Dissertations, Theses, and Capstone Projects

Date of Degree


Document Type


Degree Name



Speech-Language-Hearing Sciences


Valerie L. Shafer

Committee Members

Erika Levy

Martin Gitterman

Subject Categories

Speech and Hearing Science


The field of research in bilingualism and second language (L2) acquisition has yielded overwhelming evidence that acquiring a second language later in life will result in less accurate production and perception of consonants and vowels in the second language. These effects, in part, are a result of interference from the already formed phonetic categories shaped by early exposure to the L1 (Iverson, 2007). Phonetic categories from the L2 will, at least initially, be mapped onto phonetic categories from the L1 (Flege, 1995). Shared storage of similar lexical items from L1 and L2 may also take place resulting in differences in processing for words with similar meanings in both languages with similar meanings. Language learners of any age are able to acquire a limitless number of new vocabulary items in their L2. Whether similarities in orthography and/or phonology of semantically similar words affect access to and comprehension of these new L2 lexical items is still unclear. Another question is whether lexical items that differ only in a non-native sound contrast are processed as good or poor exemplars of the L2 word, as a poor exemplar of the L1 word, or as allophonic variation of the L2 word.

In this dissertation neural correlates of L2 words that have or do not have L1 cognates were examined. A group of monolingual English speakers and a group of late Spanish-English bilinguals were asked to decide whether pairs of cognate and non-cognate words were produced the same or differently. Words were pronounced in Standard English or with a change in the production of the stressed vowel in the word to a vowel more similar to a Spanish phoneme. The results revealed that cognate words seemed to facilitate L2 speech discrimination as evidenced by similar responses by bilinguals and monolinguals to these words and smaller or absent responses by bilingual participants to non-cognate words. This facilitation was in the form of a positive ERP response elicited by the frontal electrodes. These results provide a better understanding of why there are mispronunciations and misperceptions of lexical items in an L2 and how shared meaning influences these processes.