Empowering Economic Simulation for Massively Multiplayer Online Games through Generative Agent-Based Modeling

2025-06-05

AI Summary

Traditional agent-based models (ABMs) in massively multiplayer online (MMO) economic simulation suffer from inherent limitations in agent reliability, sociality, and interpretability. Method: This study pioneers the integration of large language models (LLMs) into MMO economic modeling, introducing generative agents endowed with role-playing, environmental perception, long-term memory, and causal reasoning capabilities. The framework unifies role-driven agent design with economic behavior simulation, enabling emergent specialization, adaptive responses to supply-demand shocks, and realistic market-like price fluctuations. Contribution/Results: Empirical evaluation demonstrates that LLM-powered agents significantly enhance behavioral fidelity, depth of social interaction, and mechanistic interpretability of simulations. This work establishes a novel paradigm for modeling complex virtual economic systems, advancing beyond conventional ABMs by embedding rich cognitive and socio-economic dynamics within autonomous agents.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Within the domain of Massively Multiplayer Online (MMO) economy research, Agent-Based Modeling (ABM) has emerged as a robust tool for analyzing game economics, evolving from rule-based agents to decision-making agents enhanced by reinforcement learning. Nevertheless, existing works encounter significant challenges when attempting to emulate human-like economic activities among agents, particularly regarding agent reliability, sociability, and interpretability. In this study, we take a preliminary step in introducing a novel approach using Large Language Models (LLMs) in MMO economy simulation. Leveraging LLMs' role-playing proficiency, generative capacity, and reasoning aptitude, we design LLM-driven agents with human-like decision-making and adaptability. These agents are equipped with the abilities of role-playing, perception, memory, and reasoning, addressing the aforementioned challenges effectively. Simulation experiments focusing on in-game economic activities demonstrate that LLM-empowered agents can promote emergent phenomena like role specialization and price fluctuations in line with market rules.

Problem

• Enhancing agent reliability and sociability in MMO economies
• Improving interpretability of agent-based economic simulations
• Simulating human-like decision-making using LLM-driven agents

Innovation

• Using LLMs for agent-based economic simulation
• Enhancing agents with human-like decision-making
• Promoting emergent economic phenomena via LLMs