There’s simply no escape from a musical world. It seems that wherever you go and however hard you try you just can’t avoid this worldwide phenomenon that touches all of our lives in some way. Dr Raymond McDonald is Professor of Music Psychology at Glasgow Caledonian University. As a musician himself, he knows only too well how different the world would be without music.
Dr Raymond McDonald [RM]
I think it would be a very strange world because music is so completely meshed into our daily life that it is ubiquitous. Now it doesn’t matter where you are or what you are doing we have music nearby all the time, whether it is in a restaurant or on the bus, and the technological revolution that’s taking place allows us now to select our own music to listen to virtually 24 hours a day and music provides such a variety of social functions as well.
The research programme at Glasgow Caledonian is attempting to find out why this ubiquitous music has such power to evoke emotional responses in all of us. As audio engineer and fellow researcher Dr Donald Knox explains this is not simply judging the overall feel of a piece of music, but an attempt to dig down into its individual nuances.
Dr Donald Knox [DK]
One of the things we are really interested in is pitch or the contour of a melody, the range of the tone in the music, as well as finer measures of rhythm patterns. You know we might have two or three songs with exactly the same tempo, but the rhythm patterns might be very different in feel and make the music feel a lot busier or a lot more relaxed. Apart from those acoustical parameters, we are looking at all of those things you might expect to be the domain of the composer and the performer and hoping to build a more coherent model out of those two future sets. Primarily our main aim is to look at those things and build a framework for how it is emotionally communicated by popular and contemporary music using those elements.
[Music playing in the background]
So this is active and it is positively valanced so I would say that it was kind of excited. The strength of the colour inside the cursor indicates the strength of the emotion. So that is expressing negativity.
So using samples of unfamiliar but contemporary popular music styles, the research team has engaged the help of volunteers to help them assess the emotional effect music has on the listener.
[Scott Beverage – PhD student speaking in the background about the research]
Using software he has designed Scott Beverage, a PhD student working on the project, explains how the evaluation process works.
Scott Beverage [SB]
So where the music is playing the participant is encouraged to move the cursor around the emotion space. What we would do is ask the participant to listen to each clip and give their judgement as to the emotion that the music expresses and we would do that using two different factors. First would be arousal and the second is valance, so the interface is represented in four quadrants almost like looking through a site and with the cross sails which give an indication of each of the quadrants. The quadrants are split by arousal and valance. High activity music which is positively balanced. Active music which has negative feelings as angry or afraid, negative music which is sort of passive, is it sort of sad or bored, and passive music which is more positive in the content quadrant or relaxed type music. So a participant plays a piece of music [music playing in the background] and they move the cursor based on what emotion they think the music is expressing, so in this case the audio is quite active and sort of negative kind of angry with the scream so the participant would just move the cursor up to the quadrant that they think, or the emotion that they think the music is expressing. So what we do is take an average of the responses for each track and we get the medium point of all of those different responses and that is the point that represents that audio track. That then creates a ground truth. We are looking for clusters of tracks that have a similar expressed emotion that’s been agreed upon by all those participants so there tends to be general agreement amongst all the participants as to what emotion each one of those pieces of music expresses.
As well as this detailed analysis of the components that make up a piece of music the team aims to study the additional impact of a song’s words on people’s emotional responses as well as how personal experience effects their judgement.
What we want to do in the next few months is to over the same piece of music employ a singer to sing in the same melody lyrics that have been changed by us, to be either positively or negatively stressed, and that will allow us to quantify the real effect or contribution of the lyrics to the emotion that’s conveyed.
From the play listed world of iTunes, Napster and Spotify to musical therapy, analysing the tools used by composers to alter rhythm melody phrasing and pitch is helping us understand how we can use music more effectively to regulate our responses to such things as pain and mental well being.
One of the projects that I was involved in recently took place in Greece where patients who were undergoing kidney dialysis brought in music to listen to during their kidney dialysis.
Dr Raymond MacDonald again.
What we found was that when they were listening to their favourite music they felt less pain and less stress in comparison to a group of people who weren’t listening to music. But why should they feel less pain, was there any commonalities in the music that the people chose, we’re not sure. But this project is helping us understand that. We can if we want take all of the music that was selected during that study and we could analyse it using Dawn and Scott’s procedures and then we can say well the music that the participants chose had these particular qualities and then we might be able to say not only should you like the music, but the music should have these particular qualities as well.
And eventually using this knowledge in a practical sense could see the development of computer programmes to identify pieces of music that could help people with mental health problems, a sort of music on prescription idea. And in a broader context being able to create playlists made up of people’s personal collections and determined by mood could add an extra dimension to our personal enjoyment of music.
MP3 players allow you to browse tags in music files and what you generally get is the artist, the genre, which can mean just about anything, how much music in your iPod is rock. Now we want to move away from those things thinking about what the listener is doing when they are listening to music where they are, what context they are in and I think they need that more Symantec interaction with their music collection. If we can start to present the music to them in different moods then we want to start to develop that relationship between the listener and their music.
But it’s easy to accuse such research of simply underlining what we already know about our emotional responses to music. But the team at Glasgow Caledonian is clear, their EPSRC -funded research is digging far deeper than previous studies have been able to go thanks to the combined disciplines of music psychology and audio engineering.
People have a strong emotional attachment to music. People use music to regulate emotion and it’s really important from a health point of view. There is so much audio now with portable media players that people need to be able to access pieces of music that are going to help them regulate their emotions. I think this is a step towards being able to do that.
If we can go from just making people enjoy their music more or access music that they want to listen to in a specific context more easily, that’s great, but the other side of these things is what if people are going through some pain or are depressed, if they can interact with their music in a more joined up way then that’s a really important area of research.