Unlocking Learning Potential in Advanced-level Mathematics: AI-Driven Spaced Retrieval for Enhanced Pedagogy

Abstract

Retrieval practice is an effective learning approach that involves students actively retrieving knowledge from memory after completing the lesson (like lectures, workshops, and labs), which improves memory retention and fosters long-term learning. In short, this study underpins Retrieval Practice Theory. Unlike the pre-test, which involves taking practice tests before learning new information as opposed to afterward, retrieval practice is a strategy where students are tested on questions about unit materials that they have already learned. However, a more interactive spaced retrieval practice using a new platform, such as the highly recommended AI-powered question generators like Edapp, is still limited. This new spaced retrieval practice was implemented two times per semester for mid-semester tests, and then the data was collected for analysis. Based on an engineering first-year unit, namely MATH1020 Calculus for Engineers, after implementing spaced retrieval practices in the first semester of 2024, a significant increase in the passing rate was observed, up to 20.33%. The 95% confidence intervals for the passing rates further reinforce this improvement, showing that the post‑intervention cohort’s performance was not only higher but also statistically more precise compared to the pre‑intervention groups. This substantial improvement demonstrates the effectiveness of the newly proposed intervention.