Impact of AI-Assisted Formative Feedback on Undergraduate Learning Outcomes: A Simulated Controlled Study

This journal-style empirical article presents a clearly labeled simulated study in Education / Learning Sciences. A simulated sample of 180 undergraduate learners divided equally between conventional feedback and AI-assisted formative-feedback conditions over eight instructional weeks. Both groups completed identical weekly assignments and quizzes. The simulated intervention group received immediate rubric-linked feedback, error explanations, and revision prompts from an AI-assisted feedback system. Outcomes included post-test score, assignment completion, revision frequency, and perceived feedback usefulness. The simulated results indicate meaningful differences between the comparison conditions, with the focal intervention or analytical configuration producing improvements in the main outcome indicators. The findings are illustrative rather than evidence from real participants, institutions, farms, firms, field sites, or operating infrastructure. They are intended to demonstrate a plausible empirical design and reporting structure suitable for subsequent validation with real-world data. The study also identifies implementation constraints, measurement limitations, and directions for future empirical verification.