| 125 | 0 | 46 |
| 下载次数 | 被引频次 | 阅读次数 |
随着生成式人工智能(GAI)的迅速发展,其在课堂教学中的应用正逐渐重塑教学方式与生态。然而,现有研究多聚焦于GAI的输出质量,而对其与人类教师在对话互动模式上的系统比较研究仍显不足。该研究基于IRF(Initiation,Response, Follow-up)对话理论框架,结合有向网络分析方法,对66名大学生在课堂教学任务中与教师和生成式人工智能导师的2016条有效在线对话数据进行比较分析。研究发现,人机交互呈现出“单线回应型”对话模式,其特征为学生提问后GAI直接回应,虽能拓展学生思维广度并激发提问意愿,但因缺乏后续延展与协商,互动深度有限,易导致学生依赖GAI而削弱独立思考能力;而师生对话则呈现“协商共建型”模式,包含单步交流与多轮追问回应,结构更为复杂,教师通过启发式提问与递进反馈引导学生聚焦逻辑与表达策略,促进高阶思维发展。未来研究可通过教育应用开发者与教师的协同设计与实践,将GAI支持下的人机交互广度与师生对话的深度引导相结合,实现优势互补,推动对话式学习在课堂教学中的有效开展,为课堂教学新形态的构建提供理论与实践支持。
Abstract：With the rapid development of Generative Artificial Intelligence(GAI), its application in classroom teaching is gradually reshaping the ways and ecology of teaching. However, existing research mostly focuses on the quality of GAI's output, while systematic comparative studies on the interactive dialogue patterns between GAI and human teachers are still insufficient. Based on the IRF(Initiation, Response, Follow-up) dialogue theory framework and combined with directed network analysis, this study comparatively analyzed 2,016 valid online dialogue data between 66 college students and teachers as well as ChatGPT in classroom teaching tasks. The findings indicate that human-machine interaction exhibits a “unilinear response” dialogue pattern, characterized by GAI directly responding to students' questions. Although it can expand students' breadth of thinking and stimulate their willingness to ask questions, the interaction depth is limited due to the lack of follow-up extension and negotiation, and students are more likely to rely on GAI, which weakens their independent thinking. In contrast, teacher-student dialogue presents a “negotiation and co-construction” pattern, which includes single-step communication and multi-turn questioning and responding, with a more complex structure. Teachers guide students to focus on logic and expression strategies through heuristic questioning and progressive feedback, thereby promoting higher-order thinking. Future research can combine the interaction breadth supported by GAI and the in-depth guidance of teacher-student dialogue through collaborative design and practice between educational application developers and teachers, to achieve complementary advantages and promote the effective implementation of dialogic learning in classroom teaching, providing theoretical and practical support for the construction of new forms of classroom teaching.
[1]Wang X,Liu Q,et al.What matters in AI-supported learning:A study of humanAI interactions in language learning using cluster analysis and epistemic network analysis[J].Computers&Education,2023,194:104703.
[2]郭颢,范佳荣等.生成式人工智能与教育的共生互构:人文守护下的实践进路[J].中国电化教育,2025,(8):75-80.
[3]蒋慧芳,曾文婕等.指向深度学习的对话式反馈:特征、方式与策略[J].开放教育研究,2024,30(2):80-88.
[4]李森,吉标.师生对话的特点及意义[J].西南师范大学学报(人文社会科学版),2004,(3):88-91.
[5]苏春景,孙晓莎.小学课堂教学中的师生“病态对话”现象及其消解策略[J].中国特殊教育,2011,(11):26-31.
[6]Vandenbroucke L,Spilt J,et al.The classroom as a developmental context for cognitive development:A meta-analysis on the importance of teacher-student interactions for children’s executive functions[J]. Review of Educational Research,2018,88(1):125-164.
[7]孔企平.教学策略的转型:从独白走向对话[J].全球教育展望,2006,35(6):42-45.
[8]郭冰.课堂中教育性对话的缺失与建构[J].中国教育学刊,2013,(1):75-78.
[9]Doyle N B, Downer J T, et al. Understanding high quality teacher-student interactions in high needs elementary schools:An exploration of teacher,student,and relational contributors[J].School Mental Health,2022,14(4):997-1010.
[10]王文洁,凌敏.高中地理优质课中的师生对话研究——以“大规模的海水运动”为例[J].地理教学,2020,(8):18-21.
[11]Pehmer A,Groschner A,et al.How teacher professional development regarding classroom dialogue affects students’higher-order learning[J].Teaching and Teacher Education,2015,47:108-119.
