Perceptual segregation of a harmonic from a vowel by interaural time difference in conjunction with mistuning and onset asynchrony.

The two experiments reported here examine how an inter-aural time difference (ITD) interacts with two other cues, mistuning and onset asynchrony, in reducing the contribution of a single frequency component to the perception of a vowel's identity. Previous experiments have shown that although ITD is generally rather ineffective at segregating a simultaneous harmonic frequency component from a vowel, it can produce some segregation when listeners have already been exposed to the isolated segregated component. A difference in ITD increases segregation overall in experiment 1 where the to-be-segregated component can also have a different onset time from the remainder of the vowel, and experiment 2 shows a similar result when the to-be-segregated component is mistuned. However, segregation by ITD is present just as strongly on trials when there is neither mistuning nor a difference in onset-time as on trials where these additional cues are present. Segregation on trials when there is neither mistuning nor a difference in onset-time is however larger in the present experiment which mixed all conditions together than in similar trials in an earlier experiment that had a blocked design [C.J. Darwin and R.W. Hukin, J. Acoust. Soc. Am. 102, 2316-2324 (1997)]. The results show that segregation by ITD increases when other more potent cues are present in the experiment.