This study examines the design and implementation of an AI-enhanced Assessment for Learning (AfL) model to support analytical thinking and self-regulated learning in higher education. Adopting a design-based research (DBR) approach, the study emphasizes iterative development, implementation, and refinement within authentic classroom contexts. The model was implemented as part of regular instruction in a university course during the second semester of the 2025 academic year, involving 120 undergraduate students across four intact groups. The instructional design integrated core AfL practices-such as feedback, self-assessment, and peer interactionwith AI-supported feedback systems and learning analytics. Data were collected using a mixed-methods approach, including learning outcome measures, student perception surveys, classroom observations, and reflective data. The findings indicate that the model was feasible to implement in real classroom settings and was associated with high levels of student engagement, active use of feedback, and the development of analytical thinking and self-regulated learning processes. This study contributes to the field by providing a theoretically grounded and practice-oriented model for integrating artificial intelligence into formative assessment. The findings highlight how AI can enhance continuous feedback processes and support adaptive learning in higher education environments.
MAHAPOONYANONT et al. (Thu,) studied this question.