Bridging Talk and Thought: Understanding Dialogue Dynamics Across Collaborative Problem-Solving Contexts

Abstract

This paper introduces a hierarchical two-layer coding scheme for collaborative problem-solving dialogue, integrating cognitive and non-cognitive problem solving with metacognitive regulatory mechanisms. Validation across nine datasets from different domains shows that metacognitive regulation is a key indicator of successful collaboration between human and artificial agents.

Min-Yen Kan

Associate Professor

WING lead; interests include Digital Libraries, Information Retrieval and Natural Language Processing.

Nancy F. Chen

Research Collaborator

ASTAR Fellow; leads the Multimodal Generative AI group and the AI for Education Programme at ASTAR