxford University Press.

Gowensmith, W. N., & Bloom, L. J. (1997). The effects of heavy metal music on arousal and anger. Journal of Music Therapy, 34, 33-45.

Greenberg, C. (1971). Art and culture: Critical essays. Boston: Beacon Press.

Griffiths, P. E. (2004). Towards a ‘Machiavellian’ theory of emotional appraisal. In D. Evans & P. Cruse (Eds.), Emotion, evolution, and rationality (pp. 89-105). New York: Oxford University Press.

Haidt, J., & Seder, P. (2009). Admiration and awe. In D. Sander & K. R. Scherer (Eds.), The Oxford companion to emotion and the affective sciences (pp. 4-5). New York: Oxford University Press.

Hanslick, E. (1986). On the musically beautiful. (Trans. G. Payzant). Indianapolis, IN: Hackett. (Original work published 1854.)

Hargreaves, D. J., Messerschmidt, P., & Rubert, C. (1982). Musical preference and evaluation. Psychology of Music, 8, 13-18.

Hargreaves, D. J., & North, A. C. (Eds.). (1997). The social psychology of music. New York: Oxford University Press.

Hargreaves, D. J., & North, A. C. (2010). Experimental aesthetics and liking for music. In P. N. Juslin & J. A. Sloboda (Eds.), Handbook of music and emotion: Theory, research, applications (pp. 515-546). New York: Oxford University Press.

Harré, R. (1997). Emotion in music. In M. Hjort & S. Laver (Eds.), Emotion and the arts (pp. 110-118). New York: Oxford University Press.

Harrer, G., & Harrer, H. (1977). Music, emotion, and autonomic function. In M. Critchley & R. A. Henson (Eds.), Music and the brain. Studies in the neurology of music (pp. 202-216). London: William Heinemann medical books.

A Unified Theoretical Framework 36

Heyduk, R. G. (1975). Rated preference for musical composition as it relates to complexity and exposure frequency. Perception and Psychophysics, 17, 84-91.

Hjort, M., & Laver, S. (Eds.). (1997). Emotion and the Arts. New York: Oxford University Press.

Hodges, D. (2010). Psychophysiological measures. In P. N. Juslin & J. A. Sloboda (Eds.), Handbook of music and emotion: Theory, research, applications (pp. 279-311). New York: Oxford University Press.

Hodges, D., & Sebald, D. (2011). Music in the human experience: An introduction to music psychology. New York: Routledge.

Höfel, L. (2003). Ästhetische schwelle [aesthetic threshold]. Unpublished diploma thesis.

Höfel, L., & Jacobsen, T. (2003). Temporal stability and consistency of aesthetic judgments of beauty of formal graphic patterns. Perceptual and Motor Skills, 96, 30-32.

Huron, D. (2006). Sweet anticipation: Music and the psychology of expectation. Cambridge, MA: MIT Press.

Huron, D. (2009). Aesthetics. In S. Hallam, I. Cross, & M. Thaut. (Eds.), The Oxford handbook of music psychology (pp. 151-159). New York: Oxford University Press.

Huron, D. (2011). Why is sad music pleasurable? A possible role for prolactin. Musicae Scientiae, 15, 146-158. (Special Issue on Music and Emotion)

Istok, E., Brattico, E., Jacobsen, T., Krohn, K., Muller, M., & Tervaniemi, M. (2009). Aesthetic responses to music: A questionnaire study. Musicae Scientiae, 13, 183-206.

Jacobsen, T. (2004). Individual and group modeling of aesthetic judgment strategies. British Journal of Psychology, 85, 41-56.

Jacobsen, T. (2006). Bridging the arts and sciences: A framework for the psychology of aesthetics. Leonardo, 39, 155-162.

Jacobsen, T., Schubotz, R. I., Höfel, L., & v. Cramon, D. Y. (2005). Brain correlates of aesthetic judgments of beauty. NeuroImage, 29, 276-285.

