TITLE:
Phase-Specific Integration of Generative AI into the 5E Instructional Model: A Design-Based Research Study in Pre-Service English Language Teacher Education
AUTHORS:
Yanjing Wu
KEYWORDS:
Generative Artificial Intelligence, 5E Instructional Model, English Language Teacher Education, Design-Based Research, Self-Determination Theory
JOURNAL NAME:
Open Access Library Journal,
Vol.13 No.10,
October
21,
2026
ABSTRACT: Generative artificial intelligence (GenAI) is reshaping instructional design in teacher education, yet existing research predominantly treats AI as a generic productivity tool rather than examining its pedagogical function at specific phases of established instructional models. This study addresses that gap by developing, implementing, and refining a phase-specific GenAI integration framework within the 5E instructional model (Engage, Explore, Explain, Elaborate, Evaluate) for pre-service English language teachers. Grounded in constructivist learning theory and self-determination theory (SDT), the framework specifies five distinct AI functions: cognitive activator (Engage), scaffold builder (Explore), critical evaluation object (Explain), creative collaborator (Elaborate), and differentiated assessment support (Evaluate). A two-cycle Design-Based Research (DBR) design was employed with 32 pre-service English language teachers across a 12-week pedagogy course. Data sources included 64 lesson plans (one lesson plan per participant per cycle; 32 in Cycle 1 and 32 in Cycle 2), 380 GenAI interaction logs, 384 reflective journal entries, and 12 semi-structured interviews. Analysis combined deductive content analysis (Cohen’s κ = 0.84) with reflexive thematic analysis. Findings show that phase-specific AI integration was associated with more frequent and more critical modification of AI-generated content, and with qualitative indicators of pedagogical reasoning, particularly in designing inquiry-oriented language activities. Because the study did not include an independent pre/post measure of pedagogical reasoning, these findings should not be interpreted as establishing a causal enhancement of reasoning. Four design principles were proposed: pedagogical goal alignment, productive friction, progressive release (tentative and not directly tested), and transparency/attribution. Three persistent tensions are analyzed: cognitive offloading, prompt construction as a barrier, and the “paradox of assistance” in SDT’s autonomy dimension. The study contributes a theoretically grounded, empirically refined framework and offers actionable design principles for teacher educators integrating GenAI into language pedagogy courses.