Auditory stream segregation in the songbird forebrain: Effects of time intervals on responses to interleaved tone sequences

For both humans and other animals, the abilities to integrate separate sound elements over time into coherent perceptual representations, or 'auditory streams', and to segregate these auditory streams from other interleaved sounds are critical for hearing and vocal communication. In humans and European starlings (Sturnus vulgaris) the ability to perceptually segregate a simple interleaved tone sequence comprised of two alternating tones differing in frequency (ABA-ABA-ABA-...) into separate auditory streams of A and B tones is promoted at larger frequency separations (ΔF) between the A and B tones. In humans, segregating A and B tones into different streams also appears to be promoted at shorter interstimulus intervals (ISI) between tones within a stream (e.g., between consecutive A tones). Here, we used the ABA experimental paradigm to investigate the influence of different time intervals between A and B tones in repeated ABA triplets on neural responses in the starling forebrain. The main finding from the study is that a ΔF-dependent effect of ISI had a large influence on the relative responses to A and B tones. Responses to B tones were suppressed, relative to A-tone responses, when the A and B tones were more similar in frequency (smaller ΔFs) and occurred at shorter ISIs. We attribute these suppressive effects to physiological forward masking and suggest that forward masking functions as a mechanism for segregating neural responses to interleaved tones in tonotopic space. We discuss the relevance of our physiological data with respect to previous electrophysiological studies of auditory stream segregation in mammals and previous perceptual studies in humans.