Does the developing lexicon constrain infants’ discrimination of English vowels?

Abstract

Infants initially discriminate native and non-native contrasts. With perceptual reorganization within the first year of life, discrimination of non-native contrasts reduces while the discrimination of native contrasts improves. Several researchers have proposed distributional learning as a domain-general mechanism by which infants’ acquire phonetic categories (e.g., Saffran et al., 1999; Maye et al., 2002). Recent proposals, however, argue for an interactive mechanism where learning words concurrently, supplements distributional learning of phonetic categories (e.g., Swingley, 2009; Feldman et al., 2013). Not only is this interactive mechanism available during the first year of life, its computational implementations outperform ones based on distributional learning alone (Feldman et al., 2013a; Feldman et al., 2013b). We evaluated predictions of the interactive and distributional learning models against infant discrimination data. Data from 4-, and 8-month-olds, learning (a) only English (b) only Spanish or (c) Spanish and English, tested on English (i) /e- ɛ/ (previously published in Sundara & Scutellaro, 2011), (ii) /i- ɪ/ and (iii) /e- ɪ/, were compared. All stimuli were produced by multiple female talkers from the Hillenbrand corpus; infants were tested using a visual fixation procedure with a habituation criterion of 50%. Our results are consistent with infants using a distributional not interactive learning mechanism.