Expertise provides resilience to sound source variability in music perception

Listeners' perception of words and tones is better for acoustically consistent sounds than for variable ones. This is seen in speech adaptation tasks, where listeners are faster to categorize target words when spoken by a single talker than multiple, changing talkers. Recently, this benefit for consistency has been extended to music adaptation tasks, where listeners are faster to categorize tones as low or high when produced by a single instrument than multiple, changing instruments. Here, two experiments sought to clarify the relationship between perceptual expertise and performance in these tasks. Experiment 1 tested listeners of heterogeneous musical backgrounds to identify a suitable musical interval for categorizing tones as low or high for the adaptation task. Experiment 2 tested nonmusicians, intermediate musicians, and experienced musicians on both speech and music adaptation tasks. Across speech and music together, results replicated that all groups of listeners were faster and more accurate in categorizing words/tones from a consistent source than a changing one. Importantly, this consistency benefit was more similar across speech and music tasks for experienced musicians than for nonmusicians. Results are discussed in terms of domain-general accounts of perceptual adaptation and expertise more generally.