Adaptive Artificial Intelligence – Supported Formative Feedback and The Development of Self-Regulated Learning in Undergraduate Education

Abstract

The increasing integration of artificial intelligence into higher education has created new opportunities for supporting student learning through personalized and data-driven feedback mechanisms. Among these innovations, artificial intelligence-supported formative assessment systems have attracted growing attention due to their potential to provide immediate, individualized, and continuous feedback that encourages students to take a more active role in managing their learning processes. Despite the growing adoption of these technologies, empirical evidence regarding their influence on learner self-regulation remains limited, particularly in large undergraduate learning environments.This study examines the impact of adaptive artificial intelligence-supported formative feedback on the development of self-regulated learning behaviors among undergraduate students. In this research, artificial intelligence refers to computational systems capable of analyzing learner performance and engagement data to generate personalized feedback, recommend learning activities, and visualize progress toward learning objectives. Such systems are designed not only to support academic performance but also to foster students' abilities to monitor, evaluate, and regulate their own learning processes.The study employs a quasi-experimental research design involving 638 undergraduate students enrolled in introductory business and social sciences courses at a large public university. Participants were divided into two groups. The control group received traditional instructor-provided formative feedback, whereas the experimental group used an AI-supported formative assessment platform integrated into the institutional learning management system. The platform delivered adaptive feedback messages, progress indicators, and personalized recommendations intended to support students' self-monitoring and strategic learning behaviors.The findings indicate statistically significant improvements in several dimensions of self-regulated learning among students who used the AI-supported formative assessment system. In particular, students demonstrated stronger capacities for goal setting, monitoring their learning progress, and adapting study strategies in response to feedback. Qualitative reflections further suggest that personalized and timely feedback enhanced learners' sense of autonomy and increased their confidence in managing academic tasks independently.At the same time, the study identifies potential challenges associated with AI-supported formative assessment. A proportion of students reported a tendency to rely excessively on automated recommendations, occasionally reducing opportunities for independent reflection and self-evaluation. These findings suggest that while artificial intelligence can effectively support self-regulated learning, its implementation should be accompanied by pedagogical strategies that preserve student agency and encourage critical engagement with feedback.The paper concludes by proposing design principles for AI-supported formative assessment systems that integrate technological efficiency with reflective learning practices and instructor guidance. Such an approach may contribute to the development of autonomous, self-regulated learners capable of navigating increasingly digital and technology-rich higher education environments.