MUSIC, MIND & BRAIN

We use big data from music platforms like Shazam and YouTube to investigate macro-level cultural dynamics in the real-world. In a recent study (Lee & Anglada-Tort et al., 2024), we collected a massive dataset of 12.8 million music discoveries from the mobile application Shazam across 1,423 cities and 53 countries. By analysing this data with network inference models, we reconstructed the global diffusion network over which songs spread around the world (see figure above, where circles represent cities and edges patterns of music diffusion).

​

We also analyse large collections of popular songs with music information retrieval techniques to examine trends in audio features (e.g., tempo, valence) and lyrical themes (e.g., feminism, diversity) over time. For example, we analysed over 23,000 chart-topping songs from the past 70 years to study whether music preferences in the UK are influenced by broad environmental factors, such as prevailing weather conditions and seasonal patterns (Anglada-Tort et al., 2023). The results revealed that collective music choices reflect weather and seasonal patterns via mood-regulation mechanisms, influencing song success in the market.