What does it mean to ‘hide’ behind music? What could listening to music on loop indicate if it makes the listener feel worse? And could the answers to these questions be connected to mental health?
A team of music psychology researchers from the International Institute of Information Technology in Hyderabad has been working on music listening patterns as indicators of mental health.
Prof Vinoo Alluri from the Institute’s Cognitive Science Lab explains that her team, comprising students Rajat Agarwal and Ravinder Singh, interacted with 300 respondents including young adults and workers in the IT sector as a part of the project.
The team used the “Healthy and Unhealthy Music Scale” (HUMS) which asks respondents 13 questions on their engagement with music. One question asked respondents if ‘I feel happier after playing or listening to music’. Another went ‘I hide in my music because nobody understands me, and it blocks people out’.
“‘Music helps me to relax’ is a ‘healthy association’. There are five such ‘healthy associations’ and eight are ‘unhealthy associations’. It was found that the higher the respondents score on the ‘unhealthy’ items, the higher they score on the Kessler Psychological Distress Scale (if one’s score is above a threshold, the higher the risk of developing mental health issues),” Prof Alluri said.
However, she was quick to clarify that this does not mean that music is bad for the listener. On the contrary about how one uses music indicates one’s current mental state.
The team sought to check whether the results of a similar HUMS study in Australia holds good in India.
“We investigated this (Australian study) in an Indian context as there could be potential differences in the sense that our association with music could be different. We did it with a young adult population — late teens to early 20s — and found very similar results. We also did it in an IT setting with the same results. What we can say is that irrespective of age or setting, our associations with music — why we listen to music and what it means to us — does not seem to change,” Prof Alluri says.
As an extension, the team then used machine learning to predict where respondents stand on the Kessler scale. Data was collected from a music streaming service which allows mining information such as what artiste or artistes the user listens to and when, over a period of time.
“After taking consent, we have users’ HUMS, Kessler measures to identify who is high risk or low risk. We also have personality traits. We are now in the middle of analysing data. It is a greater challenge to understand the patterns. Also, we know that music is beneficial in many ways. But you have to know how to regulate it if you are high risk. This is an ongoing work,” Prof. Alluri says.
