Universal and cultural emotion communication in music

Hello dear reader. It is a beautiful sunny Sunday morning here in London. One could be forgiven for assuming that spring is on the way after a long chilly winter, but I have a rule to not get excited until I see my first snowdrop flowers. I have seen shoots so far but no flowers so my spring excitement (it is my favourite season you see) will have to wait but hopefully for just a few more days….

I have been enjoying some time in the Twitter-scape recently, which is one way to hear about new, hot-off-the-press papers in music psychology. One fellow music psychologist on Twitter is Professor Tuomas Eerola (@tuomas_ee) who is an author on a new paper looking at emotional communication in music, which I will share with you today.

Big thanks to Tuomas for sending me a copy last night 🙂

Is the emotion we perceive and/or feel in response to music a result of universal reactions to sound or our own cultural exposure over a lifetime. I have written about one small aspect of this debate, the major/minor vs. happy/sad effect, this week in NME. But happy vs. sad is just one angle within a myriad of musical emotions. Can we ever hope to understand what is universal and what is culturally driven about musical emotion?

Our response to music is never, of course, going to be wholly one or the other. We all have similar auditory equipment and perceptual systems so it is only logical to assume that some reactions to musical sound are going to be the same across cultures. The basics of pitch organisation in music seem to be mapped for the properties of our auditory systems, which is either a remarkable coincidence or evidence that our choices we make when we create music follow our understanding of what we are capable of hearing.

Then again, culture is bound to have an influence, as it does in every human intellectual endeavour or technological development be it language, art or music. The authors of the present paper, headed by Petri Laukka, cite a quotation from Ian Cross to make the point that “music takes as many forms as culture” 

Let us start therefore with the viewpoint that our emotional performances of and reactions to music are going to be a mix of universal and cultural principles. The next, and far more complex question, is:

Which aspects of musical emotion expression are common and which differ depending on our musical experiences? Put another way, when do our emotional reactions show an in-group advantage and when do they not? 

Petri Laukka and his team have published an ambitious study of emotional communication in music from the perspective of the performer and the listener. Today I am going to talk about the results from their listeners. I hope next week to go into the results about emotional expressions in performers from different cultures.

THE STIMULI: Musicians from three different musical traditions (Swedish folk music, Hindustani classical music, and Japanese traditional music; 3 musicians each) performed short pieces of music (30 seconds – 1 minutes) to express different emotions. The authors also recorded 3 musicians performing Western classical music. This condition was designed to act as a control, meaning that the researchers expected everyone would react in a similar way to this kind of music due to its more mass exposure.

The musicians were asked to convey an amazing 11 different emotions in the one piece that they selected as representative of their genre: affection, anger, fear, happiness, humour, longing, peacefulness, sadness, solemnity, spirituality, and neutral. The selection was made based on theories of musical expressivity. In the end the researchers recorded 132 musical stimuli (12 musicians across 11 emotions).  I think they deserve a medal just for the stimuli creation!

THE TEST: Listeners were recruited in each of the three different represented countries; Sweden, India and Japan. Each listener was presented with a list of the 11 emotions and asked to chose which one was best represented by each of the 132 musical excerpts. Each person also had to rate their familiarity with all 4 musical traditions to make sure that the manipulation of cultural familiarity was valid. Each session in total took 1-1.5 hours. Another medal to this group for data collection!

RESULTS: As expected, listeners rated Western classical music as more familiar than the other 2 unfamiliar cultures. Only the Indian group rated their music as more familiar than Western classical. It is therefore no surprise that overall accuracy was higher in the Western classical music condition, a result which confirms that this musical form is widespread and not a great candidate for looking at specific cultural reactions (unless you can insure that someone has never heard it before).

In terms of the emotions, the best recognised were (in order, all within a small margin of each other) anger, fear, happiness, humour, and sadness.  Solemnity and spirituality were the most poorly recognised overall.

There were many interesting interactions in the data such as Swedish listeners showed higher recognition rates for fear and longing. Also, Indian music was particularly effective for conveying anger and sadness. Other examples can be found in the paper.

The main result of interest is whether there was a cultural in-group advantage for recognising emotion in music; yes, there was a small overall advantage.

However, there are complicated undercurrents, such that the Japanese listeners, for example, who did not show an in-group advantage and were in fact better with Western music than their own. Here we see the complexities of assuming that people have equal exposure to their cultures musical form just because they come from the country and provide one rating of their subjective familiarity.  Perhaps estimates of time spent actually listening to different musical forms would be helpful in the future?

The second main result of interest is whether there was any evidence for universal perception of emotion in music; yes, listeners were able to deduce the emotional intentions of musicians performing music from outside their culture at above chance levels. 

The authors suggest that this second finding supports the theory that there are universals in the way we express emotion in music across different cultures, and that these arise principally from overlaps between expression in our emotional vocalisations and musical performances.

Overall, the authors did an incredible job with this cross cultural study which must have been a very long time in the making.

Tune in next week for the results of the analysis regarding the universals in the emotional performance of music across cultures.

Paper: Laukka, P., Eerola, T., Thingujam, N. S., Yamasaki, T., & Beller, G. (2013, February 11). Universal and Culture-Specific Factors in the Recognition and Performance of Musical Affect Expressions. Emotion. Advance online publication. doi: 10.1037/a0031388