Dissertations, Theses, and Capstone Projects

Date of Degree


Document Type


Degree Name



Speech-Language-Hearing Sciences


Valerie L. Shafer

Committee Members

Isabelle Barriere

Martin Gitterman

Monica Wagner

Subject Categories

First and Second Language Acquisition | Phonetics and Phonology | Psycholinguistics and Neurolinguistics


Electrophysiology, Mismatch Response, Bilingualism, Speech Perception


Second language (L2) input in the infant and toddler years clearly affects speech processing, particularly for L2 vowels (Cheour, Shestakovab, Ceponieneb, Näätänen 2002), Moreno, Rodriguez-Fornells, Matti, 2008); Rinker, Paavo, Brosch, Kiefer 2010). However, few studies have closely examined how amount of L1 versus L2 input impacts automaticity of speech processing in young children. Greater language use of one than the other language promotes improved speech perception and production in the language of greater use (Flege & Munro 1994; Flege & MacKay 2004). Investigations have used a variety of custom-designed questionnaires to quantify amount of language use, but most have not critically examined the pattern of use is published reports. The current dissertation provides a more fine-grained assessment of language input in relation to L2 speech processing using Event-Related Potentials (ERPs). In particular, the Mismatch Negativity Response (MMN), which is a change-detection measure shows sensitivity to language experience and age of acquisition of L2 (e.g., Moreno et al.; 2008, Hisagi, et al., 2014). In this dissertation, neural mismatch responses (MMRs) to L2 vowel discrimination in relation to language experience are examined. The question posed was whether four - to five-year old Spanish-English bilingual children when compared to monolingual English children experience more difficulty with neural discrimination of the English vowel contrast /I/ in “bit” compared to /e/ in “bet”. We hypothesized that all monolingual participants would show more robust discriminative responses for the / I - e/ contrast due to their greater experience with the English language and that the amount of L1 and L2 input in bilingual children, in a specific setting (i.e. home, leisure, media, literacy) would influence the amplitude of the MMRs. Results revealed a significant effect of group in the amplitude of MMRs. In particular, the bilingual children showed a less negative MMR than monolingual children. Language experience in the Home and Literacy clusters showed a moderate negative correlation with MMR amplitude from 320-380 ms for the four-year-old subgroup. Specifically, greater input in English was associated with greater negativity (i.e., MMN). The Literacy and Media clusters showed a strong relationship with the time interval 400-460 ms in the 5-year-old group, also indicating that greater LU in English was associated with the increased negativity of the MMN. These findings suggest less automaticity of processing the English vowel contrast for bilingual children and that increased experience with English leads to increased automaticity of English vowel processing.