Patrick E. Savage · Proceedings of the National Academy of Sciences 2015 · cross-sectional comparative study using phylogenetic methods · n=304 music recordings

Statistical universals reveal the structures and functions of human music

Cited 670 times in the scientific literature.

Level 4 - case-series / case-control

Cross-sectional comparative analysis of recorded media; non-clinical design analogy

OpenAlex W1718516643 · doi:10.1073/pnas.1414495112 · record verified 2026-08-26

What was done

Researchers applied a music-classification scheme combined with statistical and phylogenetic comparative methods to analyze a global sample of 304 music recordings across nine geographic regions, evaluating candidate musical universals across acoustic, stylistic, and social dimensions.

What was found

The analysis detected no absolute universals across the sample. However, it revealed strong support for statistical universals consistent across all nine geographic regions, identifying 18 individual musical features and a network of 10 commonly co-associated features spanning pitch, rhythm, performance style, and social context (such as group singing, percussion, and dancing to repetitive structures). Exact effect sizes, prevalence percentages, and test statistics were not reported in the abstract.

Why it matters

This study provides empirical cross-cultural evidence that human music exhibits robust statistical regularities rather than absolute laws, supporting theories that music evolved to facilitate group coordination and cohesion.

Limits

The abstract reports no exact numerical values, effect sizes, or confidence intervals. The findings are limited to statistical tendencies rather than invariant universal properties, and the dataset is restricted to 304 analyzed recordings, which may not encompass all localized musical variations or performance nuances.

Cited by