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

2025-07-24

AI Summary

This study addresses representational harm in online music recommendation systems, focusing on Italian listeners’ limited psychosocial awareness of algorithmic biases—particularly gender representation imbalances. Employing qualitative methods, we conducted in-depth interviews and applied sentiment-based text analysis alongside narrative analysis to uncover implicit cultural assumptions and cognitive blind spots in users’ discourse. To our knowledge, this is the first systematic investigation of representational mechanisms in music recommendation from integrated psychological and cultural perspectives. Results reveal widespread deficits in algorithmic literacy and critical digital competence: while users perceive recommendation biases, they struggle to connect them to structural determinants, and issues of representation—such as gender—are largely absent from their technical mental models. The findings provide empirical grounding and theoretical insights for enhancing algorithmic transparency and designing culturally responsive recommendation systems. (149 words)

Technology Category

• Knowledge Representation and Reasoning: Preferences
• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Learning Preferences or Rankings

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Responsible Web: Algorithmic accountability and transparency on the web
• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content

Abstract

Recommender systems shape music listening worldwide due to their widespread adoption in online platforms. Growing concerns about representational harms that these systems may cause are nowadays part of the scientific and public debate, wherein music listener perspectives are oftentimes reported and discussed from a cognitive-behaviorism perspective, but rarely contextualised under a psychosocial and cultural lens. We proceed in this direction, by interviewing a group of Italian music listeners and analysing their narratives through Emotional Textual Analysis. Thanks to this, we identify shared cultural repertoires that reveal people's complex relationship with listening practices: even when familiar with online platforms, listeners may still lack a critical understanding of recommender systems. Moreover, representational issues, particularly gender disparities, seem not yet fully grasped in the context of online music listening. This study underscores the need for interdisciplinary research to address representational harms, and the role of algorithmic awareness and digital literacy in developing trustworthy recommender systems.

Problem

• Examining representational harms in music recommender systems
• Analyzing Italian listeners' psychosocial perspectives on algorithms
• Addressing gender disparities in online music recommendations

Innovation

• Interviewing Italian listeners for psychosocial insights
• Using Emotional Textual Analysis on narratives
• Focusing on algorithmic awareness and digital literacy