Abstract
Auditory scene analysis is an elementary aspect of music perception, yet only little research has scrutinized auditory scene analysis under realistic musical conditions with diverse samples of listeners. This study probed the ability of younger normal-hearing listeners and older hearing-aid users in tracking individual musical voices or lines in JS Bach's The Art of the Fugue. Five-second excerpts with homogeneous or heterogenous instrumentation of 2–4 musical voices were presented from spatially separated loudspeakers and preceded by a short cue for signaling the target voice. Listeners tracked the cued voice and detected whether an amplitude modulation was imposed on the cued voice or a distractor voice. Results indicated superior performance of young normal-hearing listeners compared to older hearing-aid users. Performance was generally better in conditions with fewer voices. For young normal-hearing listeners, there was interaction between the number of voices and the instrumentation: performance degraded less drastically with an increase in the number of voices for timbrally heterogeneous mixtures compared to homogeneous mixtures. Older hearing-aid users generally showed smaller effects of the number of voices and instrumentation, but no interaction between the two factors. Moreover, tracking performance of older hearing aid users did not differ when these participants did or did not wear hearing aids. These results shed light on the role of timbral differentiation in musical scene analysis and suggest reduced musical scene analysis abilities of older hearing-impaired listeners in a realistic musical scenario.
Highlights
In a process called auditory scene analysis (Bregman, 1990) the auditory system organizes sound mixtures into auditory events and streams
Considering the main results of all participants included in the analysis, the grand average of the group of young normal-hearing participants (yNH) listeners was 0.79, with 95%-confidence-interval [0.75, 0.83], which was more than 10 percentage points higher than the grand average of older hearing-aid users (oHA) users with 0.67 [0.59, 0.73] without any overlap of confidence intervals and a moderate contribution to the model, β = 0.33 [0.15, 0.53], p < 0.001
YNH and oHA listeners appeared to exhibit a differential pattern of instrumentation effects: Scores of yNH listeners were by eight percentage points lower in the homogeneous condition (0.75 [0.70, 0.79]) compared to the heterogeneous condition (0.83 [0.79, 0.87])
Summary
In a process called auditory scene analysis (Bregman, 1990) the auditory system organizes sound mixtures into auditory events and streams. In the case of polyphonic music, this allows listeners to track distinct musical voices or follow a melody in the midst of an accompaniment. Tracking Musical Voices (McAdams and Bregman, 1979). Relatively little research has tested auditory scene analysis abilities under realistic musical conditions. Coffey et al (2019) presented a music-in-noise task that had listeners hear out musical target melodies and rhythms from a masker signal consisting of four unrelated polyphonic music pieces artificially mixed together, but the ecological validity of this approach remains constrained. There is a scarcity of scene analysis research that reaches beyond young normal-hearing test participants. We attempted to address this void by comparing normal-hearing listeners’ and hearing-aid users’ ability of tracking voices in polyphonic excerpts from the music of JS Bach
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
