This main experiment in this project was designed to test the hypothesis that individual differences in working memory (WM) capacity correlate significantly with individual differences in the ability to speak a second language (L2) fluently. A pilot project was carried out to provide a set of quantifiable factors that produced a reliable description of L2 oral fluency. These factors were related to speed of delivery, pause profiles and morphosyntactic accuracy. In the main experiment, 44 native English speakers who were studying Spanish as a foreign language were tested with a set of three working memory tests, and the scores from these tests were correlated with the scores of three L2 oral fluency tests. The hypothesized strong correlations between working memory capacity and fluency were not found. Furthermore, many of the working memory scores did not correlate strongly with each other. These negative results are explained here partly by reference to the complex nature of speaking in a foreign language, which may tax other faculties more than working memory. Personal and affective variables are also mentioned as a possible explanation, as well as the relationship between working memory and long-term memory stores.

This study uses a matched guise technique to elicit evaluations of men and women from participants based solely on what they hear. Four speakers (two men and two men) created two recordings, one in which they incorporated womens language into their speech and the other using standard language. One hundred university students listened to each recording and evaluated the speaker in terms of twelve personality traits. Results showed a significant difference in how male and female speakers were perceived, regardless of the language style they employed. Womens language and standard language were also perceived differently regardless of speaker gender.

The present study focuses on the processing of relative clauses in Turkish as a second language. The specific purpose of the study is to address the gap in the previous research with regard to why certain relative clause constructions should be more difficult to process than others. For example, in English, object relative clauses such as the lion that the cow carries are more difficult to comprehend and produce than subject relative clauses such as the lion that carries the cow. It has been stated for both L1 and L2 learners that these observed differences in difficulty parallel the implicational relationships in Keenan and Comries (1977) Noun Phrase Accessibility Hierarchy Hypothesis (NPAH). Although there has been some research on this issue, the question of why the acquisition order follows this pattern has never fully been answered since different theories make the same predictions for languages that have been investigated thus far. However, in an SOV language like Turkish, because of its particular structural characteristics, the predictions of those theories diverge, and thus their separate effects can be disentangled. Therefore, the present study explores the issue using the Turkish language. The results of picture selection tasks taken by 20 English and 7 Japanese, Korean and Mongolian learners of Turkish indicate that learners have an easier time with processing object relative clauses than subject relative clauses contrary to the results in the literature for the same construction in other languages. These results have significant implications for the theory of second language acquisition. These implications include, among others, questions about the accuracy of current views of interlanguages (language learner languages) and of the role of language universals in second language acquisition.

This paper analyzes the feedback provided by teachers in three levels of English as a Second Language classes in order to answer the question: Do teachers use different features to distinguish recasts from noncorrective repetition? Three class sessions from three different levels (low-intermediate, high-intermediate, and advanced), each with a different teacher, were video recorded and transcribed. Transcription included the general and specific features that teachers use when providing recasts and noncorrective repetition. Analysis and comparison of the features accompanying recasts and noncorrective repetition show that teachers in this study do use different features when providing recasts or noncorrective repetition. Specific features identified significantly more frequently with recasts (pointing to mouth in a certain position, leaning or moving toward student, stress, over-enunciation of one word or sound, and reduction) serve to highlight the corrective purpose of recasts. On the other hand, the features more often accompanying noncorrective repetition (written support on board or overhead, rising intonation, expansion, and praise) support different purposes such as rebroadcasting an utterance, confirming or acknowledging a students response, or requesting more information.