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

Janata, P., Tomic, S. T., & Rakowski, S. K. (2007). Characterization of music-evoked autobiographical memories. Memory, 15, 845-860.

Joseph, R. (2000). Neuropsychiatry, neuropsychology, clinical neuroscience. New York: Academic Press.

Juslin, P. N. (1997). Perceived emotional expression in synthesized performances of a short melody: Capturing the listener’s judgment policy. Musicæ Scientiæ, 1, 225-256. A Unified Theoretical Framework 37

Juslin, P. N. (2000). Cue utilization in communication of emotion in music performance: Relating performance to perception. Journal of Experimental Psychology: Human Perception and Performance, 26, 1797-1813.

Juslin, P. N. (2001). Communicating emotion in music performance: A review and a theoretical framework. In P. N. Juslin & J. A. Sloboda (Eds.), Music and emotion: Theory and research (pp. 309-337). New York: Oxford University Press.

Juslin, P. N. (2005a, July). How does music arouse emotions in listeners? Paper presented at the 13th meeting of the International Society for Research on Emotions (ISRE), Bari, Italy.

Juslin, P. N. (2005b). From mimesis to catharsis: expression, perception, and induction of emotion in music. In D. Miell, R. MacDonald, & D. J. Hargreaves (Eds.), Musical communication (pp. 85-115). New York: Oxford University Press.

Juslin, P. N. (2009). Emotional responses to music. In S. Hallam, I. Cross, & M. Thaut (Eds.), The Oxford handbook of music psychology (pp. 131-140). New York: Oxford University Press.

Juslin, P. N. (2011). Music and emotion: Seven questions, seven answers. In I. Deliège & J. Davidson (Eds.), Music and the mind: Essays in honour of John Sloboda (pp. 113-135). New York: Oxford University Press.

Juslin, P. N. (2012). Are musical emotions invariant across cultures? Emotion Review, 4, 283- 284.

Juslin, P. N., & Bänziger, T. (2009). Brunswikian lens model. In D. Sander & K. R. Scherer (Eds.), The Oxford companion to emotion and the affective sciences (pp. 80-81). New York: Oxford University Press.

Juslin, P. N., Harmat, L., & Eerola, T. (in press). What makes music emotionally significant? Exploring the underlying mechanisms. Psychology of Music.

Juslin, P. N., & Isaksson, S. (2013). Subjective criteria for choice and aesthetic value of music: A comparison of psychology and music students. Manuscript submitted for publication.

Juslin, P. N., & Laukka, P. (2003). Communication of emotions in vocal expression and music performance: Different channels, same code? Psychological Bulletin, 129, 770-814.

Juslin, P. N., & Laukka, P. (2004). Expression, perception, and induction of musical emotions: A review and a questionnaire study of everyday listening. Journal of New Music Research, 33, 217-238.

Juslin, P. N., Liljeström, S., Laukka, P., Västfjäll, D., & Lundqvist, L.-O. (2011). Emotional reactions to music in a nationally representative sample of Swedish adults: Prevalence and causal influences. Musicae Scientiae, 15, 174-207. (Special Issue on Music and Emotion)

A Unified Theoretical Framework 38

Juslin, P. N., Liljeström, S., Västfjäll, D., & Lundqvist, L.-O. (2010). How does music evoke emotions? Exploring the underlying mechanisms. In P. N. Juslin & J. A. Sloboda (Eds.), Handbook of music and emotion: Theory, research, applications (pp. 605-642). New York: Oxford University Press.

Juslin, P. N., Liljeström, S., Västfjäll, D., Barradas, G., & Silva, A. (2008). An experience sampling study of emotional reactions to music: Listener, music, and situation. Emotion, 8, 668-683.

