THE IMPACT OF CONTEXT ON THE PERCEPTUAL ORGANIZATION OF SPEECH

摘要

One of the major functions of the human auditory system involves auditory scene analysis - separating a noisy environment into distinct auditory representations. Streaming is a well-studied laboratory task that is meant to model real-world object recognition, and typically involves the presentation of tonal stimuli in an ABA- type galloping pattern, with low A tones and high B tones1. When the frequency difference between A and B is small, participants typically perceive the tones as a single integrated source (ABA-ABA-...). When the frequency difference between A and B is large, participants tend to perceive two separate streams, known as streaming, with A repeating at twice the rate of B （A-A-A-A and B---B---......）. At an intermediate frequency difference, perception is more ambiguous, though a series of recent studies found that both prior stimulus and prior perception have a large impact on how individuals classify ambiguous sequences [1,2].