The advantages of both explicit and incidental vocabulary learning mechanisms have been a subject of ongoing scholarship within the field of Second Language Vocabulary Acquisition (SLVA). Most studies addressing these two types of learning mechanisms have explored them within the context of second language (L2) reading activities. Traditionally, research on explicit and incidental vocabulary has been conducted without computer technology, at least for studies involving English. This thesis examines the opportunities that intermediate ESL learners had to acquire vocabulary while reading pre-selected texts every week using a computer program known as REAP as part of their coursework in the English Language Institute. Students received an individualized series of documents containing target words in a study that was developed as an extension of an earlier study of enhanced learning conditions. The target words consisted of a list of academic words that students did not know. The list was determined by a vocabulary pre-test. Students were told explicitly to try to learn the meanings of their target vocabulary words by clicking on them in order to view online dictionary definitions. Students engaged in explicit learning of target words, though in doing so, they were given the opportunity to use the same online dictionary to look up other non-target words. The learning of non-target words proceeded via incidental learning mechanisms. Data was collected through observations of students, teacher feedback and student-student interviews. The quantitative and qualitative analyses revealed a variety of student learning outcomes and behaviors. There was no relation between non-target and target vocabulary learning outcomes. Students exhibited one of two distinct vocabulary-learning behaviors. One group of students took notes while reading and focusing more on target words. The other mainly asked their teacher vocabulary questions while reading. The results of the study are explored in terms of their pedagogical implications.

This study describes patterns of code-switching, code-mixing and borrowing in the speech and behavior of 25 participants, native Russian speakers and English bilinguals, currently residing in Pittsburgh, PA. Even though these speakers choose to code-mix naturally, I hypothesize that the choice of English verb stems is morphophonologically constrained according to the specific limitations created by Russian linguistic structure and phonemic inventory. I focus on verb borrowing and explore to what degree Russian morphological structure affects the borrowing and code-mixing of the English verbs into predominantly Russian speech. The results reveal that the speakers accept and prefer to code-mix specific verbs according to morphophonological constraints and that age of arrival and time in country correlate with the acceptability judgments of the speakers. I propose that this speech community presents a case of code-mixing, where both languages are unmarked and freely embedded into the matrix of everyday communication of the participants. I utilize Myers-Scottons (1993a) Matrix Language-Frame Model to describe the patterns of code-mixing in this speech community. Her hypotheses are described and contrasted with the findings of this thesis. I postulate that the type of code-mixing that exists in this speech community is not specific to certain practices or speech events. The study demonstrates that the pattern of speech represents a recognized norm within the group. The switch is an unmarked choice for all of the members of this speech community, which in turn defines and separates the group from the other members of the overall Russian-speaking community, specifically by age. This paper suggests that a gap exists in the literature on code-switching as it rarely describes any cases from the Russian-English bilingual community and from any speech community where neither language possesses an indexicality of power and is somehow hierarchically placed within the language matrix of the community. Specific languages appear during specific topics of discussion and discourse contexts, however I suggest that in this particular speech community, the speakers do not assign dominance to Russian or English and code-mix freely, limited and restricted only by discourse and topic issues and morphophonological constraints.

This paper investigates the effect of learning conditions on phrasal verbs in adult ESL learners. It aims to find more effective learning conditions for phrasal verbs taking into account the influence of proficiency and the learners first language. The study, which was designed as an experimental study, includes the following procedures: a pre-test/treatment/post-test. The experiment was conducted using Arabic and Korean learners, and they were divided into two proficiency levels (high and low levels) according to the results of the pre-test. The interaction between the learning condition (translation versus context), the proficiency level, and first language was studied. The semantic properties of phrasal verbs (transparent versus idiomatic phrasal verbs) were also examined. The study suggests that the context learning condition was more beneficial to Arabic participants whereas the translation learning condition was more beneficial to Korean participants.

This study examines the use of the Northern Cities Shift (NCS) and local identity in a group of speakers from the Cleveland, Ohio area. Members of a suburban recreation center participated in audio recorded interviews, during which they answered questions about their personal background, their consumption practices, and their leisure activities, and engaged in conversation about their opinions and memories of the Cleveland area. Local identity was then scored using an identity index modeled after that in Kiesling, et al. (2005). A more locally loyal interview topic received a higher point value, while a less locally loyal topic received a lower point value. The sum of these values formed an overall identity score. Tokens of both (aeh) and (e) were gathered from each speakers interview. Tokens of (o), (oh), (ey), and (iy) were also gathered as reference points within the vowel space of each speaker; these tokens were measured in a controlled, pre-fricative or alveolar stop environment. Using a Praat script, first and second formant measurements were gathered within the first third of each token. These measurements were then normalized. Using multiple regression, the vowel measurements were modeled based on the identity score and other social factors, including sex and age, and linguistic internal factors. This produced no statistically significant results. However, several subsections of the overall identity score were found to be statistically significant.

