Ai Generated Feedback Versus Teacher Feedback In Efl Writing: A Narrative Review Of Their Complementary Roles
Auteurs : Silini Ibtihel . Tihal Wafia .
Résumé
Feedback is a critical component of EFL writing development, as it serves as a mechanism for improvement. Traditionally, teacher feedback has been the primary source of support, addressing both basic writing issues, such as grammar and mechanics, and higher-order skills, including content development, organization, argumentation, and coherence. However, this approach is often constrained by large class sizes and limited instructional time. The emergence of AI-generated feedback tools offers new opportunities, enabling learners to receive immediate, detailed guidance on their writing performance. Despite growing research on AI and teacher feedback, most studies focus primarily on lower-order skills, leaving higher-order skills underexplored. Furthermore, there is currently no narrative synthesis identifying best practices for combining AI and teacher feedback to support both lower- and higher-order writing development simultaneously. This narrative review synthesizes findings from fifteen peer reviewed recent studies between 2020 and 2026, comparing AI-generated and teacher feedback, highlighting patterns, implications, and limitations. The review indicates that AI and teacher feedback contribute to writing development in complementary ways: AI excels in providing timely, objective, and detailed guidance on grammar, mechanics, and vocabulary, supporting autonomy, while teacher feedback is more effective for higher-order skills and motivational support. Hybrid approaches, which combine AI and teacher feedback, appear to be particularly promising, allowing teachers to focus on complex aspects like organization, argumentation, and content quality while AI addresses routine mechanical issues. Integrating AI effectively requires teachers to guide students in evaluating AI suggestions and reflecting critically on their revisions. Overall, this collaboration can enhance efficiency and maintain the essential depth and nuance of human feedback to develop EFL learners’ writing performance.
Mots clés
AI-generated feedback ; Teacher feedback ; EFL writing ; Hybrid feedback ; Lower- and higher-order skills
