Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

How GenAI Mentor Configurations Shape Early Collaborative Dynamics: A Classroom Comparison of Individual and Shared Agents

Siyu Zha, Weijing Liu, Fei Qin, Jie Cao, Yanjin Wang, Yujia Liu, Kaiyi Zhang, Jiangtao Gong, Yingqing Xu

Abstract

Generative artificial intelligence (GenAI) is increasingly embedded in computer-supported collaborative learning (CSCL), yet little empirical research has unpacked how different configurations of AI participation reshape collaborative processes. This study investigates how GenAI configuration shapes collaborative regulation in authentic classroom settings. Two eighth-grade classes engaged in small-group creative problem-solving under two conditions: a shared-AI configuration, in which each group interacted with a single AI mentor, and an individual-AI configuration, in which each student accessed a personal AI instance. Using multi-layer discourse coding combined with lag sequential analysis (LSA) and ordered network analysis (ONA), we examined interaction distribution, AI-student coupling, shared regulation processes, and teacher orchestration. Results reveal distinct regulatory dynamics across configurations. Shared AI access promoted convergence-oriented collaboration, with stronger alignment of shared regulatory states and more coordinated group-level reasoning. In contrast, individual AI access distributed support across learners, producing more exploratory and evaluative cycles but also more fragmented interaction patterns, accompanied by increased teacher intervention to manage divergence. These findings suggest that AI configuration functions as a structural design variable that reorganizes the regulatory ecology of classroom collaboration.

How GenAI Mentor Configurations Shape Early Collaborative Dynamics: A Classroom Comparison of Individual and Shared Agents

Abstract

Generative artificial intelligence (GenAI) is increasingly embedded in computer-supported collaborative learning (CSCL), yet little empirical research has unpacked how different configurations of AI participation reshape collaborative processes. This study investigates how GenAI configuration shapes collaborative regulation in authentic classroom settings. Two eighth-grade classes engaged in small-group creative problem-solving under two conditions: a shared-AI configuration, in which each group interacted with a single AI mentor, and an individual-AI configuration, in which each student accessed a personal AI instance. Using multi-layer discourse coding combined with lag sequential analysis (LSA) and ordered network analysis (ONA), we examined interaction distribution, AI-student coupling, shared regulation processes, and teacher orchestration. Results reveal distinct regulatory dynamics across configurations. Shared AI access promoted convergence-oriented collaboration, with stronger alignment of shared regulatory states and more coordinated group-level reasoning. In contrast, individual AI access distributed support across learners, producing more exploratory and evaluative cycles but also more fragmented interaction patterns, accompanied by increased teacher intervention to manage divergence. These findings suggest that AI configuration functions as a structural design variable that reorganizes the regulatory ecology of classroom collaboration.
Paper Structure (28 sections, 5 figures, 2 tables)

This paper contains 28 sections, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Theoretical Background
  3. Human--AI Collaboration as Distributed Cognitive Activity
  4. GenAI Mentor Configurations and the Structuring of Interaction
  5. Socially Shared Regulation of Learning in AI-mediated Collaboration
  6. Methodology
  7. Research Context and Participants
  8. Experimental Conditions: Individual-Agent vs. Shared-Agent AI Mentors
  9. AI Mentor System and Technical Background
  10. Data Collection
  11. Multi-layer Coding Framework and Interpretive Orientation
  12. Data Analyses
  13. Descriptive and Between-Condition Comparisons
  14. Lag Sequential Analysis (LSA)
  15. Ordered Network Analysis (ONA)
  16. ...and 13 more sections

Figures (5)

  • Figure 1: Mean proportions and 95% bootstrapped confidence intervals of all interaction codes in Condition A and Condition B.
  • Figure 2: Lag Sequential Analysis of Interaction Patterns Under Two AI Configurations (shared-AI (Condition A) on left, individual-AI (Condition B) on right)
  • Figure 3: Subtracted ONA network (B–A) highlighting relative interaction patterns, with red indicating stronger transitions in Condition B and blue indicating stronger transitions in Condition A.
  • Figure 4: Heatmap of differential transitions from AI functions to student shared regulation behaviors, comparing the shared-AI (Condition A) and individual-AI (Condition B) configurations.
  • Figure 5: Transition heatmap of shared regulation (SSRL) behaviors across the two AI configurations. Each cell represents the normalized frequency of transitions from one SSRL category to another (from → to).