The primary purpose of this dissertation is to discover whether two current proposals for specific indicators of morphosyntactic development can successfully predict the placement of second language learners of English (ESL learners) in an intensive English program. This research is important because most of the placement/proficiency tests that are currently in use do not include a clear, empirically-tested theory of how second language learners (L2 learners) acquire the morphosyntax of the target language, which is one essential component of L2 proficiency. In order to determine which morphosyntactic elements could be included in a new assessment measure, I examined semi-spontaneous oral production data from 48 ESL learners of mixed L1 background at an intensive English program at the University of Pittsburgh. The measures examined and methodology used were based primarily on Young-Scholten, Ijuin, & Vainikkas (2005) Organic Grammar and Pienemanns (2003) Rapid Profile, two proposals that intend to account for L2 learner development. In order to test the proposals of each, I created implicational tables based on the production data. It was found that Organic Grammar could not fully account for the order of emergence of morphosyntactic features in these data. While Rapid Profile made more accurate predictions, the predictions were not useful in distinguishing between learners at intermediate and advanced levels. Despite these problems, it was possible to combine the results from the Organic Grammar and Rapid Profile tables to produce a new table describing the order of emergence of morphosyntactic forms. It is possible that this table can be integrated into current scale measures of placement/proficiency, such as the ACTFL scale. A preliminary proposal for such a combined measure is proposed; however, further empirical research is necessary in order to determine the effectiveness and accuracy of the scale.

This thesis reports the findings of four non-linguistic experiments with participants from three second language learning groups, students in second semester American Sign Language (ASL2), in fourth semester ASL (ASL4), and students learning Spanish as a point of comparison. These experiments provide evidence that the spatial-visual modality of ASL impacts the effects of language learning. Participants completed two face-processing tasks, the Benton Facial Recognition Test (BFRT) and the Mooney Faces Closure Test (MFCT), and two spatial relations tasks, a Mirror Reversal/Mental Rotation test (MR) and the Differential Aptitude Test-Space Relations (SR). Previous research has found deaf native signers have increased facial recognition skills (McCullough & Emmorey, 1997; Bettger et al., 1997) and that hearing signers have increased face-processing skills (Arnold & Murray, 1998). Deaf late learners of ASL and hearing signers outperformed hearing non-signers on the BFRT (Bettger et al., 1997). However, on the MFCT, signers showed a slight disadvantage (McCullough & Emmorey, 1997). Existing research finds native signers have increased skills on mirror reversal tasks (Masataka, 1995) and mental rotation tasks (Emmorey, Kosslyn, & Bellugi, 1993). Some research has found that hearing ASL L2 participants outperform new ASL L2 participants and non-signers (Talbot & Haude, 1993). Research results are inconsistent about non-linguistic signing advantages. Research on ASL as an L2 is limited. This paper adjoins non-linguistic task results and begins to address when in the L2 progression effects are found. Participants scores on these four tests were analyzed using a series of one-way ANOVAs. When the language group was a significant (p<.05) factor, a post-hoc (Tukeys HSD) analysis determined which language groups significantly differed. ASL2 and ASL4 scores on the MR task were significantly different from the Spanish group. Moreover, Wilcoxon Signed Ranks Test confirmed significant, but not consistent, differences in accuracy between same and reversed test items in the higher rotation categories for each language group. These results suggest that mirror reversal and mental rotation may be separate skills that are both correlated with signing. Results also indicate that ASL may serve as spatial relations training, supporting a psycho-social response for gender differences on spatial relation tasks.