Juslin, P. N., & Lindström, E. (2010). Musical expression of emotions: Modeling listeners’ judgments of composed and performed features. Music Analysis, 29, 334-364. (Special Issue on Music and Emotion)

Juslin, P. N., & Sloboda, J. A. (Eds.). (2010). Handbook of music and emotion: Theory, research, applications. New York: Oxford University Press.

Juslin, P. N., & Västfjäll, D. (2008a). Emotional responses to music: The need to consider underlying mechanisms. Behavioral and Brain Sciences, 31, 559-575.

Juslin, P. N., & Västfjäll, D. (2008b). All emotions are not created equal: Reaching beyond the traditional disputes. Behavioral and Brain Sciences, 31, 600-621.

Juslin, P. N., & Zentner, M. R. (2002). Current trends in the study of music and emotion: Overture. Musicae Scientiae, Special Issue 2001-2002, 3-21. (Special Issue on Music and Emotion)

Kant, I. (1790/2000). Critique of the power of judgement (ed. P. Guyer, trans. P. Guyer & E. Matthews). Cambridge: Cambridge University Press.

Kinomura, S., Larsson, J., Gulyás, B., & Roland, P. E. (1996). Activation by attention of the human reticular formation and thalamic intralaminar nuclei. Science, 271, 512-515.

Kivy, P. (1990). Music alone: Philosophical reflections on the purely musical experience. Ithaca, NY: Cornell University Press.

Koelsch, S., Fritz, T., von Cramon, D. Y., Müller, K., & Friederici, A. D. (2006). Investigating emotion with music: An fMRI study. Human Brain Mapping, 27, 239-250.

Koelsch, S. (2012). Brain & music. Oxford: Wiley-Blackwell.

Koneþni, V. J. (1982). Social interaction and musical preference. In D. Deutsch (Ed.), The psychology of music (pp. 497-516). New York: Academic Press.

Koneþni, V. J. (2005). The aesthetic trinity: Awe, being moved, thrills. Bulletin of Psychology and the Arts, 5, 27-44.

Kosslyn, S. M., Thompson, W. L, & Ganis, G. (2006). The case for mental imagery. New York: Oxford University Press.

Kövecses, Z. (2000). Metaphor and emotion. Cambridge: Cambridge University Press. A Unified Theoretical Framework 39

Kraus, R. E. (1986). The originality of the avant-garde and other modernist myths. Boston: MIT Press.

Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago, IL: University of Chicago Press.

Lamont, A., & Eerola, T. (2011). Music and emotion: Themes and development. Musicae Scientiae, 20, 139-145. (Special Issue on Music and Emotion)

Lamont, A., & Greasley, A. (2009). Musical preferences. In S. Hallam, I. Cross, & M. Thaut (Eds.), Oxford handbook of music psychology (pp. 160-168). New York: Oxford University Press.

Lane, R. D. (2000). Neural correlates of conscious emotional experience. In R. D. Lane & L. Nadel (Ed.), Cognitive neuroscience of emotion (pp. 345-370). New York: Oxford University Press.

Lang, P. J. (1979). A bio-informational theory of emotional imagery. Psychophysiology, 16, 495-512.

Langer, S. K. (1957). Philosophy in a new key. Cambridge, MA: Harvard University Press.

Larsen, J. T., & McGraw, A. P. (2011). Further evidence for mixed emotions. Journal of Personality and Social Psychology.

Larson, R. (1995). Secrets in the bedroom: Adolescents private use of media. Journal of Youth and Adolescence, 24, 535-550.

Leder, H., Belke, B., Oeberst, A., & Augustin, D. (2004). A model of aesthetic appreciation and aesthetic judgments. British Journal of Psychology, 95, 489-508.

LeDoux, J. (2000). Cognitive-emotional interactions: Listen to the brain. In R. D. Lane & L. Nadel (Ed.), Cognitive neuroscience of emotion (pp. 129-155). New York: Oxford University Press.

