Type of Document	Dissertation
Author	Ellis, Dyana L. R.
Author's Email Address	dlellis@fsu.edu
URN	etd-05022008-173535
Title	Testing the Limits of Levelt's Loops with Delayed Auditory Playback
Degree	Doctor of Philosophy
Department	Modern Languages, Department of
Advisory Committee	Advisor Name Title Gretchen Sunderman Committee Chair Lara Reglero Committee Member Leigh Edwards Committee Member Michael Leeser Committee Member
Keywords	Language Learning Speech Production Speech Perception Oral Production Psycholinguistic Speech Models Linguistics Audio Playback Listening Skills
Date of Defense	2008-04-23
Availability	unrestricted
Abstract ABSTRACT “To err is human. To self-repair fortunately is also.” (Postma, 2000, p. 98) The overarching purpose of this research is to explore whether second language (L2) learners can identify speech errors in a digital recording of their own oral effort and can correct those errors in subsequent oral efforts. Contemporary psycholinguistic speech production models posit that self-monitoring is an integral and automatic function of online (real time) speech production. Pedagogical research has demonstrated that learners can utilize this self-monitoring/self-analysis mechanism in a post-hoc self-assessment, particularly insofar as writing is concerned, to identify and correct production errors. The current research proposes that in the same way that learners err while writing, they err while speaking and, further, that such errors are not a valid representation of their internal language systems. Hypothetically, therefore, L2 learners can identify their own speech errors via audio playback (AP) given time and opportunity, and providing they have a rudimentary knowledge of the target forms to begin with. To test this assumption, learners were provided a series of monologic Spot the Difference tasks designed to elicit agreement of gender and number in adjective forms. Participants in the experimental procedure group listened to their oral product immediately after task completion while participants in the control group listened to linguistically sterile musical selections. Results of the experiment support the hypothesis that second language learners can identify speech errors in their own oral text, that self-correction efforts do increase and improve, and that acquisition does improve over repeated trials. Interestingly, however, a main effect was noted for post-positional adjectives and not for pre-nominals. A second important aspect of this study involved testing whether lack of developmental readiness (DR) was a constraining factor for AP. Findings revealed that participants who were developmentally unready also improved in self-monitoring and accuracy in subsequent trials. The implications of these findings on current psycholinguistic speech production/perception models as well as their pedagogical implications are discussed at the conclusion of the study.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    EllisDDissertation.pdf 4.50 Mb 00:20:49 00:10:42 00:09:22 00:04:41 00:00:23