Rethinking reflective writing practices: From tried and tested tools to the challenges and opportunities of AI

Reflexivity is a significant part of the doctoral experience, maximising the learning potential of the write-revise-reflect cycle (Boud et al. 1985). This premise is a key feature of our writing for publication course for multidisciplinary, multilingual doctoral students. This presentation firstly examines the evaluation and subsequent adaptation of three writing reflection tools used on this course. Our findings from this iterative process underscore the importance of designing a reflective tool that not only complements course objectives but also addresses students’ needs. The second part of the talk aims to situate our results within the AI landscape. We seek to engage the community in exploring how AI-driven tools might enhance or hinder students’ engagement with writing reflection. How could AI support the reflective process and students’ needs? Should we proactively adapt reflective practice to incorporate AI for everyone? And can AI be integrated into reflective writing without compromising the writer’s agency? This paper briefly introduces our 10-week, flipped research writing course before outlining the evolution of the reflective tools employed from reflective journals (Boud, 2001) to in-text comments with a reflective paragraph (Yayli, 2012) to narrative frames (Wette & Furneaux, 2018). We will explain why we felt a change in reflexive practice was required by using Experiential Learning Theory (ELT) as the analytical framework. We evaluated each reflective tool based on the frequency and quality of reflection-in-action; reflection-on-action; and reflection-for-action (Schön 1992) found in the reflections. Our findings suggest that separating reflection from the text in journals can be problematic and whilst reflection-for-action was evident, reflection within and upon the text was limited. The second iteration, in-text comments, demonstrated evidence of reflection-in-action. However, students struggled to use this method for reflection-for action. Our third tool, narrative frames, facilitated reflection-on-action through exemplifying and commenting on text revisions and promoted reflection-for action. Whilst we believe this tool marries our objective of promoting writer’s understanding of their own process, we are now faced with considering how foolproof this strategy is against the use of AI or whether we should accept the use of AI in the write-revise-reflect cycle and adapt accordingly.