Can Human-Machine Collaborative Learning Based on Generative Artificial Intelligence Improve Student Learning Outcomes? A Meta-Analysis of 20 Experimental and Quasi-Experimental Studies

Abstract

As a result of the application of artificial intelligence (AI) in education, human-machine collaborative learning emerged as a novel learning modality, attracting much attention in academia. The advent of generative AI (Gen AI) gave new impetus to this modality. Nevertheless, there are debates on the effectiveness of Gen AIbased human-machine collaborative learning. This article synthesizes the findings of 20 experimental and quasiexperimental studies, using the meta-analytical techniques, and examines the impact of moderating variables, such as the disciplinary domain, type of knowledge, duration of intervention, on the outcomes of Gen AI-based human-machine collaborative learning.