Hierarchical Generative Agents for Simulating Sequential Human Behavior

2026-06-12

AI Summary

This study addresses the limitations of existing evacuation models, which often assume rational and homogeneous human behavior, thereby failing to capture the complex cognitive, emotional, and social dynamics observed during disasters, compounded by a scarcity of real-world data. To overcome these challenges, this work proposes a novel simulation framework that integrates personality-driven large language models with a three-layer cognitive architecture—comprising goal setting, path reasoning, and low-level navigation—to model heterogeneous and irrational sequential decision-making in dynamic, grid-based disaster environments. Calibrated against empirical data, the proposed approach significantly enhances the realism and predictive accuracy of evacuation simulations, mitigating the overly optimistic biases inherent in conventional models and offering more reliable decision support for emergency planning.

Abstract

Complex cognitive, emotional, and social processes shape human evacuations during natural disasters. Accurate modeling and understanding of human behavior in disasters or emergencies can greatly impact the evacuation process by informing more effective planning and resource allocation. However, collecting human data in these situations is very difficult, and existing computational evacuation models assume rational, homogeneous behavior, leading to unrealistic, overly optimistic predictions. To address this gap, we present a simulation framework of sequential human decision-making during an evacuation scenario, introducing cognitively grounded, persona-driven agents. Our framework models evacuation behavior in a grid-based urban environment that evolves over time, capturing fire and other hazards. Human agents are modeled as personas that make sequential decisions in response to environmental stimuli with cognition structured in three levels: high-level evacuation goals, mid-level route reasoning, and low-level navigation. Decision-making is driven by large language models (LLMs) coupled with a cognitive module and calibrated with empirical human evacuation data. We propose a dynamic, stimulus-driven disaster simulation framework that models human evacuation decision-making using persona-conditioned LLM agents and a cognitive hierarchy.

Problem

• human evacuation behavior
• computational modeling
• disaster simulation
• cognitive hierarchy
• sequential decision-making

Innovation

• hierarchical generative agents
• persona-driven LLMs
• cognitive hierarchy
• evacuation simulation
• sequential decision-making