Self-directed writing development across computer/AI-based tasks: Unraveling the traces on L2 writing outcomes, growth mindfulness, and grammatical knowledge

Abstract

As technology continues to reshape the educational landscape, understanding how autonomous engagement with AI-driven platforms influences second language (L2) acquisition is increasingly important. This study examined the impact of self-directed writing development through computer- and AI-based tasks on L2 writing outcomes, growth mindfulness, and grammatical knowledge. The research was conducted in three private English institutes in Ahvaz, Iran, involving 561 intermediate-level EFL learners that were selected using a convenience sampling method. Using a quasi-experimental design, participants were divided into two groups: an experimental group (EG) of 276 students, which engaged in self-directed AI-based writing tasks, and a control group (CG) of 285 students, which followed traditional writing instruction. Quantitative data were gathered through pre- and post-tests measuring writing proficiency, growth mindfulness, and grammatical accuracy. Additionally, qualitative data were collected through interviews and questionnaires administered to the EG to assess their attitudes toward AI-based learning. The results indicated that the EG significantly outperformed the CG in writing outcomes and grammatical accuracy. Moreover, the EG demonstrated greater growth in mindfulness, with participants showing increased awareness of their writing processes and enhanced self-regulation strategies. Interviews and questionnaires revealed that participants in the EG held positive attitudes toward AI-based tasks, emphasizing increased engagement, autonomy, enjoyment, motivation, personalized learning, critical thinking, self-efficacy, time management, and collaboration in their English learning development. These findings underscore the potential of AI-based tasks in fostering both linguistic competence and metacognitive skills in L2 learners. The study provides valuable insights into the role of AI technologies in promoting independent learning and calls for further exploration of AI-based interventions in writing education.