Levinson, J. (1997a). Emotion in response to art. In M. Hjort & S. Laver (Eds.), Emotion and the arts (pp. 20-34). New York: Oxford University Press.

Levinson, J. (1997b). Music and negative emotion. In J. Robinson (Ed.), Music and meaning (pp. 215-241). Ithaca, NY: Cornell University Press.

Levinson, J. (2001). Messages in art. In S. Davies (Ed.), Art and its messages: Meaning, morality, and society (pp. 70-83). University Park, PA: Pennsylvania State University Press.

Levinson, J. (2003). Philosophical aesthetics: An overview. In J. Levinson (Ed.), The Oxford handbook of aesthetics (pp. 3-24). New York: Oxford University Press.

Levinson, J. (2004). Musical chills and other delights of music. In J. W. Davidson (Ed.), The music practitioner (pp. 335-351). Aldershot, UK: Ashgate. A Unified Theoretical Framework 40

Levitin, D. (2010). The world in six songs: How the musical brain created human nature. London: Aurum Press.

Lindström, E., Juslin, P. N., Bresin, R., & Williamon, A. (2003). Expressivity comes from within your soul: A questionnaire study of music students’ perspectives on musical expressivity. Research Studies in Music Education, 20, 23-47.

Lorand, R. (2007). In defense of beauty. On-line article retrieved January 15, 2013 from: http://aesthetics-online.org/articles/index.php?articles_id=34

Lundqvist, L.-O., Carlsson, F., Hilmersson, P., & Juslin, P. N. (2009). Emotional responses to music: Experience, expression, and physiology. Psychology of Music, 37, 61-90.

Lyotard, J.-F. (1994). Lessons on the analytic of the sublime (trans. E. Rottenberg). Stanford, CA: Stanford University Press.

Martindale, C., & Moore, K. (1989). Relationship of musical preference to collative, ecological, and psychophysical variables. Music Perception, 6, 431-455.

McKinney, C. H., Antoni, M. H., Kumar, M., Tims, F. C., & McCabe, P. M. (1997). Effects of Guided Imagery and Music (GIM) therapy on mood and cortisol in healthy adults. Health Psychology, 16, 390-400.

McPherson, G. E., & Schubert, E. (2004). Measuring performance enhancement in music. In A. Williamon (Ed.), Musical excellence: Strategies and techniques to enhance performance. (pp. 61-82). New York. Oxford University Press.

Menuhin, Y. (1996). Unfinished journey. London: Methuen.

Meyer, L. B. (1956). Emotion and meaning in music. Chicago: Chicago University Press.

Miserendino, M. J. D., Sananes, C. B., Melia, K. R., & Davis, M. (1990). Blocking of acquisition but not expression of conditioned fear-potentiated startle by NMDA antagonists in the amygdala. Nature, 345, 716-718.

Moors, A. (2009). Theories of emotion causation: A review. Cognition & Emotion, 23, 625- 662.

Nesse, R. M. (2009). Evolution of emotion. In D. Sander & K. R. Scherer (Eds.), The Oxford companion to emotion and the affective sciences (pp. 159-164). New York: Oxford University Press.

Norris, C. J., Coan, J. A., & Johnstone, T. (2007). Functional magnetic resonance imaging and the study of emotion. In J. A. Coan & J. J. B. Allen (Eds.), Handbook of emotion elicitation and assessment (pp. 440-459). New York: Oxford University Press.

Oatley, K., Keltner, D., & Jenkins, J. M. (2006). Understanding emotions (2nd ed.). Oxford: Blackwell Publishing.

A Unified Theoretical Framework 41

Öhman, A., & Soares, J. J. F. (1994). Unconscious anxiety: Phobic responses to masked stimuli. Journal of Abnormal Psychology, 103, 231-240.

Osborne, J. W. (1980). The mapping of thoughts, emotions, sensations, and images as responses to music. Journal of Mental Imagery, 5, 133-136.

