Name:Harris Miller

School/Affiliation:McMaster University

Co-Authors:Shu Sakamoto, Emily Wood, Laurel Trainor

Virtual or In-person:In-person

Abstract:

In a multi-sound environment, we can either segregate the individual sound sources into separate streams or integrate them into one single percept. When listening to multiple musical sources, auditory streams are formed automatically in the early processing before being segregated or integrated. Previous studies suggest that bottom-up features of sound, such as timbre proximity, modulate the difficulty of segregation and integration. This study investigates how timbre affects stream formation and, in turn, how stream formation influences segregation and integration. We hypothesize that as the timbres of two streams become similar, the degree of stream formation will decrease, making segregation more difficult. To test this, we have deployed an experimental paradigm where we manipulate the proximity of timbre of two instruments by “morphing” one into the other. The subjects listened to two concurrent musical sounds and were instructed to detect a target sound from one or two concurrent instruments, to segregate or integrate the streams, respectively. The timbre proximity was varied and counter-balanced across the experiment, and their detection accuracy was collected, as well as their EEG, which will be used for a future study to analyze the degree of stream formation. Here, we will report the preliminary analysis to demonstrate the efficacy of the timbre morphing approach to modulate the subjects’ ability to segregate and integrate under a multi-sound environment.