Initiation-Response-Follow-up Pattern of Feedback Interaction between L2 writers and ChatGPT
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
ChatGPT, an AI-powered chatbot, has become increasingly integrated into L2 writing instruction, offering significant benefits by providing feedback through human-like interaction. Within the framework of the Initiation/Response/Follow-up (IRF) classroom interaction model, this study examined how L2 writers interact with ChatGPT to receive feedback for revision. Over a ten-week period, ChatGPT was used as a tool to assist students in revising their take-home writing assignments. The chat history was analyzed to identify prominent structural patterns and thematic categories in the follow-up responses that sustain the interaction. Findings showed that feedback interactions were often initiated by a student’s request for broad feedback on their writing, followed by ChatGPT’s extensive suggestions for revision, and then the student’s follow-up responses to address each suggestion based on their learning interests and needs. A thematic analysis of the follow-up responses identified four primary types: requests for elaboration and additional information, requests for confirmation of language use and grammar, resolution of miscommunications and presentation of counterargument, and expressions of appreciation and satisfaction. Paired t-tests were conducted and found that the interactive feedback led to improvement in L2 writing proficiency. Findings of this study offer implications for the effective integration of ChatGPT into foreign language education.
Abstract
ChatGPT, an AI-powered chatbot, has become increasingly integrated into L2 writing instruction, offering significant benefits by providing feedback through human-like interaction. Within the framework of the Initiation/Response/Follow-up (IRF) classroom interaction model, this study examined how L2 writers interact with ChatGPT to receive feedback for revision. Over a ten-week period, ChatGPT was used as a tool to assist students in revising their take-home writing assignments. The chat history was analyzed to identify prominent structural patterns and thematic categories in the follow-up responses that sustain the interaction. Findings showed that feedback interactions were often initiated by a student’s request for broad feedback on their writing, followed by ChatGPT’s extensive suggestions for revision, and then the student’s follow-up responses to address each suggestion based on their learning interests and needs. A thematic analysis of the follow-up responses identified four primary types: requests for elaboration and additional information, requests for confirmation of language use and grammar, resolution of miscommunications and presentation of counterargument, and expressions of appreciation and satisfaction. Paired t-tests were conducted and found that the interactive feedback led to improvement in L2 writing proficiency. Findings of this study offer implications for the effective integration of ChatGPT into foreign language education.