Hyderabad: In an effort that will help identify mental illness in a better way, the Cognitive Science laboratory at IIIT-H has validated the Healthy-Unhealthy Music Scale (HUMS) in an Indian setting. The IIIT-H wing has also employed machine learning to HUMS.
Dr Vinoo Alluri, who leads research on Music Cognition at the Cognitive Science lab, says with machine learning employed to HUMS, it will now be possible to predict if a person is high-risk or low-risk for mental health issues with an accuracy of 80 per cent.
HUMS and its importance
In 2015, a study in Australia probed what music listening strategies of people reveal about their current mental state. As part of this, the team interviewed adolescents suffering from depression in a clinical setting and came up with a 13-point questionnaire, widely known as HUMS or the Healthy Unhealthy Music Scale.
As the name suggests, it has both happy and unhappy items on it. Some of the examples of ‘healthy’ items are ‘music helps me relax’, ‘I feel happier after playing or listening to music’ and so on. The ‘unhealthy’ items include ‘I hide in my music because nobody understands me, and it blocks people out’, ‘I like to listen to songs over and over, even though it makes me worse’, and so on.
The effort was to find a correlation between HUMS and the Kessler Psychological Distress Scale, which is designed to measure anxiety and depression through a 10-item questionnaire. Researchers found out that those who scored high on HUMS unhealthy items also scored high on the Kessler. A high HUMS unhealthy score could be followed up with a screening measure for depression and suicidal ideation to conclusively find out, they said.
To check the validity of HUMS in the Indian context, Alluri and her students Rajat Agarwal and Ravinder Singh tested it out on adults whose average age was 24. They then tried it on 151 respondents working for an IT company. Both these exhibited similar results.
“Our study, which applied machine learning approaches to make accurate predictions of mental well-being from music associations, shows that no matter the age, if people are going through some distress, their musical engagement strategies are similar,” explains Alluri.
Significance of IIIT-H findings
Alluri says: “In India, people do not want to talk about mental illness due to the stigma associated with it. Some of the standard diagnostic tools to assess mental illness may have intrusive questions like, ‘In the past 2 weeks, did you have thoughts that you would be better off dead, or of hurting yourself in some way’”.
It is not easy for people to acknowledge that. “So, we are using HUMS as an indirect way to identify the risk of depression.”
The findings were published in a paper ‘Mining Mental States Using Music Associations”. It was accepted for the Speech, Music, and Mind with Audio Satellite Workshop held at InterSpeech 2019, a conference on the science and technology of spoken language processing, in September in Austria.
“Identifying one’s own listening strategies using HUMS might be suggestive of potential ill-health and can be followed up with additional clinical screening. So, it is all about potential early identification of depression or psychological disturbances so that timely treatment can be sought,” she says.
The team is monitoring music-listening preferences to identify and pre-empt red flags in mental health. “We have begun analysing music listening history of 600 willing participants using a music streaming platform. In addition, we also have their HUMS and personality scores. We are trying to understand what sort of patterns (either textual, acoustic, or habitual) exist in a more realistic music-listening setting which may be indicative of poor mental health,” she says.
While conclusive results are awaited, it is a big step towards overcoming current taboos about mental health that exist in Indian society, she says.