AI-INTEGRATED RESEARCH-BASED APPROACH FOR ENHANCING PRE-SERVICE EFL TEACHERS’ METHODOLOGICAL COMPETENCE: A LOCALIZED, MULTI-SITE IMPLEMENTATION IN UZBEKISTAN

Abstract

This study examines how an AI-integrated, research-oriented practical inquiry methodologycan strengthen pre-service EFL teachers’ methodological competence in Uzbekistan. The intervention reframed teacher education from the transmission of “ready-made” methodology knowledge to a dynamic system of inquiry, experimentation, evidence-based decision-making, and reflective improvement. A localized EAR cycle (problem identification → exploration → action → reflection → re-action) was enriched with AI analytics and digital learning evidence (e.g., error mapping, learning logs, automated feedback, and participation analytics). The model was implemented across multiple higher education institutions (UzSWLU, Namangan State Institute of Foreign Languages, Samarkand State Institute of Foreign Languages, Denov Institute of Entrepreneurship and Pedagogy, and Kokand State University) through blended instruction combining face-to-face sessions and Zoom-based classes. Data sources included classroom and Zoom recordings, learner written work, diagnostic tests, interviews, rubric-based assessments, e-portfolios, learning analytics dashboards, and AI log files. Findings indicate that AI-supported diagnostics made methodological discrepancies visible and measurable; reflective–empirical inquiry accelerated corrective cycles; and student teachers increasingly demonstrated evidence-based lesson design, differentiated task creation, and reflective practitioner behaviors. The study concludes that localized AI-integrated EAR provides a scalable methodological framework for developing teacher-researcher identity and strengthening methodological competence in varied regional contexts.