Recommender systems, representativeness, and online music: a psychosocial analysis of Italian listeners

TL;DR

This study explores Italian music listeners' perceptions of recommender systems, revealing their routinized usage, limited awareness of algorithmic processes, and the need for psychosocial insights to improve culturally sensitive recommendations.

Contribution

It provides a psychosocial analysis of listener perspectives, highlighting the gap between user experiences and technical system design in music recommendation.

Findings

Listeners engage with platforms routinized but lack understanding of algorithms.

Limited awareness of gender-related representational issues.

Emphasizes integrating psychosocial insights into system design.

Abstract

Recommender systems shape music listening worldwide due to their widespread adoption on online platforms. Growing concerns about representational harms that these systems may cause are increasingly part of the scientific and public debate, wherein music listener perspectives are oftentimes reported and discussed, but rarely contextualised through a psychosocial and cultural lens. We address this gap by interviewing a group of Italian music listeners and analysing their narratives through Emotional Textual Analysis. Our findings reveal that listeners often engage with platforms in routinized ways, yet lack a critical understanding of how recommender systems operate and experience a sense of detachment from algorithmic processes. Moreover, while listeners perceive cultural and linguistic distinctions in music, their awareness of gender-related representational issues remains relatively limited. Overall, results underscore the importance of integrating psychosocial insights with technical approaches in the design of trustworthy and culturally sensitive music recommender systems.