Using multilayer network analysis to reveal the regulated learning characteristics between the learner groups with high and low levels of socially regulated learning

Regulated learning in collaboration involves interactions among self-regulation, co-regulation, and socially shared regulation, yet empirical evidence on how they interact to support collaboration remains scarce. This gap arises because regulated learning is complex, covert, and multilayered, and previous methods have limited capacity to capture such intertwined dynamics. This research employed multilayer network analysis to empirically investigate the multi-level and intertwined characteristics of groups with high and low levels of socially regulated learning, based on multimodal data collected from a series of collaborative activities in China’s higher education. Results showed that high-level groups exhibited: (1) more influential nodes associated with adaptive-level regulation and deep cognitive strategies at three levels; (2) tightly interconnected and efficient regulation pattern; and (3) homogeneous inter-layer network structure and complementary inter-layer regulation mechanisms. Based on the findings, this research proposed pedagogical implications for supporting regulated collaborative learning, and methodological implications for using MNA to reveal the complex regulated learning mechanisms.