Academia.eduAcademia.edu

Two challenges in cognitive musicology

Two themes in music cognition research are highlighted—inspired by the contributions in this volume: (a) statistical learning and (b) evolutionary theorizing. Our ability to test alternatives to statistical learning is threatened by the rapidly diminishing opportunities for cross-cultural studies unconfounded by bimusicalism. Our ability to infer possible evolutionary origins for music is confounded by the ''hedonic plenitude'' of modern music-making—where multiple pleasure channels are activated simultaneously. Cognitively inspired music research will benefit by studying a wider range of musical cultures. Evolutionary theorizing will benefit by further work involving comparative animal behavior. The articles in this issue of topics in Cognitive Science offer something of a cross-section of the state of cognitively inspired music research. Although the reports in this volume are not a representative sample of the topic of music cognition, they nevertheless echo two themes that have been dominant in music cognition circles for the past decade or two: statistical learning and evolutionary theory. In this commentary, my aim is to draw attention to some methodological problems associated with these two enthusiasms. 1. Statistical learning Beginning with a classic paper by Krumhansl and Kessler (1982), several lines of music research have converged over the decades, pointing to the importance of simple learning from exposure in the perception of music. This continuing line of research is amply illustrated in the articles in this issue of topiCS Psyche Loui's work uses artificial grammars to demonstrate how statistical learning can be observed in melodic and harmonic perception.