Ear Differences in the Recall of Vowels Produced by Different-Sized Vocal Tracts

Abstract

The five vowels /ɪ, ε, æ, a, ʌ/ in the context /əɲ-t/ were synthesized on the Haskins parallel formant synthesizer using two different sizes of imaginary vocal tract. The formant frequencies for one voice were 25% higher than those for the other. Two experimental dichotic tapes were run on two different groups of 18 subjects. On the first tape, the two sounds constituting a dichotic pair could be both low voice or both high voice or one of each. On the second tape, only the high voice was used. Those subjects who listened to the first tape recalled significantly more vowels over-all from the right than from the left ear. On the second tape, subjects showed no significant ear difference. Moreover, this latter (insignificant) ear difference was significantly smaller than that for those trials in the first experiment, which had the high voice on both ears. A right-ear advantage for vowels thus appears when, within an experiment, there is uncertainty concerning size of vocal tract. [Support by NICHD and the Commonwealth Fund is gratefully acknowledged.]