Abstract: How do people across the globe actively shape and reflect their moods through music? This talk bridges two complementary studies that examine the intersection of mood perception, music consumption, and affective patterns on a global scale. The first study leverages a dataset of 765 million streams from 1 million individuals across 51 countries to explore diurnal and seasonal patterns of affective music preferences. We uncover globally consistent trends, such as relaxed music dominating late-night hours and energetic music peaking during business hours, alongside cultural and demographic differences. For example, music in Latin America tends to be more arousing, while Asian preferences skew toward relaxing music. These findings illustrate how musical choice reflects both shared and context-specific mood patterns. The second study investigates cross-cultural perceptions of mood in music and their alignment with automated mood detection algorithms. Analyzing responses from 166 participants in Brazil, South Korea, and the US, we found that basic moods like “cheerful” and “sad” were consistently rated across cultures, while complex moods like “dreamy” and “love” showed notable cultural divergence. Surprisingly, mood detection algorithms demonstrated uniform correlation with human ratings across all cultures, showing no detectable bias toward any particular region. These results suggest that such algorithms, despite their Western training datasets, can serve as objective measures of mood perception within popular music contexts. By combining insights into musical preferences and mood perception, this talk offers implications for cross-cultural emotion research and the design of mood-aware technologies.
Bio: Minsu Park is an Assistant Professor of Social Research and Public Policy at New York University Abu Dhabi. He develops and applies quantitative and computational methods to study the consumption and production of creative work. His current projects focus on how cultural artifacts/interests flow worldwide and how social traces, such as ratings, reviews, and reviewer identities, shape audiences' perceptions and engagements online. His research inhabits an interdisciplinary nexus between data science and social science, simultaneously drawing on and contributing to both, and has been published in top-tier interdisciplinary journals (e.g., Science Advances, Nature Human Behavior). He received his doctorate in Information Science at Cornell University, where he was a member of the Social Dynamics Lab. He is also affiliated with the Center for Data Science at New York University.