Second Language Perception of Spanish Monophthongal Vowels

This dissertation research examines the acquisition of second language vowels. For second language learners, learning a new languages' sound inventory, particularly its vowels, presents a challenge perceptually and in terms of production. Incorrectly perceiving and producing sounds in a second language results in a foreign accent in many adult learners (Flege, 1995), as mental representations of second language sounds have not been properly established in the mind of the learner (Best, 1991). Vowel sounds are particularly relevant in discussion of foreign-accented speech. Second language vowel perception has received relatively little attention in recent literature, especially for English speakers learning Spanish as a second language. Often, empirical findings are inconsistent with each other. For example, Gordon (2011) found that, perceptually, learners struggle with Spanish vowels /i/ and /e/, interpreting them as English vowels /[near-close near-front unrounded vowel]/ and /[open-mid front unrounded vowel]/, respectively. Garcia de las Bayonas (2004; 2008) indicates that learners struggle most perceiving /u/ and /e/. Most research on the acquisition of these vowels focuses on learner production rather than perception, with equally inconsistent findings (see, for example, Cobb & Simonet, 2015; Menke & Face, 2010). Additionally, the role of syllable stress is often not incorporated into these studies despite the importance of syllable stress in saliency and understanding. These gaps in the literature present an exciting opportunity to understand these acquisitional processes, beginning with the under-studied role of perception. Building on the methodology of Garcia de las Bayonas (2004) and other perceptual studies, this remotely conducted study explores the perception of Spanish vowels by English-speaking learners with special attention paid to the role of syllable stress. Efforts are made to better understand (1) at what point these learners establish the mental representations necessary for Spanish vowel sounds, (2) which English vowels interfere with this learning process, and (3) the role of syllable stress in vowel perception. Participants included 243 Spanish learners in seven different levels of Spanish learning, indicated by class enrollment. The methodology include a discrimination task and an identification task. In the discrimination task, participants hear two disyllabic Spanish words and are asked to decide if the words were different or identical. In the identification task, participants are given a disyllabic Spanish word with one vowel highlighted. Participants are provided with five vowel sounds and select which of the sound options belonged with the highlighted vowel. The options consist of both Spanish and English vowel sound options. Confidence levels (on a scale of 1-7) are reported for all responses. Results from the first task indicate that vowels in stressed syllables are easier to perceive for all groups. For less-experienced learners, discriminating between the Spanish vowels /o/ and /u/ prove to be most difficult, as these vowels differ acoustically much more than their English vowel equivalents. The results of this task indicate that students in their fourth semester of learning (and those more advanced) performed statistically similar to native speaker controls.The identification task reveals that learners struggle to correctly identify the Spanish vowels /e/ and /a/. When asked to identify the vowel /e/, many learners select the English vowels /[open-mid front unrounded vowel]/, and especially /[near-close near-front unrounded vowel]/. For items targeting /a/, many participants select /ae/, /[open back unrounded vowel]/ and the reduced vowel schwa [[schwa]]. These results indicate that these English vowels cause the most interference for learners learning Spanish, resulting in Multiple-category Assimilation (Best, 1991; Escudero & Boersma, 2002). Results from this task also indicate that learners in their 4th semester, and those who are more advanced, tend to make selections statistically similar to native Spanish speakers. The results of this study indicate that by the end of the second year of classroom study, many second language Spanish learners have established mental representations for the vowels of Spanish. Language dominance scores are also collected from each participant (using the Bilingual Language Profile (Birdsong, Gertken, & Amengual, 2012)), which were found to have a strong linear correlation with performance in both tasks, both in terms of accuracy and reported confidence levels. This supports the notion that the more a person uses a language, the better established their phonological inventory will be. Measures of language proficiency and language attitudes are also found to be correlated with participants' accuracy and confidence levels. These findings support theoretical claims made by the Perceptual Assimilation Model (Best, 1991), and shows the process by which mental representations are established for second language phones. Future studies should explore these results in conjunction with vowel production while not forgetting the effect of syllable stress. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]