Paisley, W. J. (1964). Identifying the unknown communicator in painting, literature and music: The significance of minor encoding habits. Journal of Communication, 14, 219-237.

Panksepp, J. (1998). Affective neuroscience. New York: Oxford University Press.

Pascual-Leone, A., Davey, N. J., Rothwell, J., Wassermann, E. M., & Puri, B. K. (Eds.). (2002). Handbook of transcranial magnetic stimulation. New York: Oxford University Press.

Patel, A. D. (2008). Music, language, and brain. New York: Oxford University Press.

Patel, A. D. (2010). Music, biological evolution, and the brain. In M. Bailar (Ed.), Emerging disciplines (pp. 91-144). Houston, TX: Rice University Press.

Pearce, M. T., Ruiz, M. H., Kapasi, S., Wiggins, G. A., & Bhattacharya, J. (2010). Unsupervised statistical learning underpins computational, behavioural and neural manifestations of musical expectation. NeuroImage, 50, 302-313.

Perlovsky, L. (2012). Cognitive function, origin, and evolution of musical emotions. Musicae Scientiae, 16, 185-199.

Pinker, S. (1997). How the mind works. New York: W.W. Norton.

Plato. (1923). The republic (trans. B. Jowett). New York: Random House. (Original work published around 380 BC)

Ploog, D. W. (1992). The evolution of of vocal communication. In H. Papousek, U. Jürgens, & M. Papousek (Eds.), Nonverbal vocal communication: Comparative and developmental approaches (pp. 6-30). New York: Cambridge University Press.

Plutchik, R. (1994). The psychology and biology of emotion. New York: Harper-Collins.

Preston, S. D., & de Waal, F. B. M. (2002). Empathy: its ultimate and proximate basis. Behavioral and Brain Sciences, 25, 1-72.

Price, H. E. (1986). A proposed glossary for the use of affective responses in music. Journal of Research in Music Education, 34, 151-159.

Ramachandran, V. S., & Hirstein, W. (1999). The science of art. Journal of Consciousness Studies, 6, 15-51.

Razran, G. (1954). The conditioned evocation of attitudes: cognitive conditioning? Journal of Experimental Psychology, 48, 278-282.

A Unified Theoretical Framework 42

Reber, A. S. (1993). Implicit learning and tacit knowledge: An essay on the cognitive unconscious. New York: Oxford University Press.

Reber, R. (2008). Art in its experience: Can empirical psychology help to assess artistic value? Leonardo, 41, 367-372.

Reber, R., Schwarz, N., & Winkielman, P. (2004). Processing fluency and aesthetic pleasure: Is beauty in the perceiver’s processing experience? Personality and Social Psychology Review, 8, 364-382.

Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of Neuroscience, 27, 169-192.

Robinson, J. (2005). Deeper than reason: Emotion and its role in literature, music, and art. Oxford: Clarendon Press.

Scherer, K. R. (1999). Appraisal theories. In T. Dalgleish & M. Power (Eds.), Handbook of cognition and emotion (pp. 637-663). Chichester, UK: Wiley.

Scherer, K. R., & Zentner, M. R. (2001). Emotional effects of music: Production rules. In P. N. Juslin & J. A. Sloboda (Eds.), Music and emotion: Theory and research (pp. 361-392). New York: Oxford University Press.

Scherer, K. R., & Zentner, M. (2008). Music-evoked emotions are different - more often aesthetic than utilitarian. Behavioral and Brain Sciences, 31, 595-596.

Schoenemann, P. T. (1999). Syntax as an emergent characteristic of the evolution of semantic complexity. Minds and Machines, 9, 309-346.

Schubert, E. (1996). Enjoyment of negative emotions in music. Psychology of Music, 24, 18- 28.

Schulkind, M. D., Hennis, L. K., & Rubin, D. C. (1999). Music, emotion, and autobiographical memory: They are playing our song. Memory and Cognition, 27, 948-955.

