Proceedings of the Annual Meeting of the Cognitive Science Society

Collaboration supports learning and problem-solving, but its effectiveness depends on cognitive engagement during discourse. This study applies an extended 7-point ICAP framework based on the Interactive, Constructive, Active, and Passive modes to characterize variation in cognitive engagement during collaborative dialogue. Engagement was coded by trained human annotators and compared with large language model (LLM)–based labeling approaches, including in-context learning (ICL), zero-shot prompting, and self-reflective agents. Interrater reliability among human annotators was robust across framework refinement stages (_ = 0.906–0.998), higher than the moderate agreement observed for ICL-based annotation (_ = 0.541–0.609). The human-refined framework improved agreement among human annotators (–ñ_ = 0.10), but produced only modest gains for ICL-based LLMs (–ñ_ < 0.04). Agent-refined frameworks improved cross-model agreement but remained below the human-refined framework. These findings highlight the promise of agent-based approaches and the importance of continued interaction between theory-guided human annotation and LLM-based methods in future work.