For thousands of years music has been used to entertain, to bring people together, to share stories and express emotions. More recently, research has been exploring the extraordinary relationship between music and our mental health, like how it affects our perceptions of pain and trauma.

As the digital world has embedded itself into our day to day lives, the way we interact with music has changed. Where previously we could only listen to music if we had the right equipment, now smartphones and music streaming services make almost any music available on demand. These newer technologies give us new types of data that can help us to understand the links between music and mood in different ways than we have before, which is exactly what our research project ‘Mood Music’ hopes to do.

We’re interested in using music streaming data to understand the ways that people use music to moderate their moods, as well as whether music might be a good indicator of how moods change over time. Changes in our digital behavioural patterns can be indicative of shifts in our overall mental health and well-being, so our digital footprints can be a useful way to understand how mental health works over time. They could also provide personalised wellbeing markers that people can use to help them monitor their own changing moods.

So far we have collected some preliminary data about this, which has shown us that there are some features of music data that change at a population-level in response to certain events. The overall ‘valence’ (how positive a piece of music is perceived as being), how much of a piece of music is speech and how instrumental it is were all associated with mental health outcomes in our preliminary analyses.

This figure shows the trends in the three variables that were most strongly related to mental health outcomes in our preliminary analyses over 2020-21, at the beginning of the COVID-19 pandemic.

We have also run a series of steering group meetings where students have told us their views about how acceptable this type of technology is to them, and talked about how they use music in their lives. These showed that people use music for lots of different reasons, like exercising, cooking or studying, which is likely to mean that music listening data is complex to understand without context.

The students also said that anonymisation, and their data being considered as part of a bigger population group, would make collecting this type of data more acceptable to them. Our steering group have helped us to design a study to try and collect more detailed information about music and mood that we will be running later this year!

For the study we’re hoping to recruit people who use Spotify to share their music listening data with us for two weeks, as well as complete several daily ratings of their mood. Then we can see if there are any patterns in the way that people use music, and how this is related to their mood.

We will be looking for University of Bristol students to take part in this study, and so please do get in touch with Nina at nina.dicara@bristol.ac.uk if you would like to be sent some information about taking part.