The Impact of AI-Generated Feedback Explicitness (Generic vs. Specific) on EFL Students’ Use of Automated Written Corrective Feedback

This study investigates how AI-driven Automated Written Corrective Feedback can facilitate EFL writing, teaching, and evaluation in Arab schools and universities. The study addresses a significant research gap in the Computer-Assisted Language Learning literature by examining the impact of Automated Essay Scoring systems’ feedback on EFL University students’ writing. AI-powered AWCF tools have the potential to enhance writing accuracy, provide individualized feedback, and reduce teachers’ workload by offering automated yet effective corrective feedback. By analyzing EFL learners’ and instructors’ perceptions, this study contributes to developing user-centered AWE systems that align with the pedagogical needs of L2 writing instruction. The findings will inform educators, software developers, and policymakers on optimizing AI-driven feedback for EFL learners. The Main Research Question is: How does AI-driven Automated Written Corrective Feedback influence EFL university students’ writing proficiency and feedback preferences? A mixed-methods approach was used to conduct a longitudinal controlled experiment at Northern Border University in Saudi Arabia. The study involved questionnaires with 120 students and interviews with 27 students and 4 teachers. An exploratory factor analysis was employed to analyze perceptions of AWCF. The data analyzed the impact of ease of use, clarity, usefulness, EFL learners’ perceptions, and feedback explicitness on AWCF use. The findings proved that EFL students and teachers found AWE systems easy to use and useful for corrective feedback. Furthermore, students preferred using specific AWCF strategies over generic feedback. Developers should consider diverse factors influencing EFL learners’ preferences and views of AWCF to create useful, clear, and desirable tools.