Investigating the training effectiveness of a theory-driven multi-agent system for pre-service teachers’ interpersonal communication competence

• In Event: AI-Mediated Teaching and Learning Across Educational Contexts

Abstract

Pre-service teacher communication training faces challenges in scalability and individualization. This study integrates Carkhuff’s interpersonal communication theory with self-regulated learning (SRL) theory to construct a three-phase training framework. A multi-agent system coordinated role-play, task management, and feedback to support complex skills development. In a nine-week quasi-experimental study, 35 pre-service teachers participated, with eye-tracking capturing learning processes. Results showed sustained development in the low self-efficacy group and cognitive recalibration in the high self-efficacy group. Delayed effect emerged in supportive communication and perspective taking, with eye-tracking revealing distinct attention strategies. Findings support individualized teacher education and interdisciplinary integration of educational technology and learning sciences.

Authors

• Mingzi Zhang, East China Normal University

• Min Lan, Zhejiang Normal University