Musical emotions – unique and complex

I often think very fondly of my time as a PhD student within the Psychology Department of the University of York. I had the best supervisors, the most supportive and caring roommates, and the most stimulating of research environments. In fact, I was jolly spoilt now I think about it! And all this came flooding back to me last week when we had a visiting speaker arrive from my Alma Mater.

Dr Marcel Zentner obtained his PhD from the University of Zurich in 1996. He then moved to Harvard University to carry out postdoctoral research for two years before joining the Emotion research group at the University of Geneva. There he became a research professor at the psychology department in 2001. He finally moved to the University of York in 2007. His research interests are at the cross-roads of personality, emotion and music.

Marcel arrived at York just as I was finishing my PhD so I was not able to benefit from his experience and knowledge as much as would have liked. And it was clear from his talk that I could have learned a bit about good presenting too, as his style is very relaxed and friendly, but informative and insightful

His talk was entitled ‘Music’s emotive specificity’. He presented a series of research studies he has conducted over the last 3 years which aim to characterise musical emotion.  His general argument was that music elicits emotion, no doubt, but that:

1)      Musical emotions are less common than we might think

2)      Musical emotions are harder to trigger in a lab situation than is generally recognised by researchers, because they emerge from the combination of a variety of factors

3)      Musical emotions are typically more a blend of complex emotions as opposed to easily characterised basic emotions (such as ‘happiness’ or ‘sadness’)

The first question for his research for his research was, what do people really feel when they hear music?

He began with a very descriptive, explorative study where he attempted to define a lexicon for musical emotions by asking people to describe their feelings when listening to their favourite music.  From this he generated an unbelievably huge lexicon of 515 terms for musical emotions! He reduced the list by removing synonyms and the like, eventually reaching a still impressive lexicon of 66 adjectives (Zentner et al. 2008).

This study was the starting point for the development of the Geneva Emotional Music Scale (GEMS) which is conceived as a starting descriptive model for musical emotions. The GEMS currently contains 45 labels that were consistently chosen for describing musically evoked emotive states. These states can be grouped into 9 different categories (below), which in turn condense into three “superfactors” (sublimity, vitality, and unease)

• Wonder
• Transcendence
• Tenderness
• Nostalgia
• Peacefulness
• Energy
• Joyful activation
• Tension
• Sadness

Shorter Scales, the GEMS-25 and the GEMS-9 have also been developed, and can all be downloaded from the website above. In conclusion, Marcel states (in a paper to appear in the journal Music Analysis in 2011) that the variety of emotions uncovered in his work has little in common with those defined by basic emotion theory.

In of itself, the GEMS is a great research tool. So far Marcel’s team have used it in a number of very interesting studies. One, where they targeted visitors to a 3 day free music concert held in Geneva. They received 2000 responses across a wide range of musical tastes (the festival is multi-genre). They found that the GEMS was very representative of the kind of blended, complex emotions that people reported feeling in response to music, including ‘seduced’, ‘touched’, ‘enchanted’, and ‘dreamy’.

So are musical emotions really different from al l other emotions? I don’t think that Marcel believes there to be many music-specific emotions, although the patters of emotion ascribed to music is likely to be quite unique. Rather he framed his findings regarding musical emotions within a comparison of Utilitarian and Aesthetic emotion research.

Utilitarian emotions usually involve a reaction to an object in the environment and comprise a level of arousal and physiological upset. These emotional reactions serve to preserve physical and psychological wellbeing. By contrast Aesthetic emotions can occur in the absence of reactive necessity. They possess hedonic or contemplative features and can be experienced simply for their qualities (i.e. to savour them). Marcel argues that musical emotions belong in the latter category. He suggested that their lack of absolutely defined trigger (occurring more often as a result of a combination of patterns within music) leads them to be inherently more blended in their composition. GEMS can be used to look at these blended emotions.

Another point he made regarded how easy it is to experience musical emotions. Marcel hypothesised that they are actually far rarer than some emotion researchers might like to admit. He suggested that the chances of triggering an emotional response to randomly selected clips of music was highly unlikely, as such a reaction to music requires many factors to be in the correct alignment and state of interaction. These include the structure of the music, the performance of the music, the state of the listener and the context of the listening situation.

In conclusion, music is undoubtedly full of emotional delights, but selectively so. Music elicits some emotions more than others and forms unique blends. I particularly enjoyed Marcel’s final comments – he suggested that the true universality may actually lie in our longing and need to experience emotion in music and that this is because musical emotions are actually more precious than we are often lead to believe. After all, if gold was everywhere then no-one would be excited when it was found.

Refs:

Zentner, M. (2011) Homer’s Prophecy On Music’s Emotive Selectivity. To appear in:Music Analysis Special issue on Music and Emotion 2011 (ed. by Michael Spitzer).

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