Juslin & Västfjäll's (J&V's) article is an impressive feat and will certainly have a profound influence on the scientific community's thinking about musical emotions. I agree with its basic tenet that one must care about the mechanisms by which music evokes emotions. However, identifying and distinguishing different psychological mechanisms is a delicate task. My criticism of J&V's proposal in this respect is partly based on a recent dissertation (Vickhoff Reference Vickhoff2008). Vickhoff's theory is another attempt to clarify the different ways in which music evokes emotions, and maps nicely onto that of J&V in several respects.
An important desideratum on any classification is that it uses a uniform principle. It is not obvious that J&V's classification fulfills this desideratum. In the ordinary use of the terms, brain stem reflexes refers to an anatomical locus and visual imagery describes a certain phenomenology, whereas emotional contagion stands for a sociopsychological chain of events. To be fair to the authors, it is clear that they intend these labels to stand for psychological mechanisms. However, despite their careful attempt in section 3.2 to describe the mechanisms in terms of a number of dimensions, it never becomes clear which of these dimensions are essential, in cases of conflicts between criteria, for distinguishing one mechanism from another.
Here are a few detailed comments on the three aforementioned categories:
“Brain stem reflexes” is anatomically a misnomer for events that may include thalamic neural commands (sect. 3.1.1, para. 3). Apart from that, tying psychological concepts to brain structures always entails a great risk of conceptual confusion when new discoveries change our views about brain function.
The authors' discussion of the category “visual imagery” in my view shows clearly that they are after something quite different from what the term usually represents. They should be, since explaining a musical emotion by reference to visual imagery leaves open why the music evoked the imagery in the first place. Consequently, the authors try to pinpoint a specific process leading to visual imagery that they refer to as a “metaphorical nonverbal mapping” (sect. 3.1.4, para. 4) and that they consider especially relevant for musical emotions. Therefore, it is the latter process, not the nature or significance of visual imagery per se, that the authors ought to characterize in detail. Much of their discussion about visual imagery is actually redundant, given this aim.
Regarding “emotional contagion,” the authors argue that when an emotion is perceived in music (e.g., through a similarity with the human voice), the same emotions are evoked in the listener via internal mimicry of the emotional expression. A parallel is drawn with empathy. However, empathy is not simply mimicry; it includes a complementary emotion that we may call “pity” and is also evoked without any effort or intervention of cognitive mechanisms. In my view, the explanation why we can bear hearing very sad music is that similar complementary emotions are evoked automatically by the music. And note that it is the rule rather than the exception that we like sad music. In subconsciously controlled joint action (Seebanz et al. Reference Seebanz, Bekkering and Knoblich2006), it is also the rule rather than the exception that our actions are complementary to those of the others – not the same. In line with this, many of our spontaneous emotional reactions while playing together or listening to music in a group cannot be characterized as emotional contagion or explained by simple neuronal mirroring.
Vickhoff's solution to the problem of finding a unitary principle of classification is to relate the emotion-evoking mechanisms to different ways of perceiving music. These ways of perceiving are in turn characterized in terms of “perspectives.” A perspective is a perceptual filter that selects certain traits of the stimulus as relevant. Perspectives are heavily context dependent and can therefore be identified through situations where they typically occur. For example, the dyadic perceptual perspective is activated when we perceive another person and her emotions. It is also used when we perceive music as if it were a person. The dyadic perspective entails a mobilization of all the emotion-inducing mechanisms that we know from empathy. The tribal perspective is typical of many group encounters and is active when we listen to music in a group to which we belong. Together, these two perceptual perspectives can possibly explain all the phenomena that “emotional contagion” in J&V's theory is intended to explain. Finally, the nature of the specific mechanism for visual imagery that J&V hint at may be illuminated by Vickhoff's concept of the allocentric perspective. This means perceiving the music as a landscape one can navigate in.
I think that it would be worthwhile for J&V to consider whether some of Vickhoff's conceptual innovations could be integrated in their continued effort to refine their own theory.
I end this commentary with a few minor critical remarks:
1. The psychological mechanisms are said to “become activated by taking music as their ‘object’” (sect. 1, para. 5). In the next sentence, the authors confess their adherence to the view that emotions have intentional objects in the sense of being “about” them. But this signals a confusion, because musical emotions certainly do not always have music as their intentional object: Think, for example, of film music. The source of the confusion may be the standard cognitivist error of anthropomorphizing neural mechanisms (cf. also sect. 3.2, para. 1).
2. The characterization of moods as having low intensity (Table 1) is improper. A severe depression is a mood disturbance and is as such felt very intensely.
3. The characterization of preferences as being of low intensity (Table 1) is also difficult to understand. Preferences are not essentially conscious phenomena that can be compared with other conscious phenomena. And if some kind of behavioral measure of intensity is intended here, preferences can certainly be intense.
4. Some common musically induced emotions are not covered by the theory in the target article: for example, the joy of movement, the sense of flow, and the sense of grace in a technical performance.
