The Impact of Educational Agents on the Effectiveness of Human-Machine Collaborative Learning: AMeta-Analysis Based on Learning Motivation andAcademic Performance
Published: 27 March 2026| Version 1 | DOI: 10.17632/98w59cpmng.1
Contributor:
小冬 郑Description
This study constructs a theoretical framework from an integrated perspective of cognition, affect and motivation, and conducts a meta-analysis of 32 experimental and quasi-experimental studies worldwide. Results show that educational agents exert a significant positive effect on learning outcomes (SMD = 0.901), and this effect is moderated by discipline, intervention mode and student stage. Accordingly, we suggest that in practice, attention should be paid to discipline adaptation and refined agent design. Educational agents should evolve into comprehensive learning partners supporting cognition, affect and motivation to improve the overall effectiveness of human–AI collaborative learning.
Files
Categories
Artificial Intelligence, Education, Adaptive Agent