Scruton, R. (1997) The aesthetics of music. Oxford: Clarendon Press.

Shapiro, L. (2011). Embodied cognition. New York: Routledge.

Sharpe, R. A. (2004). Philosophy of music: An introduction. Stocksfield: Acumen Publishing.

Sheridan, K. M., & Gardner, H. (2012). Artistic development: Three essential spheres. In A. P. Shimamura & S. E. Palmer (Eds.), Aesthetic science: Connecting minds, brains, and experience (pp. 276-296). New York: Oxford University Press.

Shimamura, A. P. (2012). Toward a science of aesthetics: Issues and ideas. In A. P. Shimamura & S. E. Palmer (Eds.), Aesthetic science: Connecting minds, brains, and experience (pp. 3-28). New York: Oxford University Press.

Silvia, P. J. (2006). Exploring the psychology of interest. New York: Oxford University Press. A Unified Theoretical Framework 43

Silvia, P. J. (2009). Aesthetic emotions (psychological perspectives). In D. Sander & K. R. Scherer (Eds.), The Oxford companion to emotion and the affective sciences (p. 9). New York: Oxford University Press.

Silvia, P. J. (2012). Human emotions and aesthetic experience: An overview of empirical aesthetics. In A. P. Shimamura & S. E. Palmer (Eds.), Aesthetic science: Connecting minds, brains, and experience (pp. 250-275). New York: Oxford University Press.

Simon, H. (1956). Rational choice and the structure of the environment. Psychological Review, 63, 129-138.

Simons, R. C. (1996). Boo! Culture, experience, and the startle reflex. New York: Oxford University Press.

Simonton, D. K. (2001). Emotion and composition in classical music: Historiometric perspectives. In P. N. Juslin & J. A. Sloboda (Eds.), Music and emotion: Theory and research (pp. 205-222). New York: Oxford University Press.

Simonton, D. K., & Damian, R. I. (2012). Creativity. In D. Reisman (Ed.), The Oxford handbook of cognitive psychology (pp. 795-807). New York: Oxford University Press.

Sloboda, J. A. (1989). Music as a language. In F. Wilson & F. Roehmann (Eds.), Music and child development (pp. 28-43). St. Louis, MO: MMB Music.

Sloboda, J. A. (1991). Music structure and emotional response: Some empirical findings. Psychology of Music, 19, 110-120.

Sloboda, J. A. (1992). Empirical studies of emotional response to music. In M. Riess-Jones & S. Holleran (Eds.), Cognitive bases of musical communication (pp. 33-46). Washington, DC: American Psychological Association.

Sloboda, J. A. (1998). Music: where cognition and emotion meet. The Psychologist, 12, 450- 455.

Sloboda, J. A. (1999). Music performance and emotion: Issues and developments. In S. W. Yi (Ed.), Music, mind, and science (pp. 220-238). Seoul, South Korea: Seoul National University Press.

Sparshott, F. (1982). The theory of the arts. Princeton, NJ: Princeton University Press.

Spinoza, B. (1677/1989). Ethica (trans. G. H. R. Parkinson). London: Everyman.

Stecker, R. (2003). Definition of art. In J. Levenson (Ed.), The Oxford handbook of aesthetics (pp. 136-154). New York: Oxford University Press.

Steinbeis, N., Koelsch, S., & Sloboda, J. A. (2006). The role of harmonic expectancy violations in musical emotions: Evidence from subjective, physiological, and neural responses. Journal of Cognitive Neuroscience, 18, 1380-1393.

A Unified Theoretical Framework 44

Teasdale, J. D. (1999). Multi-level theories of cognition-emotion relations. In T. Dalgleish & M. Power (Eds.), Handbook of cognition and emotion (pp. 665-681). Chichester, UK: John Wiley and Sons.

Thompson, S. (2006). Audience responses to a live orchestral concert. Musicae Scientiae, 10, 215-244.

