TY - CONF A1 - Rudolph, Eric A1 - Steigerwald, Phillip A1 - Albrecht, Jens T1 - Comparing Human Role-Players and LLM-Simulated Clients in Online Counselling Training: An Analysis of Counselling Patterns T2 - Proceedings of the International Conference on Educational Data Mining N2 - This study investigates the capabilities of Large Language Models to simulate counselling clients in educational roleplays in comparison to human role-players. Initially, we recorded role-playing sessions, where novice counsellors interacted with human peers acting as clients, followed by role-plays between humans and clients simulated by Mistrals Mixtral 8x7b using 4-bit quantization. These interactions were analysed with a counselling communication pattern system at sentence level. We investigated two key questions: (1) to what extent LLM-generated responses replicate authentic conversational dynamics and (2) whether counsellors’ communication behaviour differs when interacting with human versus LLM-simulated clients. The findings highlight both similarities and differences in the application of counselling patterns across scenarios, showing the potential of LLM-based role-playing exercises to enhance counselling competencies and to identify areas for further refinement in virtual client simulations. KW - Large Language Models KW - Counselling Training KW - Role Playing KW - Artificial Intelligence KW - Educational Technology Y1 - 2025 UR - https://opus4.kobv.de/opus4-ohm/frontdoor/index/index/docId/4762 SN - 978-1-7336736-6-2 PB - International Educational Data Mining Society ER -