UK case studies examine learner agency in student–AI interaction

Updated Sep 07, 2026

Xianghan O’Dea, Yuan-Li Tiffany Chiu, Richard Bale and Monika Rossiter published Exploring student-AI interaction in assessment through the lens of learner agency: case studies in UK higher education in Studies in Higher Education, online on 17 July 2026. Paper and DOI.

Evidence available for this briefing: publication identity and the lead author’s public account were accessible. The complete paper was not. This is a limited briefing, not a full-paper verification.

The author’s account

O’Dea describes qualitative focus-group research in two UK institutions, interpreted through self-regulated learning. Her account argues for helping students become active decision-makers in their interactions with GenAI, with attention to planning, monitoring and critical engagement.

It advocates curriculum and teaching approaches that support those capabilities. That argument is not evidence that a particular intervention improves learning, nor does the accessible account establish how common different uses are across UK students. Lead author’s account.

Questions for local student feedback

Our practical suggestion is to ask students what they used AI for and how they checked the result. A report of faster work and an account of evaluating competing explanations describe different experiences; neither should be classified from a usage tick-box alone.

Ask where human support remained necessary and what students understood about the assessment rules. Keep the task and course context alongside comments rather than assuming the same tool use has the same meaning everywhere.

Survey comments can inform questions for teaching teams, but cannot demonstrate a student’s learning depth or establish an intervention’s effect. Any claim of improved performance needs suitable assessment evidence beyond self-reported experience.

The governance checklist can help plan a bounded review of comments about AI and assessment.

Review note, 7 September 2026: replaced unverified sample counts, quotations and claims about dominant patterns with the accessible author account; revised the headline and removed product claims about diagnosing learning quality. Publication date and URL are preserved.

Reference

Xianghan O’Dea, Yuan-Li Tiffany Chiu, Richard Bale and Monika Rossiter (2026). Exploring student-AI interaction in assessment through the lens of learner agency: case studies in UK higher education. Studies in Higher Education. DOI: 10.1080/03075079.2026.2700435.