Artificial Intelligence-Enhanced Repair Strategies in Online Collaborative EFL Classrooms: Toward a New Paradigm of Interactional Competence

This study explored the integration of artificial intelligence (AI) into the identification and enhancement of repair strategies in online collaborative English as a Foreign Language (EFL) classrooms. The aim was to examine how AI tools influence the frequency and diversity of teacher self-repair and their connection to interactional competence. A mixed-methods design was employed, combining qualitative discourse analysis with AI-supported repair detection. The statistical population included three Iranian EFL instructors teaching intermediate and advanced English classes online. The sample, selected through purposive sampling, comprised three female teachers whose teaching interactions were observed for ۷۰ hours. Data were collected using classroom video recordings and analyzed based on the Self-Repair Framework by Fox and Jasperson (۱۹۹۵), alongside AI-supported classification systems. New patterns and types of repairs were identified beyond the traditional framework. The findings revealed that AI-assisted tools successfully detected complex repair patterns and provided teachers with actionable feedback for reflection and improvement. Teachers with higher reflective skills showed greater variation in repair types and used fewer but more deliberate repairs. It was concluded that AI-supported feedback enhances teacher awareness and contributes to the development of interactional competence. The study demonstrated the potential of AI to complement human pedagogical insight and improve communicative effectiveness in online EFL instruction.

کلیدواژه ها:

AI in education ، repair strategies ، online learning ، EFL ، interactional competence ، teacher reflection ، natural language processing (NLP) ، intelligent tutoring systems (ITS)