Al-Based Feedback Systems and Student Learning Outcomes in Higher Education: A systematic review (2020-2025)

Abstract

This is a systematic review of the topic of AI-based feedback systems and their impact on student learning outcomes in higher education. Through Scopus, Web of Science (WoS) and ERIC, 93 records were obtained and filtered on the PRISMA principles, leaving 15 studies that were included in the study samples. The review mentions six broad categories of AI feedback tools, namely automated writing assessment systems, intelligent tutoring systems, generative AI-based feedback, learning analytics dashboards, chatbots, and AI-assisted grading systems. Throughout these tools, the learning outcomes that were reported were academic performance improvement, improvement in writing, conceptual understanding, engagement and motivation and self-regulated learning. The results also show that AI-based feedback can also be implemented in various fields, with the best representation of them being social sciences, then education, engineering, and health sciences. Overall, the review highlights the increasing importance of AI responses to improve instructional processes, help learners to become more independent, and offer scalable, time-sensitive and individualised feedback that can give some credible perspectives on the future of teaching and learning in higher education.