We applaud Juslin & Västfjäll's (J&V's) comprehensive overview of mechanisms of music-induced emotion, which is reminiscent of our production rule framework (Scherer & Zentner Reference Scherer, Zentner, Juslin and Sloboda2001). However, whereas we distinguished between central and peripheral route mechanisms (appraisal, memory, and empathy vs. proprioceptive feedback and facilitation), the authors' description of mechanisms tends to confound levels of analysis by addressing phenomena (emotional contagion), content (memory schemata), procedures (visual imagery), and substrata (brain stem). For example, the term emotional contagion describes a phenomenon: the spread of an emotion from one person to another. Yet, the mediating procedures and substrata can be quite different. In our 2001 chapter, we showed how contagion can occur through centrally mediated empathy or peripherally routed rhythmic motor entrainment, a biological mechanism synchronizing body oscillators to external rhythms, including music. Such coupling of rhythms may have powerful emotion-inducing properties. Thus, infants between 6 and 16 months who showed vigorous and synchronized motor entrainment to music and to rhythmic sounds also smiled more frequently (Zentner & Russell Reference Zentner, Russell, Baroni, Addessi, Caterina and Costa2006). Hence, a more important role should be assigned to peripherally mediated emotion induction, especially entrainment.

The overview of potential mechanisms of musical emotion induction provides surprisingly little insight into the conditions necessary to musical emotion induction. Why does music sometimes evoke emotions but often fails to do so? To promote understanding of this important issue we formulated testable hypotheses about determinants and moderators of emotion induction by music (Scherer & Zentner Reference Scherer, Zentner, Juslin and Sloboda2001). Specifically, music emotion experience is conceived as the result of a multiplicative function of structural features (e.g., rhythm, melody, harmony), listener features (e.g., personality, music preference), performer features (e.g., performer skills, state), and contextual features (e.g., type of event, environment). Most likely, these determinants and moderators interact with the respective induction mechanism; for example, rhythm being important for entrainment, performance for empathy, and listener experience for memory.

In some cases, the significance of the authors' claims is difficult to evaluate. For example, the claim that the nature of the induction mechanism is essential for understanding musical emotions (sect. 1, para. 10) remains unsubstantiated if it means that the induction mechanism determines the nature of the ensuing emotion. Different mechanisms of musical emotion induction may lead to similar emotive states, just as different elicitation routes can lead to similar forms of guilt. The statement that “music recruits largely the same mechanisms as do other stimuli that induce emotions” (sect. 4.2, para. 2) is obvious if it means that music, like words, can evoke memories or images and that similar brain mechanisms are involved. However, the statement remains unsubstantiated if it means that there is nothing special about the process of emotion induction by music. Similarly, the second part of that sentence “and that the emotions evoked by music are the same” is obviously true if it means that hearing music can sometimes function like any other emotion-inducing stimulus (one can get angry about loud music or unpleasant grill odors coming from a neighbor's balcony) but is unsubstantiated if it means that emotional experiences evoked by music are completely interchangeable with emotions triggered by other elicitors.

Given the paucity of current empirical knowledge about the nature of music's emotions, theory-building about mechanisms of musical emotion induction can go easily astray. For example, when describing emotional contagion as a possible mechanism, the authors note that, compared with angry speech, “a musical instrument might sound extremely ‘angry’ by virtue of its even higher speed, louder intensity, and harsher timbre. This aspect should render music a particularly potent source of emotional contagion” (sect. 3.1.3, para. 5, emphasis J&V's). This implies that anger should be easy to induce by music through emotional contagion; however, evidence shows that anger is in fact only very rarely induced by music (Laukka Reference Laukka2007; Zentner et al., in press). Similarly, the musical expectancy mechanism (sect. 3.1.6) should, in theory, give surprise special prominence among musically induced emotions. However, in practice, surprise is not a frequently reported response to music (Laukka Reference Laukka2007; Zentner et al., in press).

To provide a more solid base for theory building, we empirically studied music-evoked emotions. In a study of music experts, we showed that music typically generates emotional states other than the standard basic emotions (Scherer et al. Reference Scherer, Zentner and Schacht2001–2002). In support of these initial results, we found in a series of laboratory and field studies that emotive states evoked by music can be differentiated into nine categories (wonder, transcendence, tenderness, peacefulness, nostalgia, power, joyful entrainment, tension, and sadness). We developed a scale to assess these music emotions: the Geneva Emotional Music Scale (GEMS). We showed that a domain-specific model such as that represented by GEMS accounts more powerfully for ratings of music-induced emotions than do instruments derived from basic emotions or affective circumplex theories (Zentner et al., in press).

Although the authors claim not to have seen any evidence for the existence of music-specific emotions (sect. 4.2, para. 2), our past work provides such evidence and suggests that concepts and measures from non-musical areas of emotion research cannot simply be transposed to music. In consequence, we suggested distinguishing between utilitarian and aesthetic emotions. Utilitarian emotions are triggered by the need to adapt to specific situations that are of central significance to the individual's interests and well-being. Such emotions tend to be high-intensity reactions, preparing for action. In contrast, aesthetic emotions are triggered in situations that usually have no obvious material effect on the individual's well-being and only rarely lead to specific goal-oriented responses (Scherer Reference Scherer2004; Zentner et al., in press). In the case of aesthetic emotions, appraisal tends to be intrinsic to the visual or auditory stimulus, based on forms and relationships. As a consequence, aesthetic emotions are more diffusely reactive than proactive. In aesthetic experiences, the individual can savor the nuances of the emotional stirrings for their own sake–just as the wine taster savors the delights of different vintages.

In conclusion, J&V have produced an excellent and thought-provoking survey of the issues. However, some of their claims remain unsubstantiated. Their tendency to blur the boundaries between aesthetic and utilitarian emotions risks further encouraging the widespread reliance on categorical or dimensional emotion models in the study of music's emotions, when, as we have shown, music is capable of inducing a much more nuanced range of emotive states than these traditional models of emotions imply.