Tolstoy, L. (1960). What is art? (trans. A. Maude). Indianapolis, IN: Hackett Publishing Company. (Original work published 1898)

Tulving, E. (1983). Elements of episodic memory. New York: Oxford University Press.

Tversky, B., & Zacks, J. M. (2013). Event perception. In D. Reisberg (Ed). The Oxford handbook of cognitive psychology (pp. 83-94). New York: Oxford University Press.

Vuoskoski, J. K., & Eerola, T. (2012). Can sad music really make you sad? Indirect measures of affective states induced by music and autobiographical memories. Psychology of Aesthetics, Creativity, and the Arts, 6, 204-213.

Vuoskoski, J. K., Thompson, W. F., McIlwain, D., & Eerola, T. (2012). Who enjoys listening to sad music and why? Music Perception, 29, 311-317.

Wells, A., & Hakanen, E. A. (1991). The emotional uses of popular music by adolescents. Journalism Quarterly, 68, 445-454.

Zajonc, R. B. (2001). Mere exposure: a gateway to the subliminal. Current Directions in Psychological Science, 6, 224-228.

Zatorre, R. J., Halpern, A. R., Perry, D. W., Meyer, E., & Evans, A. C. (1996). Hearing in the mind’s ear: A PET investigation of musical imagery and perception. Journal of Cognitive Neuroscience, 8, 29-46.

Zentner, M. R., Grandjean, D., & Scherer, K. R. (2008). Emotions evoked by the sound of music: Characterization, classification, and measurement. Emotion, 8, 494-521.

A Unified Theoretical Framework 45

Table 1 Definitions of key terms as used in the present review. (adapted from Juslin & Sloboda, 2010, Ch. 1)

______

Affect This term is used as an umbrella term that covers all evaluative - or ‘valenced’ (positive/negative) - states (e.g., emotion, mood, preference). The term denotes such phenomena in general. If that is not intended, a more precise term (e.g., emotion, preference) is used.

Emotion This term is used to refer to a quite brief but intense affective reaction that usually involves a number of sub-components - subjective feeling, physiological arousal, expression, action tendency, and regulation - that are more or less ‘synchronized’. Emotions focus on specific ‘objects’ and last minutes to a few hours. (e.g., happiness, sadness)

Musical emotions This term is used only as a short term for ‘emotions that were somehow induced by music’, without any further implications about the precise nature of these emotions.

Mood This term is used to denote such affective states that are lower in intensity than emotions, that do not have a clear ‘object’, and that are much longer lasting than emotions, several hours to days (e.g., gloomy).

Feeling This term is used to refer to the subjective experience of emotions or moods. One component of an emotion that is typically measured via verbal self-report.

Arousal This term is used to refer to physical activation of the autonomic nervous system. Physiological arousal is one of the components of an emotional response, but could also occur in the absence of emotion (e.g., due to exercise). Arousal is often reflected in the ‘feeling’ component (i.e., the subjective experience).

Preference This term is used to refer to more long-term affective evaluations of objects or persons with a low intensity (e.g., liking of a particular piece or style of music).

Aesthetic judgment This term is used to refer to a subjective evaluation of a piece of music as art based on an individual set of subjective criteria (explained further in Section 3).

______

Table 2: Updated hypotheses for seven psychological mechanisms through which music might arouse emotions in listeners. ______Order Survival value of brain function Information focus Mental representation Mechanism ______

Brain stem reflex 1 Focusing attention on potentially Extreme or rapidly changing (Cued) sensori-motor important changes or events in the basic acoustic characteristics representations close environment

Rhythmic entrainment 2 Facilitating motor coordination Periodic pulses in rhythms, (Cued) sensori-motor in physical work tasks especially around 2Hz representations

Evaluative conditioning 3 Being able to associate objects or Covariation between events (Cued) associative events with positive and negative representations outcomes

Contagion 4 Enhancing group cohesion and Emotional motor expression (Cued) sensori-motor social interaction, e.g. between reminiscent of human voices representations mother and infant

Visual imagery 5 Permitting internal simulations of Self-conjured visual images (Detached) pictorial events that substitute for overt and representations risky actions

Episodic memory 6 Allowing conscious recollections of Personal events in particular (Detached) hierarchically previous events and binding the self places and at particular times organized schematic and to reality pictorial representations

Musical expectancy 7 Facilitating symbolic language with Syntactic information (Detached) schematic and a complex semantics hierarchical representations ______A Unified Theoretical Framework 47

Table 2 (continued) ______Key brain regions Cultural impact Ontogenetic and learning development Mechanism ______

Brain stem reflex The inferior colliculus, the reticulospinal tract of the reticular Low Prior to birth formation, the intralaminar nuclei of the thalamus

Rhythmic entrainment Networks of multiple oscillators in the cerebellum and the Low Prior to birth sensorimotor regions (perception only)

Evaluative conditioning The lateral nucleus of the amygdala, the interpositus nucleus High Prior to birth of the cerebellum

Contagion ‘Mirror neurons’ in the pre-motor regions, right inferior frontal Low First year regions, the basal ganglia

Visual imagery Spatially mapped regions of the occipital cortex, the visual High Pre-school years association cortex, and (for image generation) left temporo- occipital regions

Episodic memory The medial temporal lobe, especially the hippocampus, and the High 3-4 years dorsal medial prefrontal cortex

Musical expectancy The left perisylvian cortex, ‘Broca’s area’, the dorsal region of High 5-11 years the anterior cingulate cortex, the orbital fronto-lateral cortex

______

A Unified Theoretical Framework 48

Table 2 (continued) ______Induced affect Temporal focus of affect Induction speed Degree of volitional influence Mechanism ______

Brain stem reflex General arousal Present High Low Surprise

Rhythmic entrainment General arousal, Present Low Low Feelings of communion

Evaluative conditioning Basic emotions Present High Low

Contagion Basic emotions Present High Low

Visual imagery All possible emotions Omnidirectional Low High

Episodic memory All possible emotions, Past Low Medium but especially nostalgia

Musical expectancy Interest, anxiety, surprise, Present/Future Medium Low chills, hope, disappointment

______

A Unified Theoretical Framework 49

Table 2 (continued) ______Availability to consciousness Modularity Dependence on musical structure Mechanism ______

Brain stem reflex Low High Medium

Rhythmic entrainment Low High Medium

Evaluative conditioning Low High Low

Contagion Low High Medium

Visual imagery High Low Medium

Episodic memory High Low Low

Musical expectancy Medium Medium High

______

A Unified Theoretical Framework 50

Figure captions

Figure 1: Schematic description of the aesthetic judgment process in music listening [COLOR FIGURE in print]

Figure 2 (a): Mean values and standard errors for listeners’ ratings of criteria for aesthetic value of music [COLOR FIGURE in print] (from Juslin & Isaksson, 2013) (b): Individual ratings of criteria for aesthetic value of music by four subjects [COLOR FIGURE in print] (based on data from Juslin & Isaksson, 2013)

Figure 3: Parallel induction of musical emotions via BRECVEM mechanism and aesthetic judgment [COLOR FIGURE in print] A Unified Theoretical Framework 51

A Unified Theoretical Framework 52

5,5

5,0

4,5

4,0

3,5

3,0

2,5

MEAN RATING 2,0

1,5

1,0

0,5

0,0 Taste Beauty Wittiness Sublimity Message Originality Use as art Expression Complexity skill Artistic Representation Artistic intention Emotion arousal CRITERIA

A Unified Theoretical Framework 53

7 Beauty Originality Expression Use as art 6 Artistic skill Emotion arousal Artistic intention

5

4 Rating

3

2

1 1234

Subject

A Unified Theoretical Framework 54