Adopting GenAI-Supported Portfolio Assessment to Promote Self-Regulated Learning: A Focus on Student Feedback Literacy

While studies have investigated the affordances of GenAI in providing personalised feedback and empowering multimodal composing, there remains a lacuna on how GenAI promotes self-regulated learning when integrated into formative assessment. This study aims to examine how GenAI-supported portfolio assessment can foster students’ self-regulated learning, with a particular focus on student feedback literacy development. The research employed a qualitative design over a 16-week semester with 50 first-year students in a Chinese university. They formed groups where they collaboratively created WordPress-based portfolios and posted multimodal writing about recommendations for domestic travel itineraries. Throughout portfolio compilation, they drafted their writing in Word documents and then received feedback from peers, the teacher, and GenAI, where they were asked to document their feedback provision and revision processes by enabling the “track changes” function, highlighting edits informed by AI-generated feedback, and preserving AI responses. Data were collected from students’ writing samples, focus group interviews and narrative inquiries. Findings reveal a gradual developmental trajectory of students’ self-regulated learning supported by GenAI, from uncritical acceptance of AI feedback to prompting refinement for more desirable responses and development of critical evaluation of AI feedback. Students also demonstrate augmented appreciation and engagement with multisource feedback in portfolio compilation, where they compared and reflected on different feedback sources and actively took action accordingly. These findings provide a deeper understanding of how and why students become feedback literate after GenAI-supported portfolio assessment and lay a foundation for future research to measure the exact impact of AI-based portfolio assessment on learning gains.