| SLA researches on "task" has been paid much attention to in the past two decades. But in China, there is still some space left for us to explore. The present study, informed by Long's Interaction Hypothesis, set out to explore the effect of task type on learner-learner interaction.Starting from the possible properties (close-ended vs open-ended; required information exchange vs optional information exchange) that may have an influence on interaction, the study adopts three task types: JS --Jigsaw Task (close-ended &required information exchange),DT -- Dictogloss Task (close-ended & required optional information exchange) and DT -- Debate Task (open-ended & optional information exchange), setting L-L interaction as the dependent variable and task type as the independent variable. This thesis attempts to answer the following research questions: 1. What are the differences of the overall quantity of L-L interaction across learners in doing dictogloss, jigsaw and debate? 2. How do task types influence the features of negotiated interaction across learners in doing dictogloss, jigsaw and debate? 3. How does negotiated interaction take place in different task types from a micro perspective? The 78 participants are freshmen from three parallel classes of a 211 university in Nanjing, receiving different treatments respectively. Their oral production was recorded, transcribed, coded and analyzed. 30 dyadic talks were collected as effective data. Quantitative analysis was achieved by means of one-way between subjects ANOVA and multiple comparisons. Furthermore,a CA-based qualitative analysis was made to deeply analyze significant episodes of learner interaction discourse in order to disclose a detailed relationship between task type and learner interaction. The following are the conclusions drawn by the researcher.Firstly, in terms of discourse quantity in L-L interaction across three task types,the word counts of debate task ranks first while the word counts of jigsaw task is the least. Moreover, by multiple comparison, there is significant difference between the JS and DT regarding the discourse quantity. There is no significant difference between JS and DG, DG and DT.Secondly, in terms of the amount of turn-taking across three task types, learners in DG produce the most amount of turn-taking while learners in DT produce the least amount of turn-taking. As for the negotiation types of the three tasks, JS participants tend to produce the most of NoM and a great amount of LREs. In DG,LREs are produced by participants most. The proportion of NoC is the highest in the DT.Regarding the negotiation strategies used by L2 learners in three task types, learners in DG use more scaffolding and metalinguistic talk to negotiate more on formal problems (NoF). Learners in JS tend to negotiate more on meaning (NoM) and form(NoF) while DT task stimulates more argument and opinion conflicts but less negotiation on meaning and even no negotiation on form.Thirdly, qualitative analysis reveals that the number and ratio of negotiation types and strategies can reflect the certain nature of a task type. DG is a close-ended task which requires optional information exchange, so learners tend to use many metalinguistic talks to engage in NoF. JS is a relatively close-ended, required information exchange task, so learners engage in much NoM by means of confirmation check, self-correction, metalinguistic talks and solicitation of assistance.DT is an open task which requires optional information exchange, so learners engage in much NoC by objection strategy.This thesis aims to explore how learners collaborate with each other to achieve co-construction of meaning through interaction in different task types, and how they notice the gaps and form modified output so as to promote language acquisition.Henceforth, some pedagogical implications based on the present study can be drawn for further task-related study in SLA and the task designing in EFL classroom. |
PDF Full Text Request