MODELING THE PERCEPTUAL SEGREGATION OF DOUBLE VOWELS WITH A NETWORK OF NEURAL OSCILLATORS

The ability of listeners to identify two simultaneously presented vowe ls can be improved by introducing a difference in fundamental frequenc y (FO) between the vowels. We propose an explanation for this phenomen on in the form of a computational model of concurrent sound segregatio n, which is motivated by neurophysiological evidence of oscillatory fi ring activity in the auditory cortex and thalamus. More specifically, the model represents the perceptual grouping of auditory frequency cha nnels as synchronised (phase-locked zero phase lag) oscillations in a neural network Computer simulations on a vowel set used in psychophysi cal studies confirm that the model qualitatively matches the performan ce of human listeners; vowel identification performance increases with increasing difference in FO. Additionally, the model is able to repli cate other findings relating to the perception of harmonic complexes i n which one component is mistuned. (C) 1997 Elsevier Science Ltd. All rights reserved.