The effect of exposure to a single vowel on talker normalization for vowels

The current work investigated the role of single vowels in talker normalization. Following initial training to identify six talkers from the isolated vowel /i/, participants were asked to identify vowels in three different conditions. In the blocked-talker conditions, the vowels were blocked by talker. In the mixed-talker conditions, vowels from all six talkers were presented in random order. The precursor mixed-talker conditions were identical to the mixed-talker conditions except that participants were provided with either a sample vowel or just the written name of a talker before target-vowel presentation. In experiment 1, the precursor vowel was always spoken by the same talker as the target vowel. Identification accuracy did not differ significantly for the blocked and precursor mixed-talker conditions and both were better than the pure mixed-talker condition. In experiment 2, half of the trials had a precursor spoken by the same talker as the target and half had a different talker. For the same-talker precursor condition, the results replicated those in experiment 1. In the different-talker precursor, no benefit was observed relative to the pure-mixed condition. In experiment 3, only the written name was presented as a precursor and no benefits were observed relative to the pure-mixed condition.