Dr Manesha Peiris, Senior Lecturer in Reflective Practice and Project Management at Queen Mary University of London, examines AI-supported reflective practice.
Much of the conversation around GenAI in higher education has focused on assessment integrity, authorship and AI-resilient assessment design. But what if we are overlooking another opportunity? As GenAI becomes increasingly accessible, higher education institutions have sought to understand how assessment practices can maintain academic standards while recognising the realities of student engagement with these technologies.
Recent studies, such as Stephenson and Armstrong (2026), suggest that learners are already incorporating GenAI into various stages of the learning process. While discussions surrounding authenticity and academic integrity remain important, an exclusive focus on these concerns risks overlooking the pedagogical opportunities that GenAI may offer. In this article I explore one such opportunity: the use of GenAI to scaffold affective development through reflective practice.
As educators, we often talk about preparing students for employment, professional practice and life beyond university. Yet many of the attributes we value most like self-awareness, resilience, professional identity and values sit within the affective domain. Even though we see affective attributes referenced in graduate attributes, employability frameworks and degree apprenticeship standards, affective development often receives less attention compared to its cognitive counterpart in higher education curricula.
One reason for this imbalance lies in the challenge of assessment. Cognitive development can often be evidenced through examinations, coursework and other observable demonstrations of knowledge and skill. Affective development, however, is internalised, personal, and context-dependent, making it more difficult to observe directly. Peiris (2026) shows that assessing affective development requires educators to reconsider traditional approaches to assessment and evidence. In this regard, reflective practice offers a possible solution, providing learners with opportunities to articulate and demonstrate changes in their attitudes, values, professional identities and self-awareness. Through reflection, learners can make visible aspects of development that might otherwise remain tacit.
Scaffolding affective development with GenAI
Operationalising GenAI as a scaffold for affective development requires educators to move beyond simply introducing AI tools into learning activities. Instead, we need to carefully consider how effective learning outcomes are embedded within the curriculum. Krathwohl’s affective taxonomy provides a useful framework in this regard, encouraging educators to consider how attributes such as receiving, responding, valuing and organising can be intentionally cultivated through learning and assessment activities. When doing this, educators need to reflect on how affective attributes complement the knowledge, skills and professional dispositions associated with a particular subject area. Once these relationships have been established, thoughtfully designed prompts can provide learners with a structured framework through which GenAI can support reflective dialogue and exploration of experience.
At the same time, the use of GenAI requires appropriate pedagogical safeguards to be put in place. Educators need to be mindful of risks, like hallucinations, over-reliance on AI-generated interpretations, and the potential for learners to disengage from the authentic meaning-making processes. Consequently, clear guidance and carefully designed guardrails are necessary to ensure that GenAI enhances rather than replaces reflective practice.
A practical example
One approach is to align reflective prompts with the stages of Krathwohl’s affective taxonomy, encouraging learners to progressively explore and articulate their developing attitudes, values and professional identities. As Jackson (2025) explains, the way questions are framed can be considered a form of prompt engineering, as simplistic prompts are more likely to generate superficial responses. This has important implications for educators seeking to scaffold affective development through reflective practice. If GenAI is to support learners in examining assumptions, interrogating biases and making sense of complex experiences, educators must carefully consider how prompts are designed and what forms of reflection they are intended to elicit. In this context, prompt engineering becomes a pedagogical design activity rather than merely a technical skill. In practice, I found myself asking a simple question: what would happen if students used GenAI not to write reflections, but to support the reflective process itself? To explore this question, I redesigned a reflective activity within a Level 4 Design Thinking module.
Figures 1 and 2 illustrate how GenAI was positioned as a reflective dialogue partner within a Level 4 Design Thinking module. Figure 1 presents the guidance provided to learners, combining Driscoll’s reflective model with Krathwohl’s affective taxonomy. Rather than instructing AI to generate a reflection, the prompt encourages the system to facilitate a structured conversation focused on affective development, assumptions, biases, values and future action.
Figure 2 demonstrates the resulting interaction. Rather than producing a completed reflection, GenAI begins by situating the learner within the “What?” stage of Driscoll’s model and the “Receiving” level of the affective taxonomy. The learner is then presented with a single open-ended question designed to encourage attention, awareness and initial sense-making. Subsequent questions emerge in response to the learner’s answers, creating an iterative and dialogic process of reflection.
What struck me most was that the value wasn’t in the AI producing a reflection. The value emerged through the questions it asked. By prompting students to unpack assumptions, examine values and consider future action, the AI became a facilitator of reflection rather than a generator of content
Final thoughts
While concerns surrounding assessment integrity, authorship and AI misuse will undoubtedly remain important in academia, they should not prevent educators from exploring the pedagogical opportunities that GenAI presents. As this blogpost has argued, affective development remains one of the most valuable yet difficult dimensions of learning to evidence within higher education. By combining established reflective frameworks with carefully designed AI-mediated dialogue, educators may be able to create new opportunities for learners to articulate changes in their attitudes, values, assumptions and professional identities. In this sense, GenAI is not positioned as a substitute for reflection or critical thinking, but as a scaffold that can support learners in making their affective development more visible. As higher education continues to focus on making assessments AI-resilient, are we overlooking opportunities to use AI to develop the very human attributes that graduate outcomes increasingly demand?
Dr Manesha Peiris is Senior Lecturer in Reflective Practice and Project Management and Director of Student Experience at Queen Mary University of London. Her work explores feminist pedagogy, affective learning, reflective practice, and educational technology, with a focus on inclusive, student-centred and transformative higher education.
References
Jackson, J. (2025). Higher order prompting: Applying Bloom’s revised taxonomy to the use of large language models in higher education. Studies in Technology Enhanced Learning, 4(1). https://doi.org/10.21428/8c225f6e.0915c17e
Peiris, M. (2026). Integrating SDGs into curriculum: Enhancing the affective domain through education. Amps 2025 Prague – Education Research & Teaching Conference Exploring Academia – From Practice to Publishing. Education Research & Teaching Conference. https://amps-research.com/wp-content/uploads/2026/02/Amps-Proceedings-Series_43.3.pdf
Stephenson, R., & Armstrong, C. (2026). Student Generative Artificial Intelligence Survey 2026 (No. 199). HEPI.
