The Impact of Gap-gpt AI–Mediated Drawing Tasks on EFL Writing Development: A Quasi‑Experimental Study

This research investigates the impact of using Artificial Intelligence (AI) tools within the Gap-gpt platform, in the form of Painting Tasks, on the development of English as a Foreign Language (EFL) writing skills in Iranian male high school students. The study, employing a quasi-experimental approach, involved ۱۲۰ students (grades ۱۰ and ۱۱; ages ۱۶-۱۷). The sample size was determined using G*Power software, considering a significance level of α = ۰.۰۵ and a medium effect size. Participants, after a pretest, were randomly assigned to two groups: an experimental group (using AI-based painting tasks on Gap-gpt) and a control group (traditional writing instruction). The educational intervention period lasted for three months, with two sessions per week. Pretest and posttest data were analyzed using SPSS version ۲۷. The results of the statistical analyses indicated that the experimental group showed a significant improvement in writing skills compared to the control group (t(۵۹) = ۶.۸۲, p < .۰۰۱, d = .۸۸). Furthermore, multiple regression analysis revealed that the visual scaffolding provided by Gap-gpt’s AI could explain ۴۱% of the variance in posttest performance. These findings underscore the potential of AI tools, particularly in facilitating multimodal learning and enhancing meaning-making processes in writing.

کلیدواژه ها:

Artificial intelligence in education ، Gap-gpt AI ، multimodal writing ، EFL writing development ، drawing tasks