Stability in perceptual assimilation: Talker and vowel context effects Sang-hee Yeon, Ph.D. (Program in Linguistics, University of Florida) Ratree Wayland, Ph.D. (Program in Linguistics, University of Florida) James D. Harnsberger, Ph.D (Dept. of Communication Sciences and Disorders, University of Florida) Jenna Silver (Program in Linguistics, University of Florida) E-mail: ratree@ufl. edu, email@example.com, 3aSC4 INTRODUCTION Cross-language speech (phonetic) perception concerns the perception of speech stimuli that are unfamiliar to the listener because they occur outside of his/her ambient language environment. Predictive models of cross-language speech perception would inform: • General models of perceptual category structure (e.g., how novel speech information is stored in long-term memory) • Models of perceptual learning, particularly second language learning Several models have been proposed to describe and/or predict the discrimination and identification (or “perceptual assimilation”) of non-native contrasts: • Speech Learning Model (SLM), oriented towards L2 perception and production • Native Language Magnet Model (NLM), oriented towards perceptual reogranization during L1 acquisition • Perceptual Assimilation Model (PAM) All three models concern the relationship between the identification of non-native speech sounds and their discrimination / ultimate acquisition. Currently, no model has been developed to predict to identification of non-native speech sounds with native speech sounds. Such a model would require: • A phonetic similarity metric • A model of perceptual categories represented in long-term memory • A mechanism that would accomodate context effects in speech perception This study examines variation in the perceptual assimilation of non-native consonants will vary as a function of the vowel context in which they were produced and the
This is the end of the preview. Sign up
access the rest of the document.