Technology-enhanced formative assessments (TEFAs) are widely promoted as mechanisms for improving student engagement, feedback use, and learning in higher education. However, educators often lack practice-facing guidance to support informed design decisions beyond tool selection or generic pedagogical recommendations. Theoretical frameworks can play an important role in helping educators conceptualise and plan pedagogical interventions, including those involving digital technologies. This paper presents the development of a practice-facing theoretical framework, integrating Learning-Oriented Assessment (LOA) (Carless, 2007; Carless et al., 2011) and the Technological Pedagogical Content Knowledge (TPACK) framework (Mishra and Koehler, 2006), to support educator decision-making in the design and implementation of TEFAs.

The framework emerged through a participatory action research study conducted across three iterative cycles within the author’s own teaching practice. Drawing on established action research traditions (Elliott, 1991; Coughlan and Coghlan, 2002), the study examined how theoretical assumptions about formative assessment, feedback, learner agency, and technology were enacted and contested in practice under real institutional conditions. Quantitative engagement data and qualitative student feedback informed cycle-by-cycle refinements to both pedagogical design and the evolving framework. The research formed part of the author’s doctoral study, which explored the role of technology-enhanced formative assessment in supporting student engagement and learning in higher education. Across the cycles, the research revealed persistent tensions between formative intentions and summative assessment structures, uneven patterns of student engagement, and the need for educators to balance pedagogical aims with institutional and technological constraints.

Rather than functioning as a technical or prescriptive model, the resulting LOA–TPACK framework is presented as a decision-making lens that highlights pedagogical intent, learner experience, and feedback processes, while treating technology as an enabling but secondary condition. The framework is explicitly informed by a state of the actual (Selwyn, 2008) account of practice, acknowledging persistent summative assessment dominance, uneven learner engagement, strategic student behaviour, and practical constraints on time and workload.

The paper argues that the value of the framework lies not in claims of optimisation or transformation, but in its capacity to support reflective professional judgement in complex educational contexts. While the framework contributes to the theoretical understandings of technology-enhanced formative assessment, it also offers a practical structure that educators can use to plan and reflect on pedagogical interventions involving digital technologies. By making visible the tensions between formative ideals and institutional realities, the framework contributes a grounded, human-centred approach to TEFA design that aligns with contemporary calls to shift EdTech research from questions of how technologies are implemented to who they meaningfully serve.

Challenges with the teaching, and learning, of Research Methods have been widely documented, including content complexity and abstractness, staff and student diversity in knowledge and experience, emotional and motivational aspects, and bridging methods and practice (Lundahl, 2008). In tandem, research suggests the need for more technological advancement into Research Methods teaching (Mitchell and Rich, 2020), and opportunities exist for AI tools to bridge a gap in this space. However, there is limited research into the perceptions of students and teaching regarding the use of AI tools in teaching and learning Research Methods.

This research explored perceptions about these tools from the perspective of staff and students to inform future development of AI tools for student support in the learning of Research Methods. Fifteen staff and students from an Arts, Humanities and Social Science background were interviewed using a semi-structured approach. Data outputs were analysed via qualitative inductive thematic analysis by two researchers.

Results from student interviews suggested conflicting negative and positive perceptions. Students were concerned about the accuracy and legitimacy of AI tools for research and described the need for cautiousness, critical thinking and human input in their use. They also emphasised the need to learn research skills themselves rather than outsourcing cognitive development to tools. Conversely, students described AI’s usefulness for identifying literature, summarising knowledge, saving time, and exploratory thinking.

Results from staff interviews revealed similar views regarding potential AI over-reliance impoverishing student learning of foundational research skills and practices. As with students, staff were concerned about the accuracy and legitimacy of AI in research, with several mentioning that ill-prepared students miss the superficial nature of AI outputs or take them at face value, and that this concern has already begun to manifest in the quality of some assessment submissions received. Time saving and efficiency were recurring themes that staff considered to be beneficial aspects of AI usage. Overall, staff perceptions of the use of AI in research were largely bimodal, but cautiously optimistic where AI outputs can be guard-railed or used as prompts for reflection, developing critical thinking or helping students speak confidently about research.

Overlapping thematic areas across student and staff centred on three shared concerns: trust in reliability and quality of AI-generated content; the risk of over-reliance or short-cuts preventing students developing foundational research competencies; and the importance of human-in-the-loop oversight to evaluate AI outputs. However, both groups noted practical benefits in efficiency, literature exploration and orientation, and as an ideation support when practicing new methods. Together, these findings indicate a tentative acceptance of AI in research contexts when framed as a cognitive support within critical practice, but such technologies are seen as problematic when positioned as a substitute for disciplinary skill development, with one staff member noting that students can’t bring their toaster (i.e. technology) to a Viva.

Any design of an AI tool for Research Methods learning should consider these concerns and integrate aligned pedagogical approaches (Constructivism, Cognitive Load Theory, Experiential Learning, Self-Efficacy, Dialogic Pedagogy) to scaffold a meaningful learning experience.

This research contributes a novel understanding of the perceptions of AI tools for Research Methods learning and provides a preliminary scaffold for the development of AI tools in this space.

References

Lundahl, B. W. (2008). Teaching research methodology through active learning. Journal of Teaching in Social Work, 28(1-2), 273-288. doi:10.1080/08841230802179373

Mitchell, A., & Rich, M. (2020). Business School Teaching of Research Methods–A Review of Literature and Initial Data Collection for Undergraduate Business School Students. Electronic Journal of Business Research Methods, 18(2), pp100‑114-pp100‑114.

Since the release of ChatGPT in November 2022, discussions about Generative Artificial Intelligence (GenAI) have dominated the discourse across Higher Education both nationally and internationally. The introduction of any new technology into the teaching space is often a catalyst for substantial changes in practice (McDonald et al., 2025). In the case of Gen AI, much effort has been expended in working to understand the pedagogical implications of its use from various perspectives, including impacts on academic integrity, assessment design, and the need for critical AI literacy for both staff and students.

Pedagogical decision making is challenging given the need to be cognisant of, and to balance the benefits and costs of GenAI use. Positive impacts on students’ creativity and self-efficacy (Wang, Sun, & Chen, 2023), and the potential for personalised learning experiences (Xu et al. 2023) sit (not always comfortably) alongside our still developing knowledge of the potential negative impacts on cognitive skills such as decision making (Heersmink 202).

The ever-growing corpus of research on Gen AI in education spaces has thus focused on impacts on student learning, and, in addition, the ways in which Gen AI is incorporated into curricula. Attention has also been directed towards inherent paradoxes of GenAI use, and the impact of polarising views (Lim et al., 2023). However, as an academic community, the contrasting views, perspectives, and research present a wicked problem. How does an educator accommodate contrasting views on the worth of Gen AI in their own practice, and deal with the inherent cognitive dissonance this may present? If, as the recent HEA National Forum Policy Framework states “it is a set of tools that, regardless of any individual professional or personal perspective, must be integrated thoughtfully into teaching and learning …..” (O’Sullivan et al., 2026, pg3), how is this understood and negotiated by educators in their day-to-day practice?

This study investigates how academics in an Irish HEI narrate their navigation of any perceived tensions in their use of GenAI in teaching, learning and assessment. Adopting a qualitative approach the study utilises the Rich Picture method (Bell, Berg, & Morse, 2016). As the method involves participants creating visual and symbolic representations of their lived experiences, it captures conceptual and emotional aspects relevant to the individual's experience in ways which may uncover the messy complexities inherent in the topic of study. Initial results will be presented focusing on academics from a range of disciplines and levels of experience

Bell, S., Berg, T., & Morse, S. (2016). Rich Pictures: Encouraging Resilient Communities (1st ed.). Routledge. https://doi.org/10.4324/9781315708393

Heersmink, R. (2024). Use of large language models might affect our cognitive skills. Nature Human Behaviour, 8, 805–806. https://doi.org/10.1038/s41562-024-01859-y

Lim, W. M., Gunasekara, A., Pallant, J. L., Pallant, J. I., & Pechenkina, E. (2023). Generative AI and the future of education: Ragnarök or reformation? A paradoxical perspective from management educators. The International Journal of Management Education, 21(2) https://doi.org/10.1016/j.ijme.2023.100790

McDonald, N., Johri, A., Ali, A., & Hingle, A. (2024). Generative artificial intelligence in higher education: Evidence from an analysis of institutional policies and guidelines. arXiv. https://doi.org/10.48550/arXiv.2402.01659

O’Sullivan, J., Lowry, C., Woods, R., & Conlon, T. (2025). Generative AI in higher education teaching & learning: Policy framework. Higher Education Authority. https://doi.org/10.82110/073e-hg66

Wang, S., Sun, Z., & Chen, Y. (2023). Effects of higher education institutes’ artificial intelligence capability on students' self-efficacy, creativity and learning performance. Education and Information Technologies, 28(5), 4919–4939. https://doi.org/10.1007/s10639-022-11443-4

Xu, W., Meng, J., Raja, S. K. S., Priya, M. P., & Kiruthiga Devi, M. (2023). Artificial intelligence in constructing personalized and accurate feedback systems for students. International Journal of Modeling, Simulation, and Scientific Computing, 14(01), https://doi.org/10.1142/S1793962323410015