Cluster analysis of student-AI dialogue identifies three distinct collaboration profiles: Delegated Reasoning, Concerted Interpretation, and Delegated Elaboration¶

Claim · All claims
Evidence · 1 study · associational r1 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Agglomerative hierarchical clustering on coded dialogue proportions yielded three student-AI interaction profiles, labeled Delegated Reasoning (DR), Concerted Interpretation (CI), and Delegated Elaboration (DE). → Zhanxin Hao 2026

Evidence¶

Zhanxin Hao 2026¶

Zhanxin Hao, Xiaobo Liu, Jiaxin Fan, Yun Long, Jifan Yu, Wenli Chen, Yu Zhang. (2026). Unpacking Interaction Profiles and Strategies in Human-AI Collaborative Problem Solving: A Cognitive Distribution and Regulation Perspective. https://arxiv.org/abs/2603.21288 doi:10.48550/arxiv.2603.21288

q2 · i? · associational · r1

Cluster analysis of proportion vectors of SEDA-adapted dialogue codes from 173 students' CPS sessions with AI agents. The article reports that "the optimal Silhouette index (0.22) was achieved when the number of clusters was set to 3," producing groups of 78, 55, and 40 students, labeled DR, CI, and DE based on their code profiles.

"the optimal Silhouette index (0.22) was achieved when the number of clusters was set to 3, resulting in three groups of students, consisting of 78, 55, and 40 individuals, respectively."

Discussion¶

Related Claims¶

• GAI-assisted DST creation produced significantly higher collaborative problem-solving skills than traditional methods among university students — the general claim this finding bears on, with the studies behind it
• The Concerted Interpretation group reported significantly higher self-regulation than Delegated Reasoning, with no profile differences in co-regulation or socially-shared regulation — a narrower finding that bears on this claim
• Semantic similarity between human and AI discourse is highest in the Delegated Reasoning profile and significantly lower in Concerted Interpretation — a narrower finding that bears on this claim
• The Delegated Reasoning group significantly outperformed the Concerted Interpretation group on collaborative problem-solving task performance — related
• Epistemic network analysis shows metacognitive behaviors co-occur with invitation actions in DR and DE but with human-initiated reasoning and elaboration in CI — related
• Learners' generative-AI use profiles are heterogeneous, with extractive delegation more common than potentially collaborative use — related
• Unsupervised hierarchical clustering of knowledge vectors recovers expert bundle structure without bundle labels — related
• A secondary ENA adding ICAP engagement codes showed proactive interactions linked constructive engagement with adequate reasoning, while reactive dialogues emphasised procedural pairings — related
• Optimal clustering of five-channel CPS sequences detected three types of collaborative patterns among graduate student groups — related
• Clustering n-gram flip sequences per task yields three distinct problem-solving strategy groups across the binary challenges
• Cluster analysis of blog networks yields five distinct social-role categories — related
• Human and AI discourse differ functionally: students lead invitations, connection, and regulation moves, while AI leads reasoning, coordination, and agreement — related
• Collaboration aids learning of mathematical elaboration skills but not procedural skills — related
• Epistemic network analysis showed Type 1 concentrating on behaviour-related code connections, Type 2 on both communication- and behaviour-related connections, and Type 3 on communication-related connections — related