華語線上互動中糾錯回饋與二語習得研究

本論文旨在探討華語學習者與華語實習教師在線上互動過程中，糾錯回饋(corrective feedback)對二語習得的關係。本研究從兩個面向討論此問題，一、糾錯回饋如何影響學習者對於目標語言形式的注意(noticing)與修復(repair)，二、周邊語言特色(paralinguistic features)，如，語調、手勢、臉部表情等，如何影響學習者與教師雙方互動，與糾錯回饋的關係為何。過去的互動研究場景多為實驗室或課堂教學，為了提供多元研究資料，本論文研究題材為捷克馬薩里克大學的華語學習者與國立臺灣師範大學華語文教學研究所碩士生的線上互動課程。研究者將線上互動課程影片轉寫為逐字稿，並應用言談分析檢視教學內容。言談分析打破傳統文本分類框架，同時記錄互動中口語及非口語訊息，適用於線上互動文本分析，符合研究目的。本論文採用Lyster 及Ranta 2007年研究中六種糾錯回饋類型，直接糾錯、重述、澄清性提問、後設語言回饋、引出以及重複學生錯誤。研究結果顯示，線上互動中，糾錯回饋類型以重述及澄清性提問為多數。以言談分析方法觀察互動後發現，華語實習教師提供糾錯回饋時，通常搭配語調、手勢、臉部表情等周邊語言特徵，引起學習者對目標語言形式的注意與修復。同樣地，學習者亦透過手勢、臉部表情傳達溝通意願，提問或者與教師持續互動。此外，從逐字稿分析中觀察到，當學習者注意到糾錯回饋的當下，即使沒立即修復，卻能在後續的話輪中，產出接近或完全正確的目標語言形式，顯示持續互動的延遲效果(delayed effects)。

關鍵字

互動假說；糾錯回饋；周邊語言特徵；注意力假說；二語學習

並列摘要

The aim of this research is to examine the relationship between corrective feedback and second language acquisition in online interaction between Chinese learners and teachers. From the two perspectives, this study investigates (1) how corrective feedback affects second language learning, and (2) if paralinguistic features, (e.g. intonation, hand gestures, facial expressions) have an influence on the teacher-learner interaction. Interactive studies have been usually conducted in laboratories or classrooms. Different from the past, the data of the current study were online interaction between Chinese learners at Masaryk University in Czech Republic and graduate students in graduate institute of Chinese as a Second Language in National Taiwan Normal University. The researcher transcribed the videotapes of the online interactions with conversational analysis. In terms of conversational analysis, verbal and nonverbal elements are recorded, presenting details of interactional moves between learners and teachers. The six types of corrective feedback adopted in the current study are explicit correction, recast, clarification request, metalinguistic feedback, elicitation, and repetition (Lyster and Ranta, 2007). The results suggested that recasts and clarification request were the two most frequently used types of corrective feedback. The data also showed that teachers employed paralinguistic features with provision of corrective feedback to prompt learners to notice the feedback and produce modification. Meanwhile, learners’ responses involved with paralinguistic features assisted themselves to reformulate, clarify themselves, or make a request. One more finding presented that although learners did not immediately modify their utterance while noticing the feedback, they yielded modification in subsequent turns, thus providing evidence of delayed effects resulting from interaction.