An artificial intelligence-driven learning analytics method to examine the collaborative problem-solving process from the complex adaptive systems perspective

Abstract

Collaborative problem solving (CPS) enables student groups to complete learning tasks, construct knowledge, and solve problems. Previous research has argued the importance of examining the complexity of CPS, including its multimodality, dynamics, and synergy from the complex adaptive systems perspective. However, there is limited empirical research examining the adaptive and temporal characteristics of CPS, which may have led to an oversimplified representation of the real complexity of the CPS process. To expand our understanding of the nature of CPS in online interaction settings, the present research collected multimodal process and performance data (i.e., speech, computer screen recordings, concept map data) and proposed a three-layered analytical framework that integrated AI algorithms with learning analytics to analyze the regularity of groups’ collaboration patterns. The results surfaced three types of collaborative patterns in groups, namely the behaviour-oriented collaborative pattern (Type 1) associated with medium-level performance, the communication-behaviour-synergistic collaborative pattern (Type 2) associated with high-level performance, and the communication-oriented collaborative pattern (Type 3) associated with low-level performance. This research further highlighted the multimodal, dynamic, and synergistic characteristics of groups’ collaborative patterns to explain the emergence of an adaptive, self-organizing system during the CPS process. According to the empirical research results, theoretical, pedagogical, and analytical implications were discussed to guide the future research and practice of CPS.

The role of learning theory in multimodal learning analytics

“…This momentary and ecologically valid information provides an opportunity for a paradigm shift in the field of learning sciences, as this new source of information might not necessarily fit existing paradigms that rely on a static notion of learning. Therefore, the produced realities from MMLA research have the potential to 'strike back' (Kuhn, 1962) and reinforce the development of new theories, extend current ones or even force the field to generate new ones that account for the dynamic nature of learning (eg, dynamic systems theory or complex adaptive systems perspective; Ouyang et al, 2022). In this way, MMLA has the capacity to challenge established 'truths' and allow different theoretical lenses to further our knowledge of how humans learn.…”

Human and artificial intelligence collaboration for socially shared regulation in learning

“…In the context of (S)SRL research, AI has enabled promising facilities for better understanding and supporting learning regulation (Molenaar, 2022; Nguyen, Järvelä, Wang, et al, 2022). Specifically, these approaches have enabled the analysis of the multi‐level and multifaceted characteristics of regulation in collaborative learning (Ouyang, Wu, et al, 2023; Ouyang, Xu, et al, 2023). In the past, the complexity and dynamics of collaborative learning often required researchers to examine different facets of regulation separately.…”

“…Although prior studies have offered valuable insights into (S)SRL processes, there is a recent call for more synergistic analysis approaches combining different facets of regulation (Nguyen & Järvelä, 2023). Recent studies have been able to address this need by using AI‐driven approaches for multi‐channel sequence analysis to examine multifaceted, adaptive and temporal characteristics of regulation in collaborative learning (Ouyang, Xu, et al, 2023). However, the integration of AI into (S)SRL research and support development has still faced several challenges related to the alignment between the theoretical and technological aspects of human‐AI interactions (Hwang et al, 2020).…”