This repository contains materials for the tutorial "Analysing Physiological Data Collected During Music Listening: An Introduction" presented at ISMIR 2023 in Milan, Italy.
The full dataset is available here.
Music has diverse effects on listeners, including inducing emotions, triggering movement or dancing, and prompting changes in visual attention. These effects are often associated with psychophysiological responses like changes in heart activity, respiratory rate, and pupil size, which can themselves be influenced by the cognitive effort exerted during music listening, e.g., when engaging with unfamiliar tracks on a web radio for music discovery.
This tutorial aims to introduce psychophysiological data analysis for a broad MIR audience, with a particular focus on the analysis of heart rate, electrodermal activity and pupillometry data. It will be structured in three parts. The first part will provide a presentation of psychophysiological data that we collected in the context of a preliminary study related to music discovery. The second part will be a hands-on tutorial during which we will guide the participants to remake two of our data analyses. In the third part, we will assist participants in undertaking their own data analysis of our data. These analyses will be demonstrated using R and Python.
Our aim with this tutorial is twofold: to promote underrepresented topics in the MIR community, especially the recognition of induced emotions from physiological data and discovery-oriented music recommendation; and to encourage researchers from those domains to interact with the MIR community. The audience we target is therefore relatively large. Participants should, however, possess sufficient knowledge of R and/or Python and standard statistical analysis methods to participate in the hands-on parts of the tutorial.