Hemispheric specialization for familiar voice recognition: Dichotic listening evidence

Abstract

Previous dichotic listening studies of voice recognition have produced ambiguous results, perhaps due to the use of unfamiliar voices as stimuli: Evidence from brain‐damaged subjects [e.g., D. Van Lancker and J. Kreiman, UCLA Working Papers in Phonetics 62, 50–60 (1985)] suggested that both hemispheres are required to discriminate among unfamiliar voices. In contrast, recognition of familiar voices was observed to require an intact right hemisphere in the unilaterally brain‐damaged population. To obtain data on ear advantages for dichotically presented voice stimuli, therefore, we used voices of politicians and entertainers familiar to many listeners. Subjects monitored one ear at a time and identified both the speaker and the word spoken on each trial. Although not significant, the ear advantage observed for voice recognition did differ significantly from the significant right ear advantage found for word identification. In light of current models of hemispheric specialization, we interpret this relative left ear advantage for voice recognition as a reflection of a right hemisphere superiority for familiar speaker recognition in normal subjects.