Exploring learners’ engagement with GenAI-generated feedback in EFL writing: a Chinese middle school case study

Feedback plays a pivotal role in English as a Foreign Language (EFL) writing instruction. While generative artificial intelligence (GenAI) shows transformative potential for language learning, existing research remains limited regarding how learners engage with GenAI-generated feedback across proficiency levels. This naturalistic case study examines 15 Chinese middle school students’ multidimensional engagement with ChatGPT-generated feedback. Through systematic analysis of draft revisions and retrospective interviews, three key findings emerged: (1) engagement levels showed strong correlation with language proficiency, (2) mid-proficiency students displayed the highest overall engagement, particularly in behavioral and affective dimensions and (3) high-proficiency learners demonstrated more critical cognitive engagement. The findings suggest that instructor facilitation is essential for providing proficiency-differentiated scaffolding and emphasize the need for structured GenAI training to optimize AI-enhanced EFL writing instruction.