Modeling Human Behavior in a Strategic Network Game with Complex Group Dynamics

2025-05-01

AI Summary

This study addresses the modeling and understanding of small-group human behavior (6–11 members) in strategic online games to foster socially beneficial outcomes. We propose hCAB, a novel framework integrating community-aware behavioral modeling with distributional learning of group-level decision-making dynamics, evaluated on the Junior High Game dataset. Methodologically, hCAB leverages community-aware graph-theoretic game modeling, distributional fitting, statistical inference, and human–machine comparative user studies. Results demonstrate, for the first time in this domain, that distribution-level modeling significantly outperforms conventional mean-based approaches, and that explicit incorporation of community structure substantially improves behavioral prediction accuracy. hCAB achieves high-fidelity reproduction of group dynamics and generates individual-level actions that pass a Turing-style test—human participants cannot reliably distinguish them from real human behavior. The core contribution is the empirical validation of two essential paradigms: distributional modeling and community awareness—as jointly necessary and effective for strategic networked behavioral modeling.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Game Theory and Economic Paradigms: Behavioral Game Theory

Application Category

• Economics, Online Markets and Human Computation: Social networks and social learning
• 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: Attacks and countermeasures in recommendation systems

Abstract

Human networks greatly impact important societal outcomes, including wealth and health inequality, poverty, and bullying. As such, understanding human networks is critical to learning how to promote favorable societal outcomes. As a step toward better understanding human networks, we compare and contrast several methods for learning models of human behavior in a strategic network game called the Junior High Game (JHG). These modeling methods differ with respect to the assumptions they use to parameterize human behavior (behavior vs. community-aware behavior) and the statistical moments they model (mean vs. distribution). Results show that the highest-performing method models the population's distribution rather than the mean and assumes humans use community-aware behavior rather than behavior matching. When applied to small societies (6-11 individuals), this learned model, called hCAB, closely mirrors the population dynamics of human groups (with some differences). Additionally, a user study reveals that human participants were unable to distinguish hCAB agents from other humans, thus illustrating that individual hCAB behavior plausibly mirrors human behavior in this strategic network game.

Problem

• Modeling human behavior in strategic network games
• Comparing methods for behavior parameterization assumptions
• Evaluating performance of community-aware vs behavior matching models

Innovation

• Models community-aware behavior in networks
• Focuses on population distribution not mean
• Uses hCAB to mirror human dynamics