Leveraging generative AI powered rubric-indexed feedback as a formative assessment strategy for enhancing medical English education

The integration of generative artificial intelligence (GenAI) into language education offers transformative potential, particularly in specialized fields such as medical English. This study examines the impact of a hybrid feedback model—combining GenAI-powered feedback from Dou Bao, a GenAI chatbot, with peer feedback—on medical students’ English writing proficiency. Conducted at a medical university in Southeast China, the study employed a mixed-methods approach involving 116 medical students divided into two experimental groups and one control group, along with five language teachers. Pre- and post-test writing assessments measured improvements in students’ writing, while reflexive thematic analysis and a three-level coding process provided qualitative insights into students’ perceptions and teachers’ experiences. Quantitative findings demonstrated that the hybrid feedback model significantly enhanced students’ writing proficiency in areas of content, structure, and language use. Qualitative findings revealed increased student engagement with medical humanities but also highlighted concerns about over-reliance on GenAI feedback and the emotional detachment associated with its suggestions. Based on these empirical findings, the study proposes the GenAI-powered, Rubric-Indexed Feedback Framework (GRFF)—a scalable model designed to enhance linguistic, professional, and critical thinking competencies among medical students. GRFF’s potential to leverage the affordances of GenAI tools for improving formative assessment practice worldwide underscores its significance for future research and educational innovation.