Comprehension monitoring in EFL reading: Text coding and AI scaffolding

Abstract

Comprehension monitoring plays an important role in English as a Foreign Language (EFL) reading, particularly at inferential and critical levels, yet it is not consistently supported in instruction. This mixed-methods study examined the effects of two comprehension-monitoring interventions, AI-assisted dialogic scaffolding and Modified Text Coding (MTC), on first-year English majors’ literal, inferential, and critical reading comprehension. Using a 13-week quasi-experimental design, 34 undergraduate EFL learners completed pre- and post-reading comprehension tests, weekly reflections, and focus group interviews. Quantitative analyses showed significant gains for both conditions; however, in this context, the MTC group outperformed the AI-assisted group on inferential, critical, and overall comprehension. Qualitative findings indicated that AI-assisted scaffolding supported meaning clarification, reduced linguistic burden, and facilitated dialogic engagement, whereas MTC fostered more structured, self-regulated monitoring routines and deeper strategic processing. The findings contribute to EFL reading research by clarifying how comprehension-monitoring instruction supports higher-order reading processes in EFL contexts.