Published: 01st November 2020
This study by IIIT-Hyderabad establishes connection between those at-risk of depression and sad music
This study by Dr Vinoo Alluri and two students, Aayush Surana and Yash Goyal, of IIIT-Hyderabad, is titled Tag2risk: Harnessing Social Music tags for characterising depression risk. Read on
The kind of music you listen to tells a lot about you. Dr Vinoo Alluri and her students Aayush Surana and Yash Goyal from the International Institute of Information Technology Hyderabad (IIIT-H) conducted a study that goes a long way in proving this. After analysing the music listening history of 500 individuals via Last.fm, a streaming platform, combined with Healthy-Unhealthy Music Scale (HUMS) and personality questionnaires, they tried identifying depressive tendencies in listeners.
Dr Vinoo Alluri | (Pic: Dr Vinoo Alluri)
Dr Alluri from the Cognitive Science department at IIIT-H has been working on this study for over one and a half years ago. For this study, they looked at the tags attached to each song. Tags like 'sadness', 'broken', 'lonely' were assigned to sad songs that fell in the genre of noise rock and dream pop; tags like 'lovely', 'beautiful', 'wonderful' fell under the genre of alternative and rock. And with these associations they tried to draw conclusions to their study. "We are not saying listening to sad songs is bad because it surely helps one cope, but if that's all that you are doing, then maybe you should be aware of the behaviour," points out Dr Alluri. The professor also draws our attention to listening to music on loop. "Music plays several roles— relaxation, changing mood or validating your feelings. But nothing in excess is good," she points out. Just to cite an example, the professor quotes the Columbine High School massacre, 1999, in the US and how the shooters were known to be fans of electro-punk music with aggressive lyrics.
Aayush Surana and Yash Goyal (in no particular order) | (Pic: Dr Vinoo Alluri)
One of the greatest challenges of conducting this study was gathering the data and understanding which questionnaire will work best. The professor, who was born and brought up in Hyderabad, hopes that they are able to engage the same participants in another survey to see how they fare now. "Eventually, there has to be some intervention in terms of recommendations maybe, so that listeners can slowly move away from their emotional state," she mulls. Maybe collaborations with streaming platforms like Spotify, which could allow them to ask a few questions to listeners or with Fitbit, that allows them to track their heart rate, will give them a lot more answers. "Looking at lyrics and getting a more holistic understanding of what is happening is also an option," she concludes.