Valerie J. Lauckner-Morano
Joan E. Sussman
Dept. of Commun. Disord. and Sci., State Univ. of New York at Buffalo, Buffalo, NY 14260
In original work Kuhl [Percept. Psychophys. 50, 93--107 (1991)] used a stimulus that was 120 mels away from the prototype (P) stimulus for demonstration of the ``magnet'' effect. Listeners in the current study who phonetically transcribed P and nonprototype (NP) [i] stimuli showed that 92% of responses for the 120-mel P token were not [i]. Further, another group of listeners rated the 64 stimuli as either the [i] in ``peep'' or ``not [i].'' Results showed that only 20% of responses rated the 120-mel token as [i] suggesting that this token along the -1 diagonal was not considreed an [i] member. Thus three other groups of listeners participated in ``change/no-change'' discrimination testing with 500-ms interstimulus intervals for stimuli along the -1 diagonal common to the P and NP distributions. Each group had a different fixed standard in each four-item trial: the Kuhl P token, the 120-mel token, and an 85-mel token. Results will be discussed according to the effects of different standards on discrimination sensitivity and prototype theories in phonetic perception.