A review of the effect of musical training on neural and perceptual coding of speech

Abstract

Several studies have reported enhanced neural coding of stimulus periodicity (frequency following response; FFR) and/or shorter neural response latencies to speech sounds in musicians than in non-musicians. Such enhanced early encoding may underlie the reported musician advantage for perceiving speech in noise. If such perceptual benefits are confirmed, it is possible that musical training may be an effective intervention for offsetting the decline of this important skill with age. This presentation will review evidence for and against the musician advantage in the neural coding and perception of speech, and will highlight several reasons to be cautious of the generalizability of the musician advantage in the published literature, including (1) small sample sizes, (2) varying definitions of musician and non-musician, (3) dichotomous between-group comparisons (e.g., comparing professional musicians to non-musicians), (4) possible speech-corpus specific effects, and (5) the multitude of degrees of freedom in FFR analyses. Efforts towards large-sample replications that use open science techniques, including preregistered hypotheses, methods and analyses, as well as open source data, should bring clarity to the association between musical expertise and the neural coding and perception of speech in noise. [Work supported by NIH under Grant R01 DC005216 and NSF-BCS under Grant 1840818.]