Enabling User-Created Multi-Agent Simulations: Interactive and Customizable 2D Environments to Study Team Dynamics with LLM Agents

Mohammed Almutairi, Charles Chiang, Haoze Guo, Nandini Banerjee, Maria Milkowski, Daniel Nguyen, Michael G Yankoski, Tim Weninger, Svitlana Volkova, Trenton W. Ford, Diego Gomez-Zara

Keywords: Multi-agent systems, large language models, simulations, team dynamics, interactive environments, social simulation, agent-based models.

TL;DR: An interactive LLM-based multi-agent system

Abstract: Enabling users to create their own simulations offers a powerful way to study how environments shape agent behavior and intelligence. We introduce VirT-Lab, a system that allows researchers and practitioners to design interactive, customizable simulations of team dynamics with LLM-based agents situated in 2D spatial environments. Unlike prior systems that restrict scenarios to predefined or static tasks, our approach empowers users to build environments, assign roles, and observe how agents coordinate, move, and adapt over time. A web-based interface makes these simulations accessible to both technical and non-technical users, supporting the design, execution, and analysis of complex multi-agent experiments without programming. By bridging team cognition behaviors with scalable agent-based modeling, our system provides a testbed for investigating how diverse environments influence coordination, collaboration, and emergent team behaviors. We demonstrate its utility by aligning simulated outcomes with empirical evaluations and a user study, underscoring the importance of customizable environments for advancing research on collective intelligence and adaptive agents.

Archival Option: The authors of this submission want it to appear in the archival proceedings.

Submission Number: 103