CoGrid & the Multi-User Gymnasium: A Framework for Multi-Agent Experimentation

2026-04-16

AI Summary

Existing research on human–AI social decision-making is hindered by the lack of accessible, multi-agent experimental tools that support seamless human–AI collaboration. To address this gap, this work introduces CoGrid, a simulation library, and Multi-User Gymnasium (MUG), an experimental platform that uniquely integrates high-performance multi-agent simulation with a low-latency, scalable online human–AI interaction system. The platform supports both server-authoritative and peer-to-peer rollback networking architectures and leverages dual NumPy and JAX backends for efficient computation, while enabling real-time multiplayer interaction through web technologies. Its effectiveness is demonstrated through multiple case studies in psychology, cognitive science, and human–AI decision-making. The open-sourced codebase significantly lowers the barrier to entry for researchers in these fields.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Humans and AI: Planning and Decision Support for Human-Machine Teams
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior

Application Category

• Economics, Online Markets and Human Computation: Research challenges in human and human-AI computation
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

The increasing integration of artificial intelligence (AI) in everyday life brings with it new challenges and questions for regarding how humans interact with autonomous agents. Multi-agent experiments, where humans and AI act together, can offer important opportunities to study social decision making, but there is a lack of accessible tooling available to researchers to run such experiments. We introduce two tools designed to reduce these barriers. The first, CoGrid, is a multi-agent grid-based simulation library with dual NumPy and JAX backends. The second, Multi-User Gymnasium (MUG), translates such simulation environments directly into interactive web-based experiments. MUG supports interactions with arbitrary numbers of humans and AI, utilizing either server-authoritative or peer-to-peer networking with rollback netcode to account for latency. Together, these tools can enable researchers to deploy studies of human-AI interaction, facilitating inquiry into core questions of psychology, cognition, and decision making and their relationship to human-AI interaction. Both tools are open source and available to the broader research community. Documentation and source code is available at {cogrid, multi-user-gymnasium}.readthedocs.io. This paper details the functionality of these tools and presents several case studies to illustrate their utility in human-AI multi-agent experimentation.

Problem

• multi-agent experimentation
• human-AI interaction
• social decision making
• research tooling
• interactive experiments

Innovation

• multi-agent simulation
• human-AI interaction
• web-based experimentation
• rollback netcode
• JAX backend