Effects Of Negative Feedback, Modified Output, Attention And Perception On Efl Development

Interaction research has come a long way since its beginnings nearly three decades ago. A large number of studies on interaction-learning relationship have shown the importance of interactions between native speakers (NSs) and non-native speakers (NNSs), between teachers and learners, and between peers in L2 development, whereas the findings are not yet consistent in the question how interaction can bring about positive effects on L2 development. This dissertation is an attempt to gain insight into the interaction-learning relationship. Specifically, the present study aims to investigate the effects of negative feedback (NF), learners'modified output (MO) to that feedback, learners'noticing target linguistic forms and perceptions about that feedback on EFL development. The research questions are mainly as follows. 1) What is the effect of NF in the form of recasts and clarification requests (CRs) alone and in combination on EFL development in the task-based oral interaction? 2) What is the effect of MO produced in response to NF on EFL development in the task-based oral interaction? 3) What is the effect of noticing target form on EFL development in the task-based oral interaction? 4) What is the effect of learners'perception about NF on EFL development? EFL development in the present study was operationalized as stage development of EFL question formation according to Pienemann and Johnston's developmental sequence for ESL question formation (Pienemann et al. 1988).The present study adopted a pretest-treatment-posttests (immediate and delayed posttests) design, together with information gap and information exchange tasks, stimulated recall protocol, learning journals, and two questionnaires, to explore the research questions. 115 second-year university students participated in the pretest. According to their performance in the pretest, 61 EFL learners at stage 4 were selected for the later experiments. Learners'noticing question forms were assessed by learning journals. Learners'perceptions about NF were assessed through stimulated recall. Data obtained in the study were processed with SPSS (Statistical Package for the Social Sciences) 11.5 for logistical regression, ANOVA, and multiple comparison analysis.The major findings of the present study are as follows. 1) Neither recasts nor CRs were found significantly predictive of EFL question development. The NF in the combining form of recasts and CRs, however, could indirectly contribute to EFL question development by distinguishing the corrective nature of the NF for learners notice, and by creating opportunities for learners to modify their output. 2) The participants did modify their previous output in response to the instructor's NF when they were provided with opportunities. Some learners could modify their output by producing stage 5 questions, while others just repeated their deviant utterance or changed to another topic. Most importantly, the participants'MO involving stage 5 questions was found significantly predictive of EFL question development. 3) The participants did report the noticing of question forms. Their noticing question form was also found a significant predictor for EFL question development. 4) The significant categories of the participants'perceptions involved an error correction, a request for utterance clarification, and a guide for task completion. Recasts followed by CRs were more likely perceived as error correction than CRs, while CRs were more likely regarded as requests for utterance clarification. It was also found that the participants'perceptions had no significant relationship with EFL question development. Nevertheless perceptions could indirectly affect EFL question development. When learners perceived NF as corrective they would be more likely to notice the gap between their utterances and the target structure, and subsequently they could modify them if provided with opportunities to do modification.On the basis of the present findings, the dissertation concludes that NF, MO, learners'noticing of the target form and perceptions about NF are interrelated with each other in the interactional processes directly or indirectly to facilitate EFL development. Despite some limitations, the present study may provide us with a better understanding of the interaction-learning relationship in the Chinese EFL context, help us gain insight into the theories related with input, output, attention, and perception, and provide some pedagogical implications for how to manipulate focus-on-form (FonF) instruction in Chinese EFL classroom.