Perception of phonetic imitation of Southern American English by Midwesterners

Abstract

Phonetic imitation has been observed across regional dialects in English, French, and Mandarin. This cross-dialect imitation is affected by social factors, including dialect prestige and social salience of the linguistic variables. The goal of the current study was to explore the perception of phonetic imitation of Southern American English by native speakers of Midwestern American English in a word shadowing task. The magnitude of the Midwesterners’ phonetic imitation was assessed using an AXB perceptual discrimination task, in which a different group of Midwestern listeners was asked to identify either a baseline read token or a shadowed token as more similar to the model token. The results of the AXB task revealed evidence of overall imitation, consistent with acoustic measures of imitation in word durations, vowel formant frequencies, and vowel formant trajectories. Moreover, perceived imitation was weaker for shadowers who were told where the model talker was from (i.e., Louisville, Kentucky) than for shadowers who were not provided with any information about the model talker, suggesting that explicit information about the model talker’s dialect background can shape the magnitude of cross-dialect phonetic imitation. The specificity of this cross-dialect phonetic imitation across acoustic domains and vowel variables will be discussed.