Effects of AI-Driven Written Direct and Indirect Feedback on Iranian Intermediate EFL Learners' Writing Complexity

This study examined the impact of AI-driven feedback on the writing complexity of Iranian intermediate EFL learners using a quasi-experimental design. Through convenience sampling, ۱۰۰ participants (male and female, aged ۱۸-۲۵) from two language institutes in Tehran were divided into four groups: two experimental groups receiving direct and indirect feedback from AI ChatGPT, and two control groups receiving the same feedback types from their teacher. Participants completed a pre-test, ten writing tasks over ۱۴ weeks, and a post-test. Results, analyzed via descriptive statistics and one-way ANOVA, indicated notable improvements in writing complexity across all groups. The AI direct feedback group showed the highest improvement, with a mean difference of ۵.۲۴ (p < ۰.۰۵), followed by the teacher direct feedback group, which also demonstrated significant gains. The AI indirect feedback group exhibited moderate progress, while the teacher indirect feedback group showed the least improvement. Analysis of syntactic measures revealed that AI feedback, particularly direct feedback, effectively enhanced sentence structures and encouraged the use of more sophisticated vocabulary. These findings highlight AI-driven feedback’s potential to enhance EFL learners’ writing complexity, with direct feedback yielding the greatest benefits.

کلیدواژه ها:

AI-driven feedback ، Writing complexity ، direct feedback ، indirect feedback ، Syntactic measures