Different voice part perceptions in polyphonic and homophonic musical textures

Separate voice part perception has been shown in polyphonic music. However, it remains unclear whether this segregation of voice parts is specific to polyphony or also occurs in homophonic music. This study compared voice part perceptions in polyphony and homophony using a redundant signals effect (RSE) paradigm. The RSE means that reaction times are shorter for two simultaneously presented signals than for one of these signals. At the final position of the four-voice homophonic and polyphonic sequences, notes in two voice parts were altered to out-of-key notes independently or simultaneously. Participants ( N = 208) responded to any deviant tones while withholding responses to non-deviant tones. All combinations of deviant voice parts (i.e., soprano–bass, tenor–bass, and alto–tenor) elicited RSEs in polyphonic and homophonic sequences, suggesting separate voice part perception, irrespective of musical texture. However, evidence of the coactivation of separate perceptual modules was obtained only for polyphonic sequences. Deviants in higher voice parts were detected faster and more accurately than those in lower voice parts in both musical textures. These results indicate that voice parts are perceived separately, with a bias toward higher voice parts in both musical textures, but voice parts are more segregated in polyphony.