[12]郑太年,胡华.社会情感学习中的活动设计和师生对话:基于课堂视频案例的分析[J].西北师大学报(社会科学版),2022,59(2):20-28.
[13]徐升,佟佳睿等.下一代个性化学习:生成式人工智能增强智能辅导系统[J].开放教育研究,2024,30(2):13-22.
[14]冷静,卢弘焕等.生成式人工智能赋能批判性思维测评——基于Chat GPT的应用实验[J].现代远程教育研究,2024,36(6):102-111.
[15]Steiss J,Tate T,et al.Comparing the quality of human and Chat GPT feedback of students’writing[J].Learning and Instruction,2024,91:101894.
[16]Guo K,Wang D.To resist it or to embrace it?Examining Chat GPT’s potential to support teacher feedback in EFL writing[J].Education and Information Technologies,2024,29(7):8435-8463.
[17]Cazden C B.Classroom discourse:The language of teaching and learning(2th ed.)[M].New York:Heinemann,2001.
[18]Wells G.Reevaluating the IRF sequence:A proposal for the articulation of theories of activity and discourse for the analysis of teaching and learning in the classroom[J].Linguistics and Education,1993,5(1):1-37.
[19]O’Connor C, Michaels S.Aligning academic task and participation status through revoicing:Analysis of classroom discourse strategy[J].Anthropology&Education Quarterly,1993,24(4):318-335.
[20]张娟娟,陈旭远等.促进深度学习的师生对话特征:基于会话分析理论的探索[J].教育科学研究,2021,(10):82-89.
[21]Doley S K.Discourse patterns in English as an L2 classroom in India[J].International Journal of Instruction,2019,12(1):1223-1238.
[22]Wang H,Liu Y.The Discourse of English reading lessons in Chinese middle schools:A sociocultural perspective[J].Chinese Journal of Applied Linguistics,2012,35(2):212-228.
[23]Estaji M,Mirzaei Shojakhanlou M.Realization of initiation,response,and feedback in teacher-student interactions in EFL classrooms:Learning realities and opportunities[J].Journal of English Language Teaching and Learning,2022,14(30):91-114.
[24]Sinclair J M,Coulthard R M.Towards an analysis of discourse:The English used by teachers and pupils[M].Oxtord:Oxford University Press,1975.
[25]吴志华,周喜欢.基于IRF话语分析理论的课堂对话教学有效性分析[J].中国教育学刊,2015,(3):71-74.
[26]Fan Y,Tang L,et al.Beware of metacognitive laziness:Effects of generative artificial intelligence on learning motivation,processes,and performance[J].British Journal of Educational Technology,2025,56(2):489-530.
[27]冷静,曹凌琳等.面向深层语言特征的智能评估系统构建与应用验证[J].远程教育杂志,2025,43(6):91-100.
[28]Pehmer A,Groschner A,et al.How teacher professional development regarding classroom dialogue affects students’higher-order learning[J].Teaching and Teacher Education,2015,47:108-119.
[29]Hwang M,Jeens R,et al.Exploring learner prompting behavior and its effect on Chat GPT-assisted English writing revision[J]. Asia-Pacific Education Researcher,2025,34(3):1157-1167.
[30]王一岩,刘淇等.对话式人机协同学习:本质内涵与未来图景[J].中国电化教育,2024,(11):21-27.
[31]左明章,张尧等.专递课堂教师言语交互策略对学习动机的影响研究——以湖北省崇阳县专递课堂为例[J].中国电化教育,2023,(6):98-105+134.
[32]Lee D,Palmer E.Prompt engineering in higher education:A systematic review to help inform curricula[J].International Journal of Educational Technology in Higher Education,2025,22(1):1-22.
[33]Luo M,Hu X,et al.The collaboration of AI and teacher in feedback provision and its impact on EFL learner’ s argumentative writing[J].Education and Information Technologies,2025,30(12):17695-17715.
基本信息:
中图分类号:G434
引用信息:
[1]冷静,王章涵,李思源,等.从单线回应到协商共建:基于IRF理论的人机与师生对话分析[J].中国电化教育,2026,No.473(06):46-54.
基金信息:
中央高校基本科研业务费项目华东师范大学哲学社会科学创新团队项目(项目编号:2024QKT004);中央高校基本科研业务费项目青年预研究项目“基于多模态大模型的批判性思维测评研究”(项目编号:2024ECNU-YYJ026)研究成果~~
2026-06-